Best Practices

Table of Contents: Problem Statement Anaplan has had challenges getting our finance, human resources, and sales teams on the same page of what our headcount is and what our headcount is forecasted to be in order to hit our revenue targets. We struggled to understand where requisitions were in the cycle and how many were currently open. As a result, we built a connected Anaplan dashboard to help transform conversations between teams to focus on understanding the different forecasts and strategic planning across those forecasts. Inconsistency in hitting hiring targets outlined in operating plans As a hypergrowth software-as-a-service (SaaS) company, the ability to hire quality candidates and stay on track with hiring targets is critical to our success. We had been consistently falling behind our sales force’s net headcount due to not hitting hiring targets, as well as having wrong assumptions on our ability to hire, generating non-achievable net headcount numbers. Misalignment of roles and responsibilities between the stakeholder group and those closest to the requisition due to lack of transparency and interaction in the process Non-subject-matter experts were responsible for entering in information regarding open headcount requisitions. This resulted in recruiters having to exceed finance’s budgeted compensation forecasts to hire quality candidates, roles taking longer to staff than forecasted, and perhaps most importantly, a lack of transparency for key stakeholders (i.e. hiring managers had little visibility into how close they were to getting a new hire, the recruiter did not know when they would be getting a batch of new jobs to recruit against, and finance could not see where recruiting was in terms of staffing active requisitions). Time-consuming and error-prone process in closing out requisition when converting a to-be-hired to an employee Our finance team was required to do hours of time-consuming manual reconciliation to make sure that each to-be-hired requisition was “deactivated” upon an employee filling that role, rather than focus on value-add activities and analysis. Inconsistent data across enterprise systems Between Anaplan, our human capital management (Workday), application tracking system (Greenhouse), and customer relationship management (Salesforce), there were myriad discrepancies for attributes tagged to to-be-hired requisitions. Which office would someone sit in? Which territory would a new sales rep cover? These are questions that oftentimes could not garner a single answer when different teams and systems were asked. We needed a single source of truth, less prone to human error. Proposed Solution The proposed solution to address the above pain points was to bring the entire headcount requisition planning process into Anaplan. This would allow finance, sales, hiring managers, and recruiting to share data freely and provide full transparency into the hiring process while allowing subject-matter experts to provide inputs where needed. Overall Workflow We have broken our workflow down into six different stages: Anaplan headcount planning systems workflow. Stage 1: Finance works with their partners in the business to plan and create new to-be-hired people requisitions in their planning, budgeting & forecasting (PBF) model. Using a numbered list, a unique, immutable code is assigned to the requisition, which will follow it through the rest of the below steps (REQ #XXXXXX). A requisition is then created in the PBF model, and either manually "released" by the finance manager overseeing that requisition, or if it is a certain number of days to the forecasted start date, the requisition is automatically "released" into the process. Stage 1.5: If the requisition is for sales, it is then imported into our sales planning management model (SPM) where our sales ops team tags the relevant assignment data (geography, territory, etc.) based on their planning and forecasting needs. Stage 2: After that step (and after stage 1, if the requisition is not for a sales rep) the requisition is then imported into the new hiring manager model (HIRE). At this stage, the hiring manager tagged to the requisition tags additional attributes to the requisition that they are best equipped to answer such as the job type, job level, and location of the requisition. Layering in the mappings imported from our Compensation model allows for a nuanced view of the total compensation that supports the req being developed . The hiring manager then clicks a Boolean line item indicating they sign off, and the requisition is "released" to the next stage. Stage 3: Once the hiring manager signs off, the requisition is imported into our recruiting management model (RECRUIT). This model contains a dashboard for recruiters to go in and enter in the final set of data necessary for the requisition to be written into our applicant tracking system (ATS): information such as the legal entity the requisition will be hired into, a template job to use to make the requisition in our ATS, etc. A series of logic determines if a requisition is ready to be created in our ATS. If all mandatory fields are filled in, as well as finance/hiring manager/recruiting sign-off has been provided, the requisition becomes unfiltered on a saved view that an integration with our ATS picks up, which runs every hour. Stage 4: Recruiting then goes through their normal hiring activities in our ATS, and eventually a candidate is hired in the requisition position. Stage 5: We have another integration running from our ATS to our Headcount Management Software (HCM) so that when a candidate is hired, they immediately have an employee profile created in our HCM. One of the attributes is the unique requisition ID mentioned in step one. We have scheduled loads of employee data from our HCM into our Anaplan Data Hub model, as well as the planning, budgeting & forecasting model. When PBF detects an open requisition ID on an employee profile line, the formulas are built to "turn off" the requisition from the forecast, and "turn on" the new employee, creating a seamless, automatic cutover from a to-be-hired to an actual employee. Model Structure Anaplan headcount planning model structure and data flow. Anaplan headcount planning model structure and data flow. Five separate models are involved in the facilitation of this process that all sit in separate workspaces: Data Hub Planning, budgeting, and forecasting (PBF) Sales planning management (SPM) Hiring manager model (HIRE) Recruiting management (RECRUIT) Why did we decide to use multiple models instead of one large model? The first reason was the ease of user management. Since several of these models are core planning models, they contain sensitive financial and employee data. Breaking up the models mitigates the risk that someone accidentally was given inappropriate access by keeping finance partners in their own planning model (PBF), hiring managers in their own model (this is the largest user base in the process), and human resources and recruiting planning in their existing recruiting management model. That way, for example, a hiring manager has no risk of mistakenly being able to see our financial projects in the PBF model. The second reason is that smaller use-case-specific models greatly reduce the recalculation time of the imports required given the integration and scheduling requirements (hourly loads from our HCM and ATS systems) . It also allows us to be more tactical with what data is imported into each model, as each model has its own unique data requirements. We have saved views set up for each part of the sign-off process, which only pick up requisitions that have been recently approved and need to be synced, which also helps with the load times between models. Smaller models also allow for less complex and simpler transition plans in the event a model owner leaves. By breaking up use cases, the models they live in are easier to understand and can be transitioned individually versus needing to transition all 10 use cases together. User Access Management We take advantage of each user-provisioning functionality Anaplan provides. We use model roles to segment users by their job role in the model and pair that with selective access and Dynamic Cell Access (DCA): Selective access: Some requisitions are so sensitive that even their existence is confidential. To deal with this, we use selective access to make sure that only the proper people can see that a requisition the finance team tags as “confidential” even exists. Dynamic Cell Access: For everything else, we leverage DCA paired with formula drivers to make sure that each role can see only their relevant data. Only hiring managers can view data for requisitions they are tagged to, finance can only view data for requisitions in cost centers under their purview and recruiters can only view data for requisitions they are recruiting on. We prefer to use DCA where applicable, because formulas are easier to audit and tweak than trying to figure out which import actions are automating selective access. Stakeholders & Responsibilities Key Stakeholder Groups This project spanned multiple departments and stakeholder groups. At a high level, the teams involved: Finance partner Hiring manager of new to-be-hired requisition Recruiter Anaplan on Anaplan partner Sales operations partner Key Activities Description of the type of activities, if applicable: Finance creates requisition and fills in basic date Within their PBF model, the finance partners work with their business partners to create new headcount requisitions. They fill in basic information about the requisition in order to be able to generate a forecast (cost center, comp estimate, location, etc.). Finance-approved requisition When ready, finance clicks a Boolean line item on the requisition to “release” it into the broader process. It’s at this point that it becomes exposed to the next group of stakeholders. We’ve put in time gates here to auto-release requisitions that are a quarter out from their forecasted start date. This helps us hit our hiring targets, so requisitions do not “slip through the cracks” and we make sure that recruiting is given enough time to recruit against. Sales ops team assigns territory The sales ops team concurrently works with the requisition list (both unapproved and approved) to tag appropriate territories to each requisition based on their sales planning model’s outputs. This allows sales ops to directly manage the data they care about and know best about each requisition. Hiring manager fills in data Once finance approves the requisitions and sales ops fills in their data (if it’s a sales requisition), the requisition is then passed to the hiring manager model where the hiring manager of the requisition goes into their own dashboard and fills in information like job level, location, and job type. With this information, the model uses lookups to generate a refined compensation range for the requisition. This not only helps cut down on back-and-forth with recruiting but also helps finance refine their forecast by receiving more accurate compensation data. Hiring manager approves requisition After the hiring manager is done with their inputs and is happy with the compensation range generated (which is also validated against finance’s initial estimate), they approve the requisition by clicking their own Boolean. This reduces the number of emails and back-and-forth trying to get a requisition approved. Recruiter fills in data After the hiring manager approves, the requisition is then made available on the recruiter’s Anaplan dashboard. They fill in the remaining information needed to populate a new job in our applicant tracking system. This allows recruiting to automate the Greenhouse (ATS) job creation process by only having to enter in fields they need to create the job, and the integration does the rest! Recruiter approves requisition Once the recruiter is done with their inputs, they click their own Boolean line item and the requisition is then made available via a saved view to be written into our applicant tracking system. Anaplan on Anaplan (AoA) The AoA team, as our internal Center of Excellence, served as technical architect advisors, as well as project managers. In order to free up time for the business groups to do their own modeling, the AoA team worked cross-functionally to make sure that all groups were heard and hit their deliverable milestones. Responsibility Matrix Finance Partner Sales Ops Partner Hiring Manager Recruiter 3oA Create New Requisition R C Approve New Requisition R R (if applicable) R R I Assigned Territory to Requisition (if applicable) C R I I Fill in Data Re: Requisition R R R R C R = Responsible, A = Accountable, C = Consulted, I = Informed Data Sources & Integration Source System Target System Description Greenhouse Anaplan/Workday Job/Applicant/Activities/Hired Candidates Workday Anaplan Employee Information Salesforce Anaplan Account and Opportunity Data Anaplan>Anaplan Multiple Models Model data syncs across involved models Synchronization Cadence Besides Workday employee data, the above integrations are scheduled via HyperConnect to run every hour. Data created in each model is shared between the models via scheduled model-to-model imports using HyperConnect. Greenhouse to Workday is configured as an event-based webhook, so when a new candidate is hired the candidate’s information is immediately sent to Workday to create a new employee. Checks and Balances As information is updated in each model, part of the hourly sync cadence is to sync all requisition data between each model. This is done by running an import process in each model that imports data from the other models. These processes run sequentially before the integration to Greenhouse runs. This allows groups to layer in their own data checks at each part of the process and share those checks with each model. They are consolidated in the recruiting management model where our “master” checks reside. It looks for data completeness, as well as all other models signing off on their own checks. For example, if a hiring manager pushes through data with a salary of $100K and last second decides to change it to $200K after finance and recruiter signs off, the logic finance has built into their checks to flag out of budget compensation would be triggered ahead of the integration. This flag would pull the saved view out of the final staging view and the requisition would not be picked up. Return on Investment Increased Speed to Begin Recruiting When a requisition was released by finance, it would take over a week (if not weeks) to be created. Now with this new process created, our average time from when finance releases a requisition to when recruiting begins recruiting against it is less than six days. In fact, once recruiting gets the information from finance and the hiring manager, the average time to begin recruiting is less than a day! The decreased time allows recruiters to do what they do best: get a quality candidate hired. With more time to have a job posted, candidates interviewed, and offer negotiations completed, they can focus on hiring quality candidates while still hitting the company-wide hiring targets. Increased Transparency Now that each stakeholder group has its own Anaplan model and dashboards, there is no ambiguity for whose court the ball is in (i.e. what part of the process is the requisition stuck in?). Hiring managers can quickly see where each of their hires is along the process, and recruiting can proactively reach out to stakeholders to guide them along in filling out their information. Automated Conversion of To-Be-Hired to Employee With a single source of truth for requisition ID numbers, and integrations automatically flowing data between systems, our finance team no longer has to spend hours reconciling which requisitions have been hired. In our planning models, a requisition is automatically “turned off” and converted to an FTE when a new hire fills an open requisition. Refined Forecasts and More Accurate Data (Within Anaplan and Across Systems) Prior to this process, a requisition could be released from finance, have its compensation profile drastically changed, or its sales territory altered, etc., without any indication or notification. Only after an offer approval came through to finance and sales ops did they find out that something was wrong, which led to last-minute scrambling and process hang-ups when they mattered most (trying to lock in a quality hire). Now that each group is responsible for and has visibility into the attributes they care about, those problems are brought to light and rectified early on, far before they become issues in crunch time. Conclusion It’s clear to see the benefits that moving Anaplan’s headcount planning process into the Anaplan platform has. Those benefits were made possible by the partnership and hard work of the finance, talent acquisition, human resources, and sales teams. If you want to learn more about how Anaplan does headcount planning, reach out to your account representative for more information and a demonstration.

