Business performance analytics sits between your ERP data and the decisions people actually make with it.
Every finance team I have worked with has the same quiet problem. The data is all there, somewhere, and getting a straight answer out of it takes three days, two exports and someone who knows which filter to untick. We are not short of numbers. We are short of numbers we trust on the morning we need them.
We use Financial reports but, this will not help if you require quantities and also Power BI reports but in this case, you will need someone technical most of the time.
That is the gap Business performance analytics — BPA, because nobody says the full name twice — is aimed at. It ships as part of Dynamics 365 Finance, it reads your finance and operations data through Dataverse, and it hands back ready-made reports in an embedded Power BI experience. No integration project, no separate warehouse to stand up first. I have spent enough time in my own environment to have opinions about where it shines and where it will surprise you, so here they are.
So what is BPA, really?
It is a solution that transforms your organisation’s data into actionable insights through integration with Dynamics 365, Dataverse and Power BI. It is three things at once: a set of curated dimensional data models over your ERP data, a library of prebuilt reports on top of those models, and a governed way of letting people slice them without handing out raw table access.
The part that matters to me is , anyone can build a Power BI report. BPA gives you a standardised model that travels with the product, which is a very different maintenance story from the bespoke datasets most of us have accumulated.
What you actually see when you open it
Illustrative view of a BPA-style finance workspace: KPI tiles across the top, trend and breakdown visuals underneath.
The first thing that struck me was how unremarkable it feels, and I mean that as a compliment. It looks like Power BI because it is Power BI, so nobody needs training on the navigation. They are genuinely usable on day one rather than being empty templates waiting for a consultant or developer.
In the old days on Dynamics, I also had these kind of reports via OLAP cubes, I am that long been in business, which was also easy but this is the most easiest reporting tool, I use.
Teach people that habit early and adoption follows on its own.
The fine print I wish I had read first
None of what follows is a reason to avoid BPA. It is a reason to set expectations with your stakeholders in the first meeting rather than the fourth, which is a much better meeting to have.
| What to check | What it means in practice |
| Refresh frequency | Two refreshes a day, at 12:00 AM and 12:00 PM UTC. Useful for management reporting, not for watching a number move during the day. |
| History available | Entitlement currently covers up to eight quarters of data. Multi-year trend analysis needs the data warehousing route. |
| Capacity | Volumes are constrained by the storage capacity of the Power BI Embedded A6 SKU. Large data sets are worth sizing before you commit. |
| Licensing | Included in the Dynamics 365 Finance licence and using shared Dataverse entitlements, so there is no separate analytics purchase to justify. |
How I would approach a rollout
If I were starting again tomorrow, this is the order I would do it in.
- Install it in a sandbox first and let it refresh for a few days before anyone forms an opinion. Empty-looking reports on day one are almost always a data question, not a product fault.
- Design security before access. Decide who sees which legal entities and which financial dimensions, and write it down as a role matrix rather than discovering it request by request.
- Pick two reports, not twenty. Spend view were enough to prove the case for me; a broad launch just gives people twenty things to ignore.
- Reconcile once, publicly. Take one number from BPA and one from the equivalent finance report, put them side by side in front of the team, and explain any difference. Trust is built in that meeting.
- Only then talk about extending. If you need more history or your own models, that is where the Fabric and data-as-a-service route earns its keep.
Where this is heading: analytics you can just ask
The development I am watching most closely is the ERP Analytics MCP server, currently in preview, which lets AI agents query the BPA models in natural language. The agent generates the query, row-level security is still enforced on the way through, and the person asking never sees a formula. It plugs into agent platforms including Copilot Studio.
I am deliberately not calling that a revolution yet, because previews have a way of humbling people who do. But it does change what a reporting backlog is for. If the routine questions can be answered by asking, the analytics team gets to spend its time on the questions nobody has thought to ask, which is where the value was hiding all along.
Useful links
- Start here for the official picture: What is Business performance analytics? (Microsoft Learn)
- Deployment steps and prerequisites: Install Business performance analytics
- Roles and dimension-level security: Set up security for Business performance analytics
- A walkthrough from the product group: TechTalk: Business performance analytics with Dynamics 365 Finance
- Ongoing community coverage: Business Performance Analytics series on the Dynamics 365 Community blog
- Wider product documentation: Dynamics 365 documentation
One last thing
BPA will not fix a reporting culture on its own. If the numbers are argued about today, they will be argued about faster tomorrow. What it does remove is the excuse that the data is too hard to get to, and once that excuse is gone, the conversation finally moves on to what the numbers actually mean. That is the part I enjoy.
Bir Cevap Yazın