Author: Werner Priels is Product Manager at Anaplan, focusing on Financial Consolidation.
Mapping source data to the right targets is the backbone of a successful consolidation process.
While traditional many-to-one mapping gets the job done, complex real-world data at times used to require workarounds, manual manipulation, or elaborate staging logic. Well, say goodbye to those workarounds! We are thrilled to introduce not one, not two, but three major enhancements to our mapping functionality:
- Sign-based mapping
- Multi-dimensional mapping
- AI-assisted mapping
In this article we’ll dive into what these new features are, how they work, and how they will make your data integration smoother than ever.
1. Sign-based mapping: Automate routing by value
Have you ever needed to route data to entirely different target accounts based solely on whether the source value is positive or negative?
With sign-based mapping, you can define distinct targets for positive (+) and negative (-) source values directly in your mapping tables.
By enabling the "sign-based mapping" option, a new Source Sign column appears, allowing you to seamlessly split incoming data without any intermediate steps.
Imagine cash accounts that usually carry a positive balance and map to your assets. At year-end, if one of those accounts happens to have a negative balance, it needs to be reclassified to liabilities. With sign-based mapping, you just set up two lines for the same source accounts: one with a "+" mapping to assets, and one with a "-" mapping to accrued liabilities. The system handles the rest dynamically.
2. Multi-dimensional mapping: Precision for complex scenarios
Sometimes, a single source field just doesn't give you enough context to determine the correct target. Enter multi-dimensional mapping, a feature designed for your most complex data routing needs.
By flagging the "multi-dimensional mapping" option during your mapping table setup, you can link a staging table and select two or more source fields to define your target. Under the hood during the load process, the system seamlessly concatenates your selected fields to find the perfect match.
Best of all, it works exactly like the traditional mapping tables you are already used to. It reads combinations from top to bottom (stopping at the first match) and is fully compatible with classic functionalities like Wildcards (*), SKIP, and BLANK.
Let’s say you need to populate a "functional area" dimension in your target module, but your source system doesn't provide that data point directly. Instead, you use a combination of both "account" and "cost center" as your source fields to drive the “functional area”. You can now tell the system: If account is PL7000 and cost center is 'SND', map the target to functional area 'FA01', but if the account is PL6000 and cost center is ‘SND’, map target to functional area 'FA03', etc.
3. AI-assisted mapping: Work smarter, not harder
Setting up a brand-new mapping table from scratch can be a daunting, manual task. Why not let AI do the heavy lifting?
With AI-assisted mapping, our intelligent engine analyzes your source value and source caption to automatically propose the most logical target values. Just click the "AI-assisted mapping" toggle in your mapping table to let the engine go to work.
We’ve designed this for maximum efficiency and full user control:
- Auto-acceptance for high confidence: Any AI suggestion with a confidence rate higher than 85% is automatically flagged for your acceptance, saving you tons of clicks. (Don't worry, you can always unflag or overwrite them!)
- Bulk or 1-by-1: Review the suggestions and approve them individually, or accept them in bulk. The mappings are only applied to your data load after you review and hit save.
- Full traceability: We know auditability is crucial. Whenever an AI-suggested mapping is accepted and saved, a specific notification is permanently recorded in the audit trail logs.
Picture this, you just imported a list of fresh source accounts for an entity that just got acquired. Instead of manually mapping their entire chart of accounts one by one, you run the AI assistant. It immediately recognizes targets based on the source captions, finds the matching consolidated chart of accounts targets, scores it with a confidence rating, and automatically flags for your approval. A multi-hour job just turned into a couple of minutes review!
Ready to get started?
These features are designed to save you time, reduce errors, and make your architecture cleaner. Head over to the Configure > Maps section in your workspace to try them out today.
Have you tried any of these new mapping capabilities yet? Which one are you most excited about? Drop a comment below and let us know how you're using them in your models!