What is Pre-Allocation in Lists? Pre-allocation in lists is a mechanism in Anaplan that adds a buffer to list lengths. It is not added by default for lists; it becomes enabled when a role is set on a list. Please follow 1.03-01 though. Only add roles when needed. When it is enabled, a 2 percent buffer is added to the list, and this includes all line items where the list is used. This means we create extra space (in memory) for each line item so that when a new list item is added, the line item does not need to be expanded or restructured. When the buffer is used up (the list has run out of free slots) another 2 percent buffer will be created and any line items using the list will be restructured. This buffer is not shown in the list settings in Anaplan, meaning if we had a list with 1,000 items, that’s what Anaplan would show as the size. But in the background, that list has an extra 20 hidden and unused items. Pre-allocation also applies to list deletions but allows for 10 percent of the list to be deleted before any line items using the list get restructured. The purpose of pre-allocation in lists is to avoid restructuring line items that use frequently updated lists. What Happens When We Restructure? Restructuring the model is an expensive task in terms of performance and time. The Anaplan Hyperblock gets its efficiency by holding your data and multi-dimensional structures in memory — memory being the fastest storage space for a computer. Creating the model structures in memory — building the Hyperblock — does take a significant time to complete. But once it's in memory, access is quick. The initial model opening is when we first build those structures in memory. Once in memory, any further model opens (by other users, for example) are quick. Restructuring is the process of having to rebuild some parts of the model in memory. In the case of adding an item to a list, that means any line item that uses that list as a dimension. When restructuring a line item, we have to recalculate it, and this is often where we see the performance hit. This is because line items have references, so there is a calculation chain from any line item changed by that restructuring. Pre-allocation is there to reduce this extra calculation caused by restructuring. An example of this was seen in a model that was adding to a list that contained trial products. These products would then have a set of forecasted data calculated from historical data from real products. The list of these new products was small and changed reasonably frequently; it contained around 100 items. Adding an item took around two seconds (except every third item took two minutes). This happened because of the difference between adding to the pre-allocated buffer and when it had to do the full calculation (and re-adjust the buffer). Without pre-allocation, every list addition would have taken two minutes. Fortunately, we managed to optimize that calculation down from two minutes to several seconds, so the difference between adding to the pre-allocation buffer and the full calculation was around five seconds, a much more acceptable difference. In summary, pre-allocation on lists can give us a great performance boost, but it works better with larger lists than small lists. Small, Frequently Updated Lists As we’ve seen, the pre-allocation buffer size is 2 percent, so on a large list — say one million items — we have a decent-sized buffer and can add many items. When we have a small list that is frequently used, then a performance characteristic that is seen is frequently changing calculation times. This is especially the case if that list is heavily used throughout the model. A list with 100 items will restructure and recalculate on every third addition. This will continue to be noticeable for quite some time. Doubling the list size is still just adding four unused items (2 percent of 200). When we have a small list that is frequently used, you will see the calculation times change from fast to slow while the buffer is frequently filled. In cases like this, it is very important to reduce and optimize the calculations as much as possible. What Can Be Done? There are a few options. You could always make the list bigger and increase the buffer so that it restructures less. How? Option 1: Create a subset of “active” items and ignoring the additional list items used to bulk out the list. The problem with this would be the size of any line items using that list would increase and so would their calculations. Changing from a 100-item list to a 10,000- or even 1,000-item list (enough to give us a bigger buffer) could greatly increase the model size. Option 2: Create a new list that is not used in any modules so we avoid any restructuring costs. This would work but it adds a lot of extra manual steps. You would have this new list used in a single data entry module, which means this data is unconnected with the rest of the model. Being connected is what gives us value. You would then need to create a manual process to push data from this unconnected module to one that is connected to the rest of the model (this way all the changes will happen at once). We do lose the real-time updates and benefits of Connected Planning, though. Option 3: Reduce the impact of restructuring by optimizing the model and the formulas. Our best option is optimizing calculations. If we have quick calculations, the difference between buffered and unbuffered list additions could be small. The best way to achieve this would be through a Model Optimization Success Accelerator. This is a tailored service delivered by Anaplan Professional Services experts who aim to improve model performance through calculation optimizations. Please discuss this service with your Anaplan Business Partner. You can also follow our best practice advice and reference the Planual to find ways you can optimize your own models.

L'application Bring Your Own Key (BYOK) vous permet maintenant de vous approprier les clés de chiffrement de vos données de modèle. Si vous avez accès à l'outil Anaplan Administration, vous pouvez chiffrer et déchiffrer des espaces de travail sélectionnés à l'aide de vos propres clés AES-256. À la différence des clés principales système, les clés créées par BYOK vous appartiennent et vous en assurez la sécurité. Aucun mécanisme ne permet au personnel Anaplan d'accéder à vos clés. Bring Your Own Key (BYOK) - Guide de l'utilisateur Bring Your Own Key (BYOK) est un produit complémentaire que votre organisation peut acheter si elle possède l'édition Enterprise.

Assume the following Non-Composite list, ragged hierarchy, needs to be set to Production Data. We need to refer to the ultimate parent to define the logic calculation. In the example, we have assumed that children of Parent 1 and Parent 3 need to return the 'logic 1' value from the constants module below, and those under Parent 2 return 'logic 2,' and we apportion the results based on the initial data of the children. Select Proportion: Data / IF PARENT(ITEM('Non-Composite List')) = 'Non-Composite List'.'Parent 1' THEN Data[SELECT: 'Non-Composite List'.'Parent 1'] ELSE IF PARENT(ITEM('Non-Composite List')) = 'Non-Composite List'.'Parent 2' THEN Data[SELECT: 'Non-Composite List'.'Parent 2'] ELSE IF PARENT(ITEM('Non-Composite List')) = 'Non-Composite List'.'Parent 3' OR PARENT(ITEM('Non-Composite List')) = 'Non-Composite List'.'Child 3.1' THEN Data[SELECT: 'Non-Composite List'.'Parent 3'] ELSE 0 Select Calculation: Select Proportion * IF PARENT(ITEM('Non-Composite List')) = 'Non-Composite List'.'Parent 1' OR PARENT(ITEM('Non-Composite List')) = 'Non-Composite List'.'Parent 3' OR PARENT(ITEM('Non-Composite List')) = 'Non-Composite List'.'Child 3.1' THEN Parent Logic Constants.'Logic 1' ELSE IF PARENT(ITEM('Non-Composite List')) = 'Non-Composite List'.'Parent 2' THEN Parent Logic Constants.'Logic 2' ELSE 0 These “hard references” will prevent the list from being set as a production list. Solution: Create a Parents Only list (this could be imported from the Non-Composite list). As we don't need the sub-level parents, we do not need to include 'Child 3.1,' even though it is technically a parent. To calculate the proportion calculation without the SELECT, a couple of intermediate modules are needed: Parent Mapping Module This module maps the Non-Composite parents to the Parents Only list. Due to the different levels in the hierarchy, we need to check for sub levels and use the parent of Child 3.1. In this example, the mapping is automatic because the items in the Parents Only list have the same name as those in the Non-Composite list. The mapping could be a manual entry if needed. The formula and “applies to” are: Non-Composite Parent: PARENT(ITEM('Non-Composite List')) Applies to: Non-Composite List Parent of Non-Composite Parent: PARENT(Non-Composite Parent) Applies to: Non-Composite List Parent to Map: IF ISNOTBLANK(PARENT(Parent of Non Composite Parent)) THEN Parent of Non Composite Parent ELSE Non Composite Parent Applies to: Non-Composite List Parents Only List FINDITEM(Parents Only List, NAME(Parent to Map)) Applies to: Parents Only List Parents Only Subtotals An intermediary module is needed to hold the subtotals. Calculation: Parent Logic Calc.Data[SUM: Parent Mapping.Parents Only List] Parent Logic? Module We now define the logic for the parents in a separate module. Add Boolean line items for each of the “logic” types. Then you can refer to the logic above in the calculations. Lookup Proportion: Data / Parents Only Subtotals.Calculation[LOOKUP: Parent Mapping.Parents Only List] Lookup Calculation: Lookup Proportion * IF Parent Logic?.'Logic 1?'[LOOKUP: Parent Mapping.Parents Only List] THEN Parent Logic Constants.'Logic 1' ELSE IF Parent Logic?.'Logic 2?'[LOOKUP: Parent Mapping.Parents Only List] THEN Parent Logic Constants.'Logic 2' ELSE 0 The list can now be set as a production list as there are no “hard references”. Also, the formulas are smaller, simpler and now more flexible should the logic need to change. If Parent 3 needs to use Logic 2, it is a simple change to the checkbox. Appendix: Blueprints

Model Load: A large and complex model such as 10B cells can take up to 10 minutes to load the first time it's in use after a period of inactivity of 60 minutes. The only way to reduce the load time is by identifying what formula takes most of the time. This requires the Anaplan L3 support (ask for a Model Opening Analysis), but you can reduce the time yourself by applying the formula best practices listed above. One other possible leverage is on list setup: Text properties on a list can increase the load times, and subsets on lists can disproportionately increase load times. It is best practice not to use List Properties but house the attributes in a System model dimensioned by the list. See Best practice for Module design for more details. Model Save: A model will save when the amount of changes made by end-users exceeds a certain threshold. This action can take several minutes and will be a blocking operation. Administrators have no leverage on model save besides formula optimization and reducing model complexity. Using ALM and Deployed mode increases this threshold, so it is best to use Deployed mode whenever possible. Model Rollback: A model will roll back in some cases of an invalid formula, or when a model builder attempts to adjust a setting that would result in an invalid state. In some large models, the rollback takes approximately the time to open the model, and up to 10 minutes worth of accumulated changes, followed by a model save. The recommendation is to use ALM and have a DEV model which size does not exceed 500M cells, with a production list limited to a few dozen items, and have TEST and PROD models with the full size and large lists. Since no formula editing will happen in TEST or PROD, the model will never rollback after a user action. It can roll back on the DEV model but will take a few seconds only if the model is small.

