Author: Josh Baker is the Head of Finance Operations and Transformation at PJT Partners, Master Anaplanner and Workday Financials – Pro. Currently leads accounts payable, accounts receivable, expense, and finance transformation.
Reporting is not actionable or intuitive
Most finance operations teams have the same problem, and it isn't a data problem. Workday knows the status of critical documents: which invoices are unpaid, which expense reports are aging for days, or which supplier records were created without required data. The information exists. It just doesn't reach anyone in a form that provokes action.
We tried the usual answers first. Scheduled reports that arrive as attachments and worklets to visualize the critical data. Reports are static and worklets show a count but not a story. Dashboards require each team member to work in the same structure and are limited by several factors. Each one told us something true and left us doing the same thing we had always done exporting to a spreadsheet, adding columns, and emailing it around.
So, we stopped trying to make Workday do the monitoring and used Anaplan for that layer instead.
How it's wired
The mechanics are deliberately simple. A Report-as-a-Service (RaaS) report in Workday exposes the process data we care about. Azure Data Factory calls it on a daily schedule, handles authentication and any light reshaping, and pushes the result into Anaplan through the Integration API v2.0. No middleware to babysit, no manual export, and finance controls the source and the target data.
The point of describing it plainly is that the interesting part isn't the pipe. Almost anyone reading this could stand up the same pipeline in a sprint.
Filtering should follow the business question, not the report definition
This is the key differentiator. In Workday, a hub or a worklet is largely bound to how the report was defined. Want to see the same population sliced a different way, or compared to last week, or filtered by a threshold that varies by business unit? You are usually setting up a series of custom prompts, global objects, and not usually visible to end users.
In Anaplan, filtering logic lives in the model, and the model is owned by finance operations. We can hold multiple days of the same extract and show movement rather than a snapshot, so an aging bucket becomes a trend instead of a number. We can apply approval thresholds that differ by business unit and by rank without rebuilding the source report. We can flag exceptions using rules that combine data from more than one Workday process, which no single report was ever going to do on its own.
The dashboards naturally sync on each click to users can follow the natural flow of data and process to get to an answer and add context.
A link removes the excuse
The single change that moved adoption most was also the simplest. Every row in our monitoring views carries a hyperlink back to the exact object in Workday: the specific invoice, the specific expense report, the specific supplier record.
Before, finding what approval step or which approver an item was stuck at started a small scavenger hunt. They had to open Workday, search individual records, open in new tab, click through business process.
Now the item is one click away with its full context intact and a user can triage large populations with dynamic sync. Nobody has to trust the data, because they can click the relative link and verify at the source. That turned out to matter more than any formatting choice we made. If you want a monitoring tool to be used, make the distance between seeing a problem and acting on it as close to zero as you can.
Enrich what the ledger isn't allowed to hold
Every finance organization has knowledge that is useful but has no legitimate home in the general ledger.
Examples include: Why is this vendor an exception? Which of these coding errors are the same root cause? Who is the blocker on this follow-up vs who happens to be the approver of record?
Putting that into Workday would mean utilizing precious custom worktags and creating audit question we do not want to answer. Leaving it out means it lives in a spreadsheet or email and is not consistent or reportable.
Anaplan unlocks these capabilities, we layer commentary, custom tags, and internal categorizations on top of the Workday data without touching the ledger. The ledger stays clean and authoritative. The monitoring layer carries the institutional knowledge that makes the data actionable. When an item resolves, the tag history explains what happened, and the native audit trail proves useful for understanding when data points changed.
Advice to someone looking to launch this
Build the pipeline last. Start by writing down the three questions you want to answer every day then work backwards to the extract that supports them.
Resist the urge to move everything. A monitoring model that covers key processes well beats one that mirrors thirty and gets ignored.
Keep Workday as the system of record; the goal is not to replicate it. The goal is to make the exceptions impossible to miss and easier to remediate.
Move spreadsheets to the UX worksheet to improve collaboration, audit trail, advanced search.
Address exposed process problems that may have gone unnoticed, and tweak upstream and downstream workflows accordingly.
Some key metrics that improved from our deployment in accounts payable, was Avg Days Past Due from 45 to 5 days, Avg time to approve from 30 days to 15 Days, % of Invoices Sent back from 20% to 10%. Many of these patterns were not visible when we looked at static reporting because of the volume but with logic based, dynamic dashboards they became obvious.
Questions? Leave a comment!