Reducing the number of calculations will lead to quicker calculations and improve performance. However, this doesn’t mean combining all your calculations into fewer line items, as breaking calculations into smaller parts has major benefits for performance. Learn more about this in the Formula Structure article. How is it possible to reduce the number of calculations? Here are three easy methods: Turn off unnecessary Summary method calculations. Avoid formula repetition by creating modules to hold formulas that are used multiple times. Ensure that you are not including more dimensions than necessary in your calculations. Turn off Summary method calculations Model builders often include summaries in a model without fully thinking through if they are necessary. In many cases, the summaries can be eliminated. Before we get to how to eliminate them, let’s recap on how the Anaplan engine calculates. In the following example we have a Sales Volume line-item that varies by the following hierarchies: Region Hierarchy Product Hierarchy Channel Hierarchy City SKU Channel Country Product All Channels Region All Products All Regions This means that from the detail values at SKU, City, and Channel level, Anaplan calculates and holds all 23 of the aggregate combinations shown below—24 blocks in total. With the Summary options set to Sum, when a detailed item is amended (represented in the grey block), all the other aggregations in the hierarchies are also re-calculated. Selecting the None summary option means that no calculations happen when the detail item changes. The varying levels of hierarchies are quite often only there to ease navigation, and the roll-up calculations are not actually needed, so there may be a number of redundant calculations being performed. The native summing of Anaplan is a faster option, but if all the levels are not needed it might be better to turn off the summary calculations and use a SUM formula instead. For example, from the structure above, let’s assume that we have a detailed calculation for SKU, City, and Channel (SALES06.Final Volume). Let’s also assume we need a summary report by Region and Product, and we have a module (REP01) and a line item (Volume) dimensioned as such. REP01.Volume = SALES06 Volume Calculation.Final Volume is replaced with REP01.Volume = SALES06.Final Volume[SUM:H01 SKU Details.Product, SUM:H02 City Details.Region] The second formula replaces the native summing in Anaplan with only the required calculations in the hierarchy. How do you know if you need the summary calculations? Look for the following: Is the calculation or module user-facing? If it is presented on a dashboard, then it is likely that the summaries will be needed. However, look at the dashboard views used. A summary module is often included on a dashboard with a detail module below; Effectively, the hierarchy sub-totals are shown in the summary module, so the detail module doesn’t need the sum or all the summary calculations. Detail to Detail Is the line item referenced by another detailed calculation line item? This is very common, and if the line item is referenced by another detailed calculation the summary option is usually not required. Check the Referenced by column and see if there is anything referencing the line item. Calculation and staging modules If you have used the D.I.S.C.O. module design, you should have calculation/staging modules. These are often not user-facing and have many detailed calculations included in them. They also often contain large cell counts, which will be reduced if the summary options are turned off. Can you have different summaries for time and lists? The default option for Time Summaries is to be the same as the lists. You may only need the totals for hierarchies, or just for the timescales. Again, look at the downstream formulas. The best practice advice is to turn off the summaries when you create a line item, particularly if the line item is within a Calculation module (from the D.I.S.C.O. design principles). Avoid Formula Repetition An optimal model will only perform a specific calculation once. Repeating the same formula expression multiple times will mean that the calculation is performed multiple times. Model builders often repeat formulas related to time and hierarchies. To avoid this, refer to the module design principles (D.I.S.C.O.) and hold all the relevant calculations in a logical place. Then, if you need the calculation, you will know where to find it, rather than add another line item in several modules to perform the same calculation. If a formula construct always starts with the same condition evaluation, evaluate it once and then refer to the result in the construct. This is especially true where the condition refers to a single dimension but is part of a line item that goes across multiple dimension intersections. A good example of this can be seen in the example below: START() <= CURRENTPERIODSTART() appears five times and similarly START() > CURRENTPERIODSTART() appears twice. To correct this, include these time-related formulas in their own module and then refer to them as needed in your modules. Remember, calculate once; reference many times! Taking a closer look at our example, not only is the condition evaluation repeated, but the dimensionality of the line items is also more than required. The calculation only changes by the day, as per the diagram below: But the Applies To here also contains Organization, Hour Scale, and Call Center Type. Because the formula expression is contained within the line item formula, for each day the following calculations are also being performed: And, as above, it is repeated in many other line items. Sometimes model builders use the same expression multiple times within the same line item. To reduce this overcalculation, reference the expression from a more appropriate module; for example, Days of Week (dimensioned solely by day) which was shown above. The blueprint is shown below, and you can see that the two different formula expressions are now contained in two line items and will only be calculated by day; the other dimensions that are not relevant are not calculated. Substitute the expression by referencing the line items shown above. In this example, making these changes to the remaining lines in this module reduces the calculation cell count from 1.5 million to 1500. Check the Applies to for your formulas, and if there are extra dimensions, remove the formula and place it in a different module with the appropriate dimensionality .

Thinking through the results of a modeling decision is a key part of ensuring good model performance—in other words, making sure the calculation engine isn’t overtaxed. This article highlights some ideas for how to lessen the load on the calculation engine. Formulas should be simple; a formula that is nested, or uses multiple combinations, uses valuable processing time. Writing a long, involved formula makes the engine work hard. Seconds count when the user is staring at the screen. Simple is better. Breaking up formulas and using other options helps keep processing speeds fast. You must keep a balance when using these techniques in your models, so the guidance is as follows: Break up the most commonly changed formula Break up the most complex formula Break up any formula you can’t explain the purpose of in one sentence Formulas with many calculated components The structure of a formula can have a significant bearing on the amount of calculation that happens when inputs in the model are changed. Consider the following example of a calculation for the Total Profit in an application. There are five elements that make up the calculation: Product Sales, Service Sales, Cost of Goods Sold (COGS), Operating Expenditure (Op EX), and Rent and Utilities. Each of the different elements is calculated in a separate module. A reporting module pulls the results together into the Total Profit line item, which is calculated using the formula shown below. What happens when one of the components of COGS changes? Since all the source components are included in the formula, when anything within any of the components changes, this formula is recalculated. If there are a significant number of component expressions, this can put a larger overhead on the calculation engine than is necessary. There is a simple way to structure the module to lessen the demand on the calculation engine. You can separate the input lines in the reporting module by creating a line item for each of the components and adding the Total Profit formula as a separate line item. This way, changes to the source data only cause the relevant line item to recalculate. For example, a change in the Product Sales calculation only affects the Product Sales and the Total Profit line items in the Reporting module; Services Sales, Op EX, COGS and Rent & Utilities are unchanged. Similarly, a change in COGS only affects COGS and Total Profit in the Reporting module. Keep the general guidelines in mind. It is not practical to have every downstream formula broken out into individual line items. Plan to provide early exits from formulas Conditional formulas (IF/THEN) present a challenge for the model builder in terms of what is the optimal construction for the formula, without making it overly complicated and difficult to read or understand. The basic principle is to avoid making the calculation engine do more work than necessary. Try to set up the formula to finish the calculations as soon as possible. Always put first the condition that is most likely to occur. That way the calculation engine can quit the processing of the expression at the earliest opportunity. Here is an example that evaluates Seasonal Marketing Promotions: The summer promotion runs for three months and the winter promotion for two months. There are more months when there is no promotion, so this formula is not optimal and will take longer to calculate. This is better, as the formula will exit after the first condition more frequently. There is an even better way to do this. Following the principles from above, add another line item for no promotion. And then the formula can become: This is even better because the calculation for No Promo has already been calculated, and Summer Promo occurs more frequently than Winter Promo. It is not always clear which condition will occur more frequently than others, but here are a few more examples of how to optimize formulas: FINDITEM formula The Finditem element of a formula will work its way through the whole list looking for the text item, and if it does not find the referenced text, it will return blank. If the referenced text is blank, it will also return a blank. Inserting a conditional expression at the beginning of the formula keeps the calculation engine from being overtaxed. IF ISNOTBLANK(TEXT) THEN FINDITEM(LIST,TEXT) ELSE BLANK Or IF BLANK(TEXT) THEN BLANK ELSE FINDITEM(LIST,TEXT) Use the first expression if most of the referenced text contains data and the second expression if there are more blanks than data. LAG, OFFSET, POST, etc. If in some situations there is no need to lag or offset data, for example, if the lag or offset parameter is 0. The value of the calculation is the same as the period in question. Adding a conditional at the beginning of the formula will help eliminate unnecessary calculations: IF lag_parameter = 0 THEN 0 ELSE LAG(Lineitem, lag_parameter, 0) Or IF lag_parameter <> 0 THEN LAG(Lineitem, lag_parameter, 0) ELSE 0 The use of formula a or b will depend on the most likely occurrence of 0s in the lag parameter. Booleans Avoid adding unnecessary clutter for line items formatted as BOOLEANS. There is no need to include the TRUE or FALSE expression, as the condition will evaluate to TRUE or FALSE. Sales>0 Instead of IF Sales > 0 then TRUE ELSE FALSE

Line item subsets are one of the most powerful and efficient features in Anaplan, yet one of the least understood. The COLLECT() function is probably the only “black box” function within Anaplan as is it not immediately apparent what it is doing or where the source values are coming from. In the following article, I will cover how to understand line item subsets more easily, and also explain their many uses, some of which do not need COLLECT() at all. For more information on creating line item subsets see Line Item Subsets in Anapedia. A line-item subset is a list of items drawn from one or more line items from one or more modules. Put simply, it converts line items into a list on which calculations can be performed. There are some restrictions: Line item subsets can only contain numeric formatted line items. Only one line item subset can be used as a dimension in a module. Although line items can contain formulas, the items in a line item subset can only aggregate to a simple subtotal. Styles on the line items are not transferred over to the line item subset. Line item subsets can be used for many different areas of functionality. For the examples used, I have based them on the final model from the new Level 1 training. Download the model and follow the instructions to practice on the same structures. These examples are deliberately simplified, but I hope you find these insightful and easy to transfer into your models to simplify the formulae and provide more flexibility to your users. Table of Contents: Calculations on Calculations This is the classic use of line item subsets. A source module contains line items, and subsequently, you need to perform additional calculations on these line items. While in some cases this can be managed through complex formulae, normally these workarounds break most of the best practice guidelines and should be avoided. Use Case example: The source module contains forecast data with line items for the profit and loss lines in U.S. dollars. We need to convert these values into local currency based on the Country dimension. The source modules are as follows: The first step is to create the line item subset, and for this report, we only want summary values. In the settings tab, choose Line Item Subsets and click insert. We recommend prefixing with LIS: the name of the module and simple description. Clicking on the Line Item Subset header item (in settings) will display the Line Item Subsets screen. Click on the newly created line item subset and the … and select the module(s) required; in this case, it is REP03. Select which line items you wish to include in the line item subset. Now that the line item subset has been created, it is available to be used in a module. Create a module with the following dimensions: LIS: REP03 Currency G2 Country Time (Years) Add the following line items: Base Currency Exchange Rate Local Currency In the Base Currency line item, enter the formula: COLLECT() Note that the values are the same as those in REP03 and the line items are now shown in list format (no formatting). Also note that these values are from the Forecast version, as the target module does not have versions, so the Current Version is used as the source automatically. Add the following formulae to the remaining line items to complete the calculation. Exchange Rate = 'DATA02 Exchange Rates'.Rate[LOOKUP: 'SYS03 Country Details'.Currency Code] Local Currency = Base Currency / Exchange Rate Note that the Exchange Rate line item should be set as a Subsidiary view (excluding the line item subset from the applies to) because we are showing it on the report for clarity. If this display was not required, the calculation could be combined with the Local Currency formula. Transformation You can also use a line item subset to help with the transformation between source and target modules. Use Case example: We want to summarize costs (from the reporting P&L) into Central and Locally controlled costs. Create a list (Controllable Costs) containing two members. Central Local Create a line item subset (as before) using just REP03 as the source module. Create a staging module with the following dimensions: LIS: REP03 Cost Reporting G2 Country Time (Years) Add a line item (Data) and enter COLLECT() as the formula. Set the Summary method to None; we do not need subtotals in this module. Create a mapping module, dimensioned by LIS: REP03 Cost Reporting. Add a line item (Mapping) formatted as the Controllable Costs list. Map the lines as applicable. Create a reporting module with the following dimensions. Controllable Costs G2 Country Time (Years) Add a line item called Costs. Add the formula: 'REP07 Cost Reporting Staging'.Data[SUM: 'SYS14 Cost Mapping'.Mapping] We use the SUM formula because the source dimension and the mapping dimension are the same. So, “If the source is the same, it’s a SUM.” Multiple Source Modules Line item subsets can contain line items from multiple modules. There is a caveat though; all modules must share at least one common dimension/hierarchy and/or have a Top Level set for non-matching dimensions. Use case example: Based on user-entered settings, we want to compare the values from two time periods for metrics from three different modules and calculate the absolute and % variances. The source modules all share a common dimension: REV03 Margin Calculation: G2 Countries, P2 Products, Month EMP03 Employee Expenses by Country: G2 Countries, Month OTH01 Non-Employee Expenses: G3 Location, E1 Departments, Month Note: G3 Location has a G2 Country as a parent The module for the user parameters is: And the metrics required are: Margin Salary Bonus Rent Utilities We could solve this problem without using a line item subset: Create a list (Reporting Metrics) containing the list items above. Create a module with the following dimensions. Reporting Metrics G2 Country Users The formula for Month 1 is: IF ITEM(Reporting Metrics) = Reporting Metrics.Margin THEN 'REV03 Margin Calculation'.Margin[LOOKUP: 'SYS11 Time Variance Reporting'.'Month 1'] ELSE IF ITEM(Reporting Metrics) = Reporting Metrics.Salary THEN 'EMP03 Employee Expenses by Country'.Salary[LOOKUP: 'SYS11 Time Variance Reporting'.'Month 1'] ELSE IF ITEM(Reporting Metrics) = Reporting Metrics.Bonus THEN 'EMP03 Employee Expenses by Country'.Bonus[LOOKUP: 'SYS11 Time Variance Reporting'.'Month 1'] ELSE IF ITEM(Reporting Metrics) = Reporting Metrics.Rent THEN 'OTH01 Non Employee Expenses'.Rent[LOOKUP: 'SYS11 Time Variance Reporting'.'Month 1'] ELSE IF ITEM(Reporting Metrics) = Reporting Metrics.Utilities THEN 'OTH01 Non Employee Expenses'.Utilities[LOOKUP: 'SYS11 Time Variance Reporting'.'Month 1'] ELSE 0 I won’t repeat the formula for Month 2, as it is effectively the same, just referencing the Month 2 line item in the source. You can see, that even for a small set of metrics, this is a large complex formula, going against best practices. So, let’s not do that. Create the line item subset as before. For multi-module line item subsets, it is best practice to use Multi> to represent the various modules. Open the line item subset and chose the three modules. Create a staging module (this is best practice following the DISCO principle), with the following dimensions. LIS: Multi>Variance Reporting G2 Country Time (Months) Add a line item (Data) and enter COLLECT() as the formula. Set the Summary method to None; we do not need subtotals in this module. Create a mapping module, dimensioned by Reporting Metrics. Add a line item formatted LIS: Multi>Variance Reporting. Map the lines accordingly. In the reporting module from above, change the Month 1 and Month 2 line item formulae to. 'REP05 Variance Report Staging'.Data[LOOKUP: 'SYS11 Time Variance Reporting'.'Month 1', LOOKUP: 'SYS12a Reporting Metrics Mapping'.Mapping] 'REP05 Variance Report Staging'.Data[LOOKUP: 'SYS11 Time Variance Reporting'.'Month 2', LOOKUP: 'SYS12a Reporting Metrics Mapping'.Mapping] Note, this time we are using LOOKUP rather than SUM because the source dimension doesn’t match the dimension of the mapping module. I think you’ll agree that the formula is much easier to read and it is more efficient. However, we can do even better; but note that there now are two ‘lookups’ in the formula. The more “transformations” there are in the formulae, the more work the engine needs to do. We can remove one of these by changing the target module dimensionality. Copy the reporting module from above. Remove the formulae for Month 1 and Month 2. Replace Reporting Metrics with LIS: Multi>Variance Reporting as the dimension (applies to). Add the following formulae for Month 1 and Month 2 respectively. Month 1 = 'REP05 Variance Report Staging'.Data[LOOKUP: 'SYS11 Time Variance Reporting'.'Month 1'] Month 2 = 'REP05 Variance Report Staging'.Data[LOOKUP: 'SYS11 Time Variance Reporting'.'Month 2'] Note, only one lookup is needed in the formula. Filters Another use case that line item subsets can be used for is filtering. And this functionality has nothing to do with staging data or mapping modules. It is possible to filter line items and these can also be filtered based on other dimensions too. Use Case example: Based on user-entered settings, for the reporting module (REP03) we want to show different line items for each year and version. We already have set up the Years to Versions filter module We now want to set up the user-driven parameters. To ensure that the users’ settings do not affect each other, we need to use the system generated Users’ list. Create a line item subset based on REP03 Select all line items Create a new module with the following dimensions: LIS: REP03 Filters Users Versions Add a single line item (Show?) formatted as a Boolean Enter values as you wish Note that Employee expenses and Other Costs are not available to check. This is because, in REP03, they are a simple aggregation and are shown as Parents of the other line items. So, how do we resolve this? You can “trick” the model by turning these setting off. The subtotals are now available to check in the filter module. It is worth noting, be careful when doing this. If you are using the line item subsets as a dimension in a data entry module, the totals will not calculate correctly. See Final Thoughts for more details. To set up the filter In REP03, set the following filters The module will now filter line items and years when the version page selector is changed. Note the subtotals work correctly in this module because it is not data entry. Dynamic Cell Access Line item subsets can be used in conjunction with Dynamic Cell Access to provide very fine-grained control over data; again, without any mapping modules or COLLECT() statements Use Case Example: In the following module The following rules apply: Bonus % is set by the central team so it needs to be read only. All metrics for Exec are not allowed to be edited. Car Allowances are not applicable for Production. Phone Allowances are not applicable for Production, Finance or HR, and the allowances for Sales should be read only. To set up the access: Create a line item subset based on EMP01 Select all line items Create an Access Driver module with the following dimensions: LIS: EMP01 DCA G2 Country E1 Departments Add two Boolean formatted line items Read? Write? Enter the values as below Now in EMP01 assign the Read Access and Write Access drivers to the module The module now looks like this: Line Items Subsets with Line Item Subsets I mentioned at the outset that you can lose formatting when using a line item subset. However, in some cases, it is possible to keep formatting along with calculations Use Case Example: Using the values from REP03, we want to classify Sales and Costs and then calculate a cost % to Sales. Yes, we could do this in the module itself as a separate line item, but we also want to be able to reclassify the source line items from a dashboard using mappings rather than change the blueprint formula. We also want to maintain formatting. For this example, I have just changed the styles to illustrate the point Create a line item subset based on REP03. Create a staging module with the following dimensions: LIS: REP03 Cost% G2 Country Time (Years) Add a line item call Data and enter COLLECT() as the formula and set the Summary method to None. Create a second line item subset based on REP10 (the target module). Create a mapping module dimensioned by the LIS: REP03 Cost% Create a line item formatted as LIS: REP10 Map the lines accordingly In the target module set following formula for both Sales and Costs line items (Yes, it is the same formula for both!) 'REP09 LISS vs LISS - Staging'.Data[SUM: 'SYS20 Cost% Mapping'.Mapping] Note the formatting is preserved. Version Formula Finally, I want to mention a piece of functionality that is not well known but very powerful; Version Formula. Utilizing line item subsets in conjunction with versions, Version Formula extends the ‘Formula scope” functionality. It is possible to control formulae using Formula Scope, but there are limited options. Use Case example: Let’s assume that we have actuals data in one module, the budget data in another and we want to enable the forecast to be writeable. The current version (in the versions setting) is set to Forecast For this example, there is only one line item in the target module, but this functionality allows the flexibility to set different rules per version for each line item Create a line item subset based on the above and select the line item(s). Now in the blueprint view of the target module click Edit>Add Version Formula. Now choose the Version to which the formula applies. You will now see a different formula bar at the top of the blueprint view. Enter the following formula: 'DATA01 P&L Actuals & Budget'.Revenue Repeat the above for Budget with the following formula: 'REV03 Margin Calculation'.Revenue Note that now at the top, you can see that there is a Version Formula set. Final Thoughts We mentioned the aggregation behavior and the ‘Is Summary’ setting earlier. Let me show you how this and the construction of the formulas affect the behavior of the line item subset We will use the following module as an example. This module is only used to set up the line item subset, so no dimensions are needed. Note that the subtotal formulae are simple aggregations. This means the subtotal lines: Calculate correctly when used as a dimension in a module. Are not available for data entry. The following module is dimensioned by the line item subset to highlight 1. and 2. above. If we decide we don’t want the Employee costs in the line item subset, two things happen: The indentation changes for the detailed cost lines because they are now not part of a parent hierarchy on display. The Costs subtotal doesn’t calculate. This is because the Costs subtotal needs the intermediate subtotals to exist within the line item subset. To mitigate the latter point there are two remedies. Include the subtotals and hide them – The lines are still calculating and taking space. If possible, adjust the formula structure. Remove the subtotals formula. Add in the Costs formula as to use the detailed items; no intermediate totals. Re-add the subtotal formulas. Note the 'Parent' and 'Is Summary' settings, the Costs subtotal now calculates correctly. If we change the formulae to be something other than simple addition, you will see that calculation is fine in the source module, but not in the line item subset module. Why is this? Remember the 'Is Summary' setting we changed in the Filters section when we adjusted the formula the 'Is Summary' is now unchecked This means that the line item subset doesn’t treat the line as a calculation, hence the data entry 0 shown instead. If your costs need to be positive (as in this example), it is possible to calculate correctly using a ratio formula. This works for normal line items/lists as well as line item subsets. See Changing the sign for Aggregation for more details Additional Resources

You may have heard about a model called a data hub, but perhaps you aren’t confident that you understand the fundamentals, primary functions, or considerations when architecting one. There are three main advantages to incorporating a data hub: Single source of truth: Stores all transactional data from the source system. Data validations: Ensures all data is correct and valid before the data gets to the spoke model(s). Performance: It is always faster to load data from a model rather than a file. Additionally, the administrator can ensure the correct granularity of data in the spoke model(s) when using a data hub. For example, the source system may only contain transactional data at the daily level, but the planners may need the data aggregated to the month. The data hub can summarize the data and export only the data needed. The following information is designed to further define a data hub and support you in your journey of building your own. Table of Contents Definition of the Data Hub First, we need to define what a data hub is. This can be split into four sections: Use cases: The data hub should be the first model built, whether you have a single use or multiple use cases. The data should be automatically refreshed on a schedule, whether it is nightly, weekly, monthly, etc., from the source system—often an Enterprise Data Warehouse (EDW). All modules and views that create hierarchies or lists should be stored in the data hub, which enables your models in having one version of truth, as well as reducing the duplication of data. Model connectivity: Anaplan Connect , one of our 3 rd party vendors (Informatica Cloud, Dell Boomi, Mulesoft, or SnapLogic), or our REST API can be used to automate the loading of data to the data hub from the source system, as well as transferring data from the data hub to the spoke model(s). Additionally, transitional data should not be loaded directly into the spoke module, especially if there is a large volume of data. Functions: Often, simple ETL (Extract, Transform, and Load) functions can be utilized within your data hub to transform the data for the spoke model(s). This is helpful when consolidating data from multiple sources where you have different “codes” and need a mapping module to ensure the correct data gets mapped correctly. Team: The management of the data hub should have a designated team of experts who understand what data is stored in the data hub (to ensure duplication doesn’t happen), as well as the how and when the data gets loaded. Anaplan Architecture with a Data Hub There are several ways your Anaplan architecture could look, depending on the number of workspaces you currently have and the type of security your company requires. The following are illustrations of common architectures. Master Hub Model: Across Workspaces The most common, and recommended, architecture is when the data hub is in its own workspace. Not only does this have the advantage of not interfering with the other models, but it also adds an additional security layer, with a segregation of duties. In this view, the Anaplan Workspace Admin(s) can limit the access to the data hub workspace to only the people who require it. Master Hub Model: Within a Workspace The simplest depiction is where your data hub is within the same workspace as your spoke models. While this can be accomplished, it is not best practice as there is no segregation of duties and there is a possibility, upon heavy loads from the source system, of performance issues. Additionally, when adding users, the Anaplan Workspace Administrator (Admin) would need to ensure users don’t have access to the data hub, as well as any users of the data hub not having access to the spoke models Multiple Data Hubs Finally, the data hub doesn’t necessarily have to be the only model in the workspace. You can have additional data hubs, if needed. Factors to Consider When Implementing a Data Hub There are six main elements to think about when architecting a data hub: User stories. Source systems. Lists. Modules. Data validation. Exporting data to spoke model(s). User Stories One of the cornerstones of The Anaplan Way is data (process, model, and deployment being the others), which is critical to all implementations. You will need to know what data is needed for a certain use case. Consider the following, common, data questions that need to be answered in order to be successful: What granularity of the data is needed? How much history is needed? How much history do you have? Does the source system only have transactional data, but the use case needs the data at the month level? Can the source system do the aggregation for you? After all data questions have been answered, shift your focus to the source system and consider the following: Consider the source system. Where is the data coming from? What is the source system, and is it a trusted environment? Is it Excel? Typically, you should stay away from Excel as the “source” because Excel cannot be audited. Define the data source owners. Who has access to this data? Who is preparing it? Are they part of the project? These are often-overlooked questions that are critical to success. Ideally, the data source owners need to be part of the project from the start to understand the file specifications and prepare the initial load of the data, as well as towards the end of the project to do a final load of the data. Define file specifications. How many files will be needed? Typically, you will need master data, as well as transactional data. Instead of having one file with all of this data, determine if the data can be split between different files (one for transactional, one for the unique members of the master data). It will be better for Anaplan (for performance reasons) to split these to reduce warnings during the data load process. Analyze the data. Understand what makes each record unique (date/period and transactional amounts should not be part of this), and make sure the data owners don’t give you everything (Select * From Employee) when you only need five columns. Remember, it is better to ask for additional columns midway through the project than getting all columns in the beginning and only using a select amount. Consider custom codes in the source system. Find more on this in the transactional lists section. This is a great trick for transactional data. After you have analyzed the data to understand what makes each record/row unique, concatenate the “codes” of the metadata into one transactional code, but remember, you will need to be under the 60-character threshold. Define the schedule. When is the data available? Is the data on a certain schedule? What is the schedule required with this use case? Determine the ETL medium to be used. Will Anaplan Connect be sufficient, will one of our 3 rd parties be used, or will a more custom application be needed, such as REST API? Does your company already have this experience inhouse, or will training be required? These will need to be factored into all data stories. Transactional Usually, the largest lists are those containing transactional data. There can be millions of transactional ID’s with several list properties defined. First, properties should not be defined on a transactional list (or any list, except for Display Name, as they do get accounted for in the workspace memory). Secondly, instead of loading metadata to list properties (Cost Center and Account as properties), try to figure out a way to incorporate them into the code. If the transactional data is defining a transactional amount at the intersection of Cost Center and Account for a particular month, attempt to use the code of the Cost Center and the code of the Account concatenated together (0100_57000). Not only will this decrease your list size, but it will also create a healthier model. In the below example, the model builder did not create a custom code, but rather used a combination of properties to make the record unique, which included the date/period, as well as the transactional amount. Notice the original number of records vs. the number of records after a custom code was created. By incorporating the date/time period, as well as the transactional amount, it inflated the list size exponentially based on the number of years that were loaded. Doing this not only caused the model to be bigger, but also caused poor model opening performance. See the Appendix for a simple worked example to explain further. Learn more about sparsity in the two-part series The Truth about Sparsity: Part 1 and The Truth About Sparsity: Part 2 . Flat Lists Similar to transactional lists, flat lists are not part of a hierarchy and are a series of records grouped in a list, like Products, Companies, Cost Centers, or Employees. These are your “legends” or “anchor” for all metadata about this unique record. Again, the only property that should be defined is a Display Name, if needed. It is best practice, from a model builders’ perspective, to suffix the name with “Flat” or “- Flat”. This helps identify whether the list is part of a hierarchy or flat list (Employee – Flat, Cost Center – Flat, Product – Flat). These lists can be used for data validation, which will be described later in this article. Modules Ideally, you should have three types of modules in the data hub: Transactional: A Transactional module will store the transactional data by the time series, whether that be by day, week, month, quarter, or year. The only data, or line items, should be transactional data. No other line items should be defined. Additionally, to keep the size down, make sure the summaries on the line items are turned off, or None, as there is no reason to sum the data within the module. System: System (SYS) modules, or the “S” in DISCO , do not have time associated with them and should only be dimensionalized by the same list (Employee Flat, Cost Center Flat, Product Flat). These modules store the metadata or attributes about the list item that doesn’t change over time, for example the employee’s start date. Another example of a SYS module would be any kind of mapping that is required, whether it be SYS Time Filter module or a mapping from one source system to another. Export modules: If the data from the source system is being loaded at a lower granularity than needed in the spoke model(s), export modules can aggregate the data to the specified need (month, quarter, or year level), which will lead to more efficient data load performance to the spoke model(s). Additionally, it is better to only load the granularity of data needed instead of loading all data to the spoke model, but only using a portion of it. Loading Data vs. Using Formula’s in SYS Modules If you can devise a custom code where all of the attributes of the data are accounted for, you can greatly increase the performance of your data load, especially on very large data volumes. It is actually faster to use formulas to derive the data from the custom code than it is to load the data. Why? A couple of reasons. First, when data is loaded, the load is triggering the change log, and every change is being recorded in the model history . Second, loading data to another module is an additional action. If you didn’t need that action, you would save processing time. In the example below, the exact same data was loaded four different ways: Import Properties to a List: A list was created with all attributes, including the transactional data, and was loaded to list properties (not best practice and against DISCO). Import to List, Attribute, and Trans: A list was created, the transactional data was loaded to a transactional module, and all of the attributes were loaded to a SYS Attribute module. Import to List, Trans, Calculate Attribute: A list was created, the transactional data was loaded to a transactional module, but the SYS Attribute model was calculated using two different methods: One Line Item: Using FINDITEM() with several functions parsing out the information from within the FINDITEM(). For example, FINDITEM(Cost Center, RIGHT(LEFT(Trans Details.Code, '2nd Group’), 3)). Multiple Line Items: Parsing of the member spread across multiple line items and using FINDITEM() with only the list and code as the parameter. First, you do the parsing to get the correct piece of the code (one line item), and then the FINDITEM() of that code (2 nd line item). Load Performance Notice, the best performing data load was the last one, Import to List, Trans, Calculate Attribute (multiple line items), where the parsing out of the data was spread over multiple line items. This is due to the fact that the data load was able to take advantage of Anaplan’s multithreading capabilities. The worst performing data load occurred when data was loaded to the Attribute module because, due to the sheer size of the data, a save had to be performed. Exporting to Spoke Models One of the most important concepts to remember when exporting data is to use a view from a module. Lists should not be exported because you lose control over what you export. It is either all or nothing. By using views, you can employ a filter (should always be a Boolean) to render exactly which data needs to be exported. If you need more than one filter, combine both into one line item and use that line as the filter. You will have much better performance if you are only using one Boolean line item as a filter vs. having multiple filters defined. Another important concept to remember is to only export detailed information, as there is no reason to export parent information (quarter, year, etc.). Not only will you get warnings when exporting parent information, but the performance of the export will decrease because the system will have to create a debug log. The goal is to make sure a debug log is not created, all green checks, so if there ever is an issue, you will know it truly is an issue that needs attention. Line items in the data hub formatted as text should not be exported as text, but actually as list formatted line items in the spoke model (text->list formatted line item). The goal is to reduce the number of text formatted line items in the spoke model. Some say they need to do validation in the spoke model, therefore they need to import the data as text. Actually, this is false, because the validation should have already been done in the data hub, so there should be no need to do the validation again. Lastly, you should think about what really needs to be exported. Do you really need to export historical data that hasn’t been changed? Instead, just export the newly loaded data, or delta data. This can be accomplished by using one of two methods: From the source system, request IT to only send the updated information, not the full load every time. Additionally, request IT to create a column in the source file with a hardcoded value of “TRUE.” This will tell Anaplan which row is new or has been updated and can be used as a filter for an export. Just know, before the import of the source data gets loaded, make sure the first action within the process clears out the previous true records (set this up via a view using a filter where the view only shows members with a value of true). Utilize the current period function to only export the current period data. In the SYS Time Filter module, create a line item named Current Period with the formula CURRENTPERIODSTART(). In the export views, filter the data on this line item. Tips and Tricks A few of tips and tricks to be aware of include the following: Hierarchies should not be in the data hub. Analytical modules should not be in the data hub. Do not delete and reload lists. Data Validations Model Why should hierarchies not be in the data hub? To answer that question, you need to understand why hierarchies are used in the first place. Essentially, hierarchies are only needed to aggregate data for analytical purposes, and since users will not normally login to the data hub, the lists essentially take up space. With that said, it is perfectly okay to create the hierarchies for testing purposes to ensure your actions from the meta modules are building the hierarchies correctly, but as soon as the actions are working correctly and have been verified, you can remove the list structures from the data hub. A case can be made that certain implementations may need the hierarchies created in the data hub for validation purposes of several sources. If this is the case in your implementation, just be sure to only use the hierarchies for validation purposes. In addition to the above, there are two more reasons to not have hierarchies built in the data hub—cluttered data, and spoke models that pull data from the lists. Data hubs need to be clean and clutter free to ensure optimal performance, which also makes it easier for the administrators to understand exactly what data is stored in the data hub. Additionally, when you have lists—especially hierarchical lists—spoke model builders will sometimes build their lists from the lists within the data hub instead of from a view. It is best practice to always build lists from views from within a module so the action can benefit from filters (there are no filters when importing from lists). Analytical modules should not be in the data hub since end users don’t normally access the data hub. There really isn’t a reason to have products by versions by time in the data hub, that belongs to the spoke model. Remember, the data hub should only be used to store data from the source system(s). Within your nightly data load process, do not delete and reload data, including the list structures. If you have a proper code, you shouldn’t need to do this. Additionally, not only does this impact the overall performance of the process (adding an additional action to delete the list, which then deletes all data associated with that list), but the process is essentially filling up the change log with the exact same data that it had before the delete. When a certain threshold is surpassed, the model will require a save, thus taking up even more time. Ultimately, you are forcing the model to re-aggregate all of the data, instead of just the new data. Lastly, if you know you will have to do a lot of transformations on your data (consolidating multiple source systems or your data is not clean), think about creating a Data Validations model. This model’s sole purpose would be to clean the data and then feed the data to the data hub, thus keeping the transformations to a minimum in the data hub as well as keeping the data hub clean. Worked Example Use Case: Transaction Data is by Store and SKU and Month Bad Way The code for the Transaction list is a three-part code Store_SKU_Month Attributes for Store, SKU and Month are imported as Text and matched against the Store list, SKU list and Time period respectively An additional line item is needed for the Store and SKU code (for export). This is the screenshot of the bad way: Notice the repetition of the attributes. STR07 and SKU031 are repeated each month. Good Way Two data files Unique combinations of Store and SKU (two-part code) Store SKU code by month for the quantity. The transaction details are stored in a module dimensioned by Transactions The Store and SKU attributes are calculated using the “_” delimiter The quantity is stored in a module dimensioned Transactions and by month The additional line item is needed for the Store and SKU code (for export). This is a subsidiary view in the module as it is not dimensionalized by Time. These are the screenshots of the good way: Below lists out the breakdown of the model in terms of List size, Line items and the associated member usage of the various structures. The main reasons for the improvement are because lists themselves account for approximately 500b for each member and also there is repetition of the attributes per “month” in the transaction data (as mentioned above). Hopefully, this article has shed some light on data hubs, how they should be used, and what you can do to ensure they perform at their peak level. Remember, analyze the data to understand what makes the row unique and use that as the code. Every list should have a code—every list! Additional Resources

PLANS is the new standard for Anaplan modeling—“the way we model.” This covers more than just the formulas and includes and evolves existing best practices around user experience and data hubs. It is a set of rules on the structure and detailed design of Anaplan models. This set of rules will provide both a clear route to good model design for the individual Anaplanner and common guidance on which Anaplanners and reviewers can rely when passing models amongst themselves. In defining the standard, everything we do will consider or be based around: Performance – Use the correct structures and formula to optimize the Hyperblock Logical – Build the models and formula more logically – See D.I.S.C.O. below Auditable – Break up the formula for better understanding, performance, and maintainability Necessary – Don’t duplicate expressions. Store and calculate data and attributes once and reference them many times. Don't have calculations on more dimensions than needed Sustainable – Build with the future in mind, thinking about process cycles and updates The standards will be based around three axes: Performance - How do the structures and formula impact the performance of the system? Usability/Auditability - Is the user able to understand how to interact with the functionality? Sustainability - Can the solution be easily maintained by model builders and support? We will define the techniques to use that balance on the three areas to ensure the optimal design of Anaplan models and architecture. D.I.S.C.O As part of model and module design, we recommend categorizing modules as follows: Data – Data hubs, transactional modules, source data; reference everywhere Inputs – Design for user entry, minimize the mix of calculations and outputs System – Time management, filters, list attributes modules, mappings, etc.; reference everywhere Calculations – Optimize for performance (turn summaries off, combine structures) Outputs - Reporting modules, minimize data flow out Why build this way? Performance Fewer repeated calculations Optimized structures and formulas Logical Data and calculations reside in logical places Model data flows can be easily understood Auditable Model structure can be easily understood Simplified formula (no need for complex expressions) Necessary Formulas and structures are not repeated Data is stored and calculated once, referenced many times, leading to efficient calculations Sustainable Models can be adapted and maintained more easily Expansion and scaling simplified Recommended Content: Video Performance Dimension Order Formula Optimization in Anaplan Formula Structure for Performance The Truth About Sparsity: Part 1 The Truth About Sparsity: Part 2 Data Hubs: Purpose and Peak Performance To Version or Not to Version? Line Item Subsets Demystified Logical Best Practices for Module Design Data Hubs: Purpose and Peak Performance Auditable Formula Structure for Performance Necessary Reduce Calculations for Better Performance Formula Optimization in Anaplan Sustainable Dynamic Cell Access Tips and Tricks Dynamic Cell Access - Learning App Personal Dashboards Tips and Tricks Time Range Application Ask Me Anything (AMA) sessions The Planual The Planual Rises

The process of designing a model will help you: Understand the customer’s problem more completely Bring to light any incorrect assumptions you may have made, allowing for correction before building begins Provide the big picture view for building This process includes the following steps: Let’s take a closer look at each of the steps, why it is important to the overall process, and the results of completing each step. Note: This document uses the terminology end-user experience, pages, and boards when referring to dashboards. For more information, check out Training on the New UX. Understand the Requirements and the Customer’s Technical Ecosystem Purpose: To solve a problem, you must completely understand the current situation. Performing this step provides this information and the first steps toward the solution. Process: When you begin a project, you gather information and requirements using tools that include: SOW. Definition of the project scope and project objectives/high-level requirements. Manifesto. Goal of the project—big picture view of what needs to be accomplished. IT ecosystem map. Which systems will provide data to the model and which systems will receive data from the model? What is the Anaplan piece of the ecosystem? What other Anaplan models will this model connect with? Current business process definition. Remember that if the current process isn’t working, it needs to be fixed before design can start. The process steps should be clearly defined for any changes to an existing process or a new process. The organization should plan communication and education programs to inform and train the end-users on the process changes. Business logic definition. What key pieces of business logic will be included in the model? Distributed model needed? High user concurrency. Security where the need is a separate model. Regional differences that are better handled by a separate model. Model size and end-user and roles in the process. Does the input or start of a second process rely on the output of another? Do the model outputs need to be real-time? Reporting and process steps should be clearly defined. Is the organization using ALM to effectively manage development, testing, deployment, and maintenance of applications? This functionality requires a premium subscription or above. User stories. These have been written by the client—more specifically, by the subject matter experts who will be using the model. Results: Understand the goal of the project. Understand: The organizational structure and reporting relationships (hierarchies). Where data is coming from and how much data clean-up might be needed. What data is organized into categories (for example, product families) or what data relationships exist that need to be carried through to the model (for example, salespeople only sell certain products). What lists currently exist and where are they are housed. Which systems (including other Anaplan models) the model will either import from or export to. What security measures are expected. What time and version settings are needed. Plan the End-User Experience for Each Role Purpose: This is an important step in the model design process. Ultimately, the information or data that the end-user needs to make a good business decision is what drives the entire structure of the model. Process: On some projects, you may be working with a project manager or a business consultant to flesh out the business process for the user. You may have user stories, or it may be that you are working on design a bit earlier in the process and the user stories haven’t been written. In any case, identify the user roles, the business process that will be completed in Anaplan, and create a high-level design of the user experience. Verify the design with the users to ensure that you have the correct starting point for the next step. Front to back design has been identified as the preferred method for model design. This approach puts the focus on the end-user experience. We want that experience to align with the process, so users can easily adapt to the model. During this step, focus on: User Roles – who are the users? You may want to create a RACI matrix to identify who is responsible, who is accountable, who needs to be consulted, and who must be kept Informed. Identify the business process that will be done using Anaplan. Review and document the process for each role, focusing on the main steps. You can document the process in any way that works for you. Here is a step-by-step process for how to document the business process: Gather information from the end-users. For each different role, ask these questions: What is the start of the process? What is the result or output of the process? What are the process inputs (what data are needed) and where do they come from? What are the actions the user takes? Write these using a verb and an object—for example, approve the request, enter sales amount, add new hire, etc. No need to organize during this step. Use post-it notes to capture them. Keep this discussion focused on the process as it is completed most of the time. Use this information to document the process, putting the actions in the correct order. You can use any process mapping software or PowerPoint. If there are multiple roles, use swim lanes to identify the roles. Create a timeline to map the user roles to see if there are any overlaps or dependencies that could cause concurrency issues. These are often missed because concurrency is often thought about by role, not across the whole user base. Verify the process with the end-users and Subject Matter Experts (SME) and make any corrections. Ask the end-users to sketch out what they would like to see on their pages and boards. Many processes will require multiple pages and boards. Keep this at a high level. Some questions to guide this discussion include: What data do you need to see? What types of charts or graphs might help you better understand the data? What actions do you need to take? What decisions do you make? When making decisions, is there data that you need to compare? Results: List of user roles. Process steps for each user role. Timeline mapping of user roles. High-level page and board designs for each user role. Use the User Experience Design From the Previous Step to Determine What Output Modules are Needed in the Model. Purpose: Think D.I.S.C.O. The Output modules are needed to support the end-user experience or export to another system. This is what should guide your design – all the Input modules and Calculation modules are added with the purpose of providing these output modules with the information needed for the pages and boards or export. Process: Some questions to help you think through the definition of your Output modules: What information (and in what format) does the user need to make a decision? If the end-user is using pages and boards for reporting purposes, what information is required? Are there modules that will serve to move data to another system? What data and what format is necessary? Results: List of outputs and desired format needed for each page or board. Identification of how the data should be categorized on the page or board, or what hierarchies are needed. For example, employees are organized into departments, departments into business units, etc. Identify What Data Is Available and Needed to Support the Output Modules. Determine the Data Modules the Model Requires. Purpose: Identify what data the model requires and where it is held. Determine the hierarchies or structures needed in the model. Process: Organize the data using Data modules, using each module to hold similar types of data, for example, employee, product, or sales data. Results: Lists needed in the model. Hierarchies needed in the model. Data and where it is coming from. Data modules identified. Identify What Inputs From End-Users Are Needed Purpose: Planners want to be able to see how changes will affect overall results. By entering new data, planners can try different scenarios and make decisions based on data. What if manufacturing costs go up? What effect do higher shipping costs have on the bottom line? Process: Look at the end-user experience designs from step 2 and identify the data types that end-users will be able to add or change. Results: List of pages or boards where data input is required. Input modules identified. These modules should not include a lot of calculations. Determine What Calculations Are Needed to Transform Source Data and Inputs to the Data Needed for Outputs. Identify the Calculation and Systems Modules Needed. Purpose: Transform data from the Data and Input modules to the data needed for Output modules, using business rules, logic, and formulas. Process: Some questions to use to think through what calculations are needed: What is the output? How should the output data be formatted? What dimensionality is needed? What data is available? Is the source data in the same dimensionality as the output, or is it different? If the dimensionality is different, what intermediate steps are needed? What Calculation or System modules are needed? Will the source data need to be converted to a different data type? Use Anapedia to locate functions. The reverse lookup page provides a list of behaviors and descriptions in plain language of many Anaplan functions. You can also use the search function in Community. As you are thinking about the calculations needed, pay attention to functions that are used in multiple calculations. Examples might include calculations for pulling data from the current period, the parent in a product hierarchy, or pulling data from a forecast version. These can be included in Systems modules and entered once and referenced many times. Systems modules can also hold lists and list item attributes. Also, think about how to organize the calculations into Calculation modules. Keep in mind the different types of data (for example, Revenue, Employee, Other Costs) used in the model and keep the calculations for each type together. Results: List of Calculation and System modules. List of possible calculations (to be tested when model building). Using Lucid Charts (Or Some Other Tool), Create Your Model Schema Purpose: Provides a graphic representation of the model design that is used for communicating with the model-building team. The model schema is required documentation for the project. Process: Designing your model using a schema means that you must think through all the information you have about the current situation, how it all ties together, and how the model provides what is needed to meet the needs of the end-users. Begin by sketching your ideas on a whiteboard. Whiteboards are excellent collaboration tools and model design is a collaborative process. Be sure to: Identify the Input, Outputs, and Calculation modules needed for each functional area. Identify the Data and Systems modules needed for each functional area. Show the data flow between the functional areas. Identify time and versions where appropriate. Transfer the final version to an electronic format. Results: A model schema that provides the big picture view of the solution. It should include imports from other systems or flat files, the modules, or functional areas that are needed to take the data from the current state to what is needed to support the end-user experience that was identified in step two. Time and versions should be noted where required. Include the lists that will be used in the functional areas/modules. Your schema will be used to communicate your design to the customer, model builders, and others. While you do not need to include calculations and business logic in the schema, it is important that you understand the state of the data going into a module, the changes or calculations that are performed in the module and the state of the data leaving the module, so that you can effectively explain the schema to others. Final Design Check Purpose: A final check of your model to ensure that it is well-designed. We recommend that you have a peer check over your model and provide feedback. It is easy to fall into using the functions you are familiar with, which may or may not be the best solution. You will build your model design skills by participating in a Model Design Check-in, which allows you to talk through the tougher parts of the design with a peer. This check-in is included as part of The Anaplan Way process. How-to: When your schema is complete, give it a final check to ensure: You have used DISCO to organize your modules and each module has a defined purpose. Your schema has been kept simple. “Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius — and a lot of courage to move in the opposite direction.” ― Ernst F. Schumacher “Simple is better than complex.” –Tim Peters The model aligns with the manifesto. The business process has been defined and works well within the model.

As an ad planner, sales representative, or merchandise planner, you might be interested in planning your sales or invoices using tier pricing. This solution will address a user-story that allows you to set the prices by tier range and automatically lookup the right price based on the units ordered. This solution was first introduced by David Smith in his Community Site response to solve a customer satisfaction score and inspired by a question asked by Vignesh Mohan in this post.

General Recommendations It is important as model calculations increase to ensure that the calculations and structures are as optimal as possible. Easy checks include: Ensure the dimension order in the "applies to" is consistent. Consider if all of the dimensions for the calculations are necessary. Reduce the number of line items with the same formulas (calculate once, reference many times). Consider the use of subsets and line item subsets. Use a balanced approach to sparsity. Ensure "text joins" are optimal. See the following articles for more information on the above: The Truth about Sparsity (part 1) The Truth About Sparsity (part 2) Best practice for Dimension ordering Reduce calculations for better performance Formula Optimization in Anaplan Customer Requirements General recommendations also include, whenever possible, challenging your customer’s business requirements when the customer requires a large list (>1M), a large amount of data history (this should be kept out of the planning model if possible), and high number of dimensions used at the same time for a line item (>5) Other Practices Once these general recommendations have been applied, you can optimize performance in different areas. The articles below will expand on each subject: Imports and exports and their effects on model performance Dashboard settings that can help improve model performance Formulas and their effect on model performance Model load, Model Save, Model Rollback and their effect on model performance

As a business operations manager on the Anaplan on Anaplan (AoA) team—an internal team, focused on bringing Connected Planning to life within Anaplan—I help to oversee our internal Anaplan model ecosystem and assist in the solutioning and development of Anaplan models across all of our functional business groups. As Anaplan's largest customer, one of the numerous requirements we must address is user access and security. Utilizing Anaplan's user roles functionality typically gets the job done for granting users access to specific models. Occasionally, we must go one step further and leverage Anaplan's selective access feature. Roles and selective access are powerful tools and address our needs nearly all of the time. However, as we scale our own use of Anaplan, we have begun to encounter the need to provision user's access to lists based on multiple criteria, rather than just a single condition. In Real Life A real-life user provisioning challenge we’ve encountered is in our headcount planning model. As this model provides real-time reporting on our employees, there are inherent sensitivities and considerations around who can see information for specific employees—taking into consideration visibility to things like compensation and personally identifiable information (PII). We have multiple use cases built out within the model, including recruiting capacity and analysis, attrition reporting, hiring reporting, etc., and the access to specific employee data depends on the end user of the model. Sample employee roster: Joey manages Usain, Eluid, and Meb; Americas Geo; HR Cost Center. In this example model, we have our complete employee roster included. If an HR business partner accesses the model, we want them to see only employees that are tagged to the functional area they support (e.g. finance, sales). Additionally, if a business manager goes into the model, they should only see information for employees where they are the manager, or employees downstream on their management chain. But wait! If the HR business partner is in Europe, they shouldn’t be able to see PII fields for their employees. Do you see how this could get complicated quickly? Additionally, some dashboards that contain non-sensitive employee information are perfectly fine to open up broadly to all users, while others contain sensitive data we need to provision. What’s Next So, how do we handle this? We can’t provision access by roles because all of the aforementioned users need access to the same modules/dashboards as it relates to the employees they manage. Additionally, no single user should be able to see all data for all employees. Selective access could be considered as a solution, but given the levels of complexity and multiple logical drivers—as well as the requirement to not hide reporting of non-sensitive data for employees—that option also has limitations. Enter Dynamic Cell Access (DCA). Since DCA allows us to base read/write access off of formulae logic, it offers us the ability to layer on multiple levels of logic ahead of deciding whether or not someone should be able to read or write on a particular item in a list. It’s dynamic (who would have thought with that name?), which means it adjusts live as data within the model changes. Additionally, it offers us the flexibility to apply the provisioning logic to the exact modules we want to, rather than blanket provision users across the model. DCA In Action The following is a high-level example of how to leverage the power of DCA: Load employee roster data into Anaplan, ensuring the data contains the employee email—the same email that is used to log in to Anaplan. This allows for the mapping of Anaplan users to the employee roster. Set up a System module with the ‘applies-to’ list of the user list. User meta-data staging module: Rows represent model users (Joey, in this example) and the line items represent meta-data off of the roster module. Within this module, we can join the employee roster data and the user list to map the employee’s meta-data to their Anaplan user profile (e.g. cost center, location, management chain, etc.) Using a series of Boolean line items, we can write whatever logic we want to base our DCA on. In our example, this could include: Is HR business partner? Is Euro? Basically, this is a staging module for all of the employee meta-data we want to leverage to create our DCA drivers. Set up a second System module with the ‘applies-to’ list of whatever list you want to apply DCA against, as well as the user list. In our case, this would also be our employee roster list. Create a series of Boolean line items, testing different attributes of the User System module we just set up against the meta-data of the employees. An example would be (Employee Cost Center = User’s Cost Center). DCA logic module for the employee roster list (rows in this module): Line items represent the logic used to determine whether the user (Joey— in the page selector) can see the employee. The key here is to consolidate all of your logic into a single “Master” line item, which is on the far right. Daisy chain your conditions together as desired, with the end result being a master Boolean line item, which is the driver for whether or not a particular user has read or write access to a particular item within the list. In this dashboard you can see that the information is masked for those employees that did not meet all of the criteria identified in the master DCA line item. Select which modules you’d like to apply DCA to. The nice thing about DCA is you can go down to the line item level to map the master Boolean driver against. The incredible power of the process described above is not only the complete control over and ability to customize your user provisioning, but also that as new roster data is loaded into Anaplan, the DCA automatically adjusts itself to account for changes. So, if someone changed cost centers or a manager on an employee changed, the formulas that we set up above would be referencing the new employee meta-data, and would automatically adjust the DCA drivers, allowing for a much more hands-off, sustainable approach to user provisioning. Another inadvertent benefit we discovered with using this methodology is that Anaplan treats cells that are blank as a result of DCA drivers as being blank for filtering purposes. So, if you want to set up a dashboard that auto-filtered employees for the end user based on the logic above, you just have to add a line item hardcoded to contain values for every list item, and then filter that line item for not-blanks on your dashboards. Then you have a dynamic filter based on the user that is viewing the model…pretty slick! Take this one step further and filter for not-blanks on a line item that will always contain data for an employee, and you get completely custom reporting based on which end user is viewing the dashboards.

Filters can be very useful in model building and are widely used, but they can come at the expense of performance—often very visible to users through their use on dashboards. Performance can also hit imports and exports, which in turn may lead to the blocking of other activity, causing a poor perception of the model. There are some very simple guidelines to design well-performing filters: Using a Single Boolean Filter on a Line Item That Does Not Have Time or Versions Applied and Does Not Have a Summary Is Fastest Try to create a Boolean line item that incorporates all the filter criteria you want to apply. This allows you to re-use the line item and combine a series of Boolean line items into a single Boolean for use in the filter. For example, you may want to filter on three data points: Volume, Product Category, and Active Status. Volume is numeric, Product Category is a list formatted line item matching a user selection, and Active Status is a Boolean. Create a line item called Filter with the following formula: Volume > Min Vol AND Product Cat = User Selection.Category AND Active Status Here’s a very simple example module to demonstrate this: A Filter line item is added to represent all the filters we need on the view. Only the Filter line needs to be dimensioned by Users. A User Selection module dimension only by Users is created to capture user-specific filter choices: Here’s the data before we apply the filter: Here's the data with the filter applied: A best practice suggestion would be to create a filter module and line items for each filter part. You may want other filters and you can then combine each filter as needed from this system module. This should reduce repetition and give you control over the filters to ensure they can all be Boolean. What Can Make a Filter Slow? The Biggest Performance Hit for Filters Is When Nesting Dimensions on Rows. The performance loss is significantly increased by the number of nested dimensions and the number of levels they contain. With a flat list versus nested dimensions (filtering on the same number of items) the nested filter will be slower. This was tested with a 10,000,000 list versus 2 lists of 10 and 1,000,000 items as nested rows; the nested dimension filter was 40% slower. Filtering on Line Items With a Line Item Summary Will Be Slow. A numeric filter on 10,000,000 items can take less than a second, but with a summary will take at least five seconds. Multiple Filters Will Increase Time. This is especially significant if any of the preceding filters do not lower the load because they will take additional time to evaluate. If you do use multiple filter conditions, try to order them so the most effective filters are first. If a filter doesn’t often match on anything, evaluate whether it's even needed. Hidden Levels Act as a Filter. If you hide levels on a composite list, this acts like a filter before any other filter is applied. The hiding does take time to process and will impact more depending on the number of levels and the size of the list. Avoid Nested Rows for Export Views Using nested rows can be a useful way to filter a complex set of data for export, but, as discussed above, the filter performance here can be poor. The best way around this is to pivot the dimensions so there is only one dimension on rows and use the Tabular Multi Column export option with a Filter Row based on Boolean option. Some extra filter tips include the following: Filter duration will affect saved views used in imports, so check the saved view open time to see the impact. This view open time will be on every use of the view, including imports or exports. If you need to filter on a specific list, create a subset of those items and create a new module dimensioned by the subset to view that data.

Overview The following is a g uide for the new Statistical Forecasting Calculation Engine Models (monthly and weekly). It includes enablement videos, practice data import exercise, model documentation, and specific steps when using the model for implementations . 1. Enablement Videos & Practice Exercise # Item Details Link 1a Intro and Overview Video Model overview and review of new key features. Video Below 1b Initial Model & Data Import Steps Steps on how to setup model, product hierarchy, customer list and multi-level forecast analysis. Video Below 1c Practice Exercise—Import data to setup stat forecast Two sets of load files included to practice setup for single level product set or multi-level product set w/ customers, product and brand level. Start on "Initial App Setup" dashboard and load either Single OR Multi Level files into model, and use Import video as guide if needed. .Zip File Attached 2. Documentation # Item Details Link 2a Lucidchart Process Maps Lucidchart Process Map document includes High-Level process flow for end-user navigation and detailed tabs for each section. **Details & links also on "Training & Enablement" dashboard. Process Maps 2b High-Level Process Map PDF High-level process map PDF format. Attached 2c Forecast Methods PDFs High-level version with forecast algorithms list and overview. Detailed version which includes a slide for each forecast method, m ethod overview, advantages/disadvantages, equation and graph example output. **These slides are also included on "Forecast Methods Overview & Formulas" dashboard. Attached 3. Implementation Specifics # Item Details 3a Training & Enablement Dashboard Training & Enablement dashboard contains details on process map navigation. 3b Initial Model Setup Initial Setup: current model staged with chocolate data from data hub, execute CLEAR MODEL action prior to loading customer-specific data. 3c Changing Model Time Scale— align Native & Dynamic Time Settings If a Time Settings change is required, need to review Initial App Setup dashboard to align Native Time with Dynamic Time setup in model. 3d Monthly Update Process After initial setup, use Monthly Data History Upload dashboard to update prior period actuals and settings . 3e Single Level vs. Multi-Level Forecast Setup Two implementation options & when to use: Single Level Forecast: Forecast at one level of product hierarchy (i.e. all stat forecasts calculated at Item level). Most use cases will leverage single level forecast setup. Multi-Level Forecast : Ability to forecast at different levels of the product hierarchy (i.e. Top Item | Customers, Item and Brand level can all have stat forecast generated). This requires a complex forecast reconciliation process, review "Multi-Level Forecast Overview" dashboard if this process is needed. 3f Troubleshooting Tips Follow troubleshooting tips on Training & Enablement dashboard if having issues with stat forecast generating before reaching out for support. 3g Model Notes & Documentation Module Notes—includes DISCO classification and module purpose. 3h "Do Not Modify" Items Module notes contain DO NOT MODIFY for items that should not be changed during the implementation process. 3i User Roles & Selective Access Demo, Demand Planner, Demand Planning Manager ro les can be adjusted After Selective Access process run on Flat List Management dashboard; then users can be given access to certain product groups/brands etc. 3j Batch Processing Details on daily batch processing and how to prepare a roadmap of your batch processes – files, queries, import actions/processes in Anaplan (see attachment). 4. Videos Intro & Model Intro and Overview Video. Data Import and Setup Steps. 5. Model Download Links Monthly Statistical Forecasting Calculation Engine Weekly Statistical Forecasting Calculation Engine

Dimension Order Affects Calculation Performance Ensuring consistency in the order of dimensions will help improve the performance of your models. This consistency is relevant for modules and individual line items. Why does the order matter? Anaplan creates and uses indexes to perform calculations. Each cell in a module where dimensions intersect is given an index number. Here are two simple modules dimensioned by Customer and Product. In the first module, Product comes first and Customer second, and in the second module, Customer is first and Product is second. In this model, there is a third module that calculates revenue as Prices * Volumes. Anaplan assigns indexes to the intersections in the module. Here are the index values for the two modules. Note that some of the intersections are indexed the same for both modules: Customer 1 and Product 1, Customer 2 and Product 2, and Customer 3 and Product 3, and that the remainder of the cells has a different index number. Customer 1 and Product 2 is indexed with the value of 4 in the top module and the value of 2 in the bottom module. The calculation is Revenue = Price * Volume. To run the calculation, Anaplan performs the following operations by matching the index values from the two modules. Since the index values are not aligned, the processor scans the index values to find a match before performing the calculation. When the dimensions in the module are reordered, these are the index values: The index values for each of the modules are now aligned. As the line-items of the same dimensional structure have an identical layout, the data is laid out linearly in memory. The calculation process accesses memory in a completely linear and predictable way. Anaplan’s microprocessors and memory sub-systems are optimized to recognize this pattern of access and to pre-emptively fetch the required data. How does the dimension order become different between modules? When you build a module, Anaplan uses the order that you drag the lists onto the Create Module dialog. The order is also dependent on where the lists are added. The lists that you add to the 'pages' area are first, then the lists that you add to the 'rows' area, and finally the lists added to the 'columns' area. It is simple to re-order the lists and ensure consistency. Follow these steps: On the Modules pane, (Model Settings>Modules) look for lists that are out of order in the Applies To column. Click the Applies To row that you want to re-order, then click the ellipsis. In the Select Lists dialog, click OK. In the Confirm dialog, click OK. The lists will be in the order that they appear in General Lists. When you have completed checking the list order in the modules, click the Line Items tab and check the line items. Follow steps 1 through 3 to re-order the lists. Subsets and Line Item Subsets One word of caution about Subsets and Line Item subsets. In the example below, we have added a subset and a Line Item Subset to the module: The Applies To is as follows: Clicking on the ellipsis, the dimensions are re-ordered to: The general lists are listed in order first, followed by subsets and then line item subsets. You still can reorder the dimensions by double-clicking in the Applies to column and manually copying or typing the dimensions in the correct order. Largest vs. Smallest? This is the normal follow up question, and unfortunately, the answer is "it depends." Through research we have found that it all depends on the data within the module. Also, it can get very confusing if subsets are used; the Customer list might be bigger than the Products list, but if a subset of Customers is used that is smaller than Products, then what? Also, we don't advocate ordering the lists in the General Lists setting in size order; the lists should be ordered in hierarchical order top to bottom, so, by definition, that will be smallest to largest. So our advice is be consistent. Think about how you describe the problem. Does the business talk about Customer by Product, or Products for Customers? Agree to a convention, and stick to it. Other Dimensions The calculation performance only relates to the common lists between the source(s) and the target. The order of separate lists in one or other doesn’t have any bearing on the calculation speed.

As a model builder, you have to define line item formats over and over. Using a text expander/snippet tool, you can speed up the configuration of modules. When you add a new Line Item, Anaplan sets it by default as a Number (Min Significant Digits: 4, Thousands Separator: Comma, Zero Format: Zero, etc.). You usually change it once and copy it over to other line items in the module. Snippet tools can store format definition of generic formats (numbers, text, boolean, or no data) and by a simple shortcut, paste it in the format of the desired line items. Below is an example of a number format line item with no decimal and hyphens instead of zeros. On my Mac, I press Option + X, I type "Num..." and get a list of all Number formats I pre-defined. I press Enter to paste it. It also works if several line items were selected. The value stored for this Number format is : {"minimumSignificantDigits":-1,"decimalPlaces":0,"decimalSeparator":"FULL_STOP","groupingSeparator":"COMMA","negativeNumberNotation":"MINUS_SIGN","unitsType":"NONE","unitsDisplayType":"NONE","currencyCode":null,"customUnits":null,"zeroFormat":"HYPHEN","comparisonIncrease":"GOOD","dataType":"NUMBER"} Here is the result of a text format snippet. {"textType":"GENERAL","dataType":"TEXT"} Or a Heading line item (No Data, Style: Heading 1). ---- false {"dataType":"NONE"} - Year Model Calendar All false false {"summaryMethod":"NONE","timeSummaryMethod":"NONE","timeSummarySameAsMainSummary":true,"ratioNumeratorIdentifier":"","ratioDenominatorIdentifier":""} All Versions true false Heading1 - - - 0 This simple trick can save you a lot of clicks. While we are unable to recommend specific snippet tools, your PC or Mac may include one by default, while others are easy to locate for free or low-cost, online.

This article provides the steps needed to create a basic time filter module. This module can be used as a point of reference for time filters across all modules and dashboards within a given model. The benefits of a centralized Time Filter module include: One centralized governance of time filters. Optimization of workspace, since the filters do not need to be re-created for each view. Instead, use the Time Filter module. Conforms with the D.I.S.C.O. methodology as a 'System' module. More on D.I.S.C.O. can be found here. Step 1: Create a new module with two dimensions—time and line items. The example below has simple examples for Weeks Only, Months Only, Quarters Only, and Years Only. Step 2: Line items should be Boolean formatted and the time scale should be set in accordance to the scale identified in the line item name. The example below also includes filters with and without summary methods, providing additional views depending on the level of aggregation desired. Once your preliminary filters are set, your module will look something like the screenshot below. Step 3: Use the pre-set Time Filters across various modules and dashboards. Simply click on the filters icon in the toolbar, navigate to the time tab, select your Time Filter module from the module selection screen, and select the line item of your choosing. Use multiple line items at a time to filter your module or dashboard view.

If you have a multi-year model where the data range for different parts of the model varies, (for example, history covering two years, current year forecast, and three planning years) then Time Ranges should be able to deliver significant gains in terms of model size and performance. But, before you rush headlong into implementing Time Ranges across all of your models, let me share a few considerations to ensure you maximize the value of the feature and avoid any unwanted pitfalls. Naming Convention Time Ranges As with all Anaplan models, there is no set naming convention, however, we do advocate consistency and simplicity. As with lists and modules, short names are good. I like to describe the naming convention thus—as short as practical—meaning you need to understand what it means, but don’t write an essay! We recommend using the following convention: FYyy-FYyy. For example, FY16-FY18, or FY18 for a single year Time Ranges available are from 1981 to 2079, so the “19” or the “20” prefixes are not strictly necessary. Keeping the name as short as this has a couple of advantages: It has a clear indication of the boundaries for the Time Range It is short enough to see the name of the Time Range in the module and line items blueprint The aggregations available for Time Ranges can differ for each Time Range and also differ from the main model calendar. If you take advantage of this and have aggregations that differ from the model calendar, you should add a suffix to the description. For example: FY16-FY19 Q (to signify Quarter totals) FY16-FY19 QHY (Quarter and Half Year totals) FY16-FY19 HY (Half Year totals only) etc. Time Ranges are Static Time Ranges can span from 1981 to 2079. As a result, they can exist entirely outside, within, or overlap the model calendar. This means that there may likely be some additional manual maintenance to perform when the year changes. Let’s review a simple example: Assume the model calendar is FY18 with two previous years and two future years; the model calendar spans FY16-FY20 We have set up Time Ranges for historic data (FY16-FY17) and plan data (FY19-FY20) We also have modules that use the model calendar to pull all of the history, forecast, and plan data together, as seen below: At year end when we “roll over the model,” we amend the model calendar simply by amending the current year. What we have now is as follows: You see that the history and plan Time Ranges are now out of sync with the model calendar. How you change the history Time Range will depend on how much historical data you need or want to keep. Assuming you don’t need more than two year’s history, the Time Range should be re-named FY17-FY18 and the start period advanced to FY17 (from FY16). Similarly, the plan Time Range should be renamed FY20-FY21 and advanced to FY20 (from FY19). FY18 is then available for the history to be populated and FY21 is available for plan data entry. Time Ranges Pitfalls Potential Data Loss Time Ranges can bring massive space and calculation savings to your model(s), but be careful. In our example above, changing the Start Period of FY16-FY17 to FY17 would result in the data for FY16 being deleted for all line items using FY16-FY17 as a Time Range. Before you implement a Time Range that is shorter or lies outside the current model calendar, and especially when implementing Time Ranges for the first time, ensure that the current data stored in the model is not needed. If in doubt, do some or all of the suggestions below: Export out the data to a file Copy the existing data on the line item(s) to other line items that are using the model calendar Back up the entire model Formula References The majority of the formula will update automatically when updating Time Ranges. However, if you have any hard-coded SELECT statements referencing years or months within the Time Range, you will have to amend or remove the formula before amending the Time Range. Hard-coded SELECT statements go against best practice for exactly this reason; they cause additional maintenance. We recommend replacing the SELECT with a LOOKUP formula from a Time Settings module. There are other examples where the formula may need to be removed/amended before the Time Range can be adjusted. See the Anapedia documentation for more details. When to use the Model Calendar This is a good question and one that we at Anaplan pondered during the development of the feature; Do Time Ranges make the model calendar redundant? Well, I think the answer is “no,” but as with so many constructs in Anaplan, the answer probably is, “it depends!” For me, a big advantage of using the model calendar is that it is dynamic for the current year and the +/- years on either side. Change the current year and the model updates automatically along with any filters and calculations you have set up to reference current year periods, historical periods, future periods, etc. (You are using a central time settings module, aren’t you??) Time ranges don’t have that dynamism, so any changes to the year will need to be made for each Time Range. So, our advice before implementing Time Ranges for the first time is to review each Module and: Assess the scope of the calculations Think about the reduction Time Ranges will give in terms of space and calculation savings, but compare that with annual maintenance. For example: If you have a two-year model, with one history year (FY17) and the current year (FY18), you could set up a Time Range spanning one year for FY17 and another one year Time Range for FY18 and use these for the respective data sets. However, this would mean each year both Time Ranges would need to be updated. We advocate building models logically, so it is likely that you will have groups of modules where Time Ranges will fall naturally. The majority of the modules should reflect the model calendar. Once Time Ranges are implemented, it may be that you can reduce the scope of the model calendar. If you have a potential Time Range that reflects either the current or future model calendar, leave the timescale as the default for those modules and line items; why make extra work? SELECT Statements As outlined above, we don’t advocate hard-coded time selects of the majority of time items because of the negative impact on maintenance (the exceptions being All Periods, YTD, YTG, and CurrentPeriod). When implementing Time Ranges for the first time, take the opportunity to review the line item formula with time selects. These formulae can be replaced with lookups using a Time Settings module. Application Lifecycle Management (ALM) Considerations As with the majority of the Time settings, Time Ranges are treated as structural data. If you are using ALM, all of the changes must be made in the Development model and synchronized to Production. This gives increased importance to refer to the pitfalls noted above to ensure data is not inadvertently deleted. Best of luck! Refer to the Anapedia documentation for more detail. Please ask if you have any further questions and let us and your fellow Anaplanners know of the impact Time Ranges have had on your model(s).

Dynamic Cell Access (DCA) controls the access levels for line items within modules. It is simple to implement and provides modelers with a flexible way of controlling user inputs. Here are a few tips and tricks to help you implement DCA effectively. Access control modules Any line item can be controlled by any other applicable Boolean line item. To avoid confusion over which line item(s) to use, it is recommended that you add a separate functional area and create specific modules to hold the driver line items. These modules should be named appropriately (e.g. Access – Customers > Products, or Access – Time etc.). The advantage of this approach is the access driver can be used for multiple line items or modules, and the calculation logic is in one place. In most cases, you will probably want read and write access. Therefore, within each module it is recommended that you add two line items (Write? and Read?). If the logic is being set for Write?, then set the formulas for the Read? line item to NOT WRITE? (or vice-versa). It may be necessary to add multiple line items to use for different target line items, but start with this a default. Start simple You may not need to create a module that mirrors the dimensionality of the line item you wish to control. For example, if you have a line item dimensioned by customer, product, and time, and you wish to make actual months read-only, you can use an access module just dimensioned by time. Think about what dimension the control needs to apply to and create an access module accordingly. What settings do I need? There are three different states of access that can be applied: READ, WRITE, and INVISIBLE or hidden. There are two blueprint controls (read control and write control) and there are two states for a driver (TRUE or FALSE). The combination of these determines which state is applied to the line item. The following table illustrates the options: Only the read access driver is set: Read Access Driver Driver Status True False Target Line Item READ INVISIBLE Only the write access driver is set: Write Access Driver Driver Status True False Target Line Item WRITE INVISIBLE Both read access and write access drivers are set: Read Access Driver Write Access Driver Driver Status True False True False Target Line Item READ INVISIBLE WRITE Revert to Read* *When both access drivers are set, the write access driver takes precedence with write access granted if the status of the write access driver is true. If the status of the write access driver is false, the cell access is then taken from the read access driver status. The settings can also be expressed in the following table: WRITE ACCESS DRIVER TRUE FALSE NOT SET READ ACCESS DRIVER TRUE Write Read Read FALSE Write Invisible Invisible NOT SET Write Invisible Write Note: If you want to have read and write access, it is necessary to set both access drivers within the module blueprint. Totals Think about how you want the totals to appear. When you create a Boolean line item, the default summary option is NONE. This means that if you used this access driver line item, any totals within the target would be invisible. In most cases, you will probably want the totals to be read-only, so setting the access driver line item summary to ANY will provide this setting. If you are using the Invisible setting to “hide” certain items and you do not want the end user to compute hidden values, then it is best to use the ALL setting for the access driver line item. This means that only if all values in the list are visible then the totals show; otherwise, the totals are hidden from view.