Microsoft PerformancePoint Planning is dead. Long live PerformancePoint Planning.
I'm shocked! In 2026, coming out of FabCon2026, as part of Fabric IQ, we've now got Fabric Plan in preview.
https://learn.microsoft.com/en-us/fabric/iq/plan/overview
Writeback in preview
https://learn.microsoft.com/en-us/fabric/iq/plan/infobridge-how-to-write-back-data
Everything else in preview
https://learn.microsoft.com/en-us/fabric/fundamentals/whats-new?source=recommendations
And about 20 features that went GA in March, 2026 as part of FabCon announcements.
Medallion architecture and multi-company consolidations as a union-all statement in Fabric indeed!
Organizations rely heavily on data to run their operations, but key processes such as planning, reporting, and analytics are often managed with separate tools. This strategy leads to data duplication, increased manual effort, and a risk to data consistency. Planning is frequently done in spreadsheets/specialized SaaS, reporting in separate systems, and dashboards provide only a limited view of overall business performance.
There is still some documentation for PerformancePoint Services in Sharepoint to brush up on.
https://learn.microsoft.com/en-us/sharepoint/administration/performancepoint-services-overview
EOL Sharepoint on-prem, hello Sharepoint Online
Next month, April 2026, Sharepoint 2010 and 2013 workflows will stop working in Sharepoint Online.
In July 2026, Sharepoint 2016 and 2019 on-prem will be no longer supported.
SharePoint Server Subscription Edition and Sharepoint Online are the paths teams should have already adopted at this point.
Next I'll look for when Microsoft adopts PEL (PerformancePoint Expression Language) again.
PEL (PerformancePoint Expression Language)
PEL was an awkward yet powerful language for financial planning, consolidation and complex rules. Generally it was replaced by DAX, Python, MDX and various other components, though I still feel it along with hierarchy building and writeback from Excel in PerformancePoint Planning were nuggets of value that MS didn't transition properly.
PEL features that are not supported by MDX:
PEL supports parameters in rules. The values of the parameters can be supplied at run time.
PEL supports variables that represent model properties. Rules can use variables. The assigned value for a variable persists until it is explicitly changed.
PEL provides explicit support for computations that are related to consolidation, currency conversion, intercompany reconciliation, shares and ownership values, and exchange-rate accounting.
Perhaps in 2026 we'll see a unified language, a blend of MDX, DAX, BIML, KQL, SQL, Excel formulas, VBA, C#, Typescript, Markdown, Powershell and Python.
Actually we do have this, it's called Agentic AI.
Agents and AI Actions
In 2026, we have Agents and AI actions, and hundreds of preview features.
The Financial Reconciliation Agent
https://learn.microsoft.com/en-us/copilot/finance/reconcile/reconcile-data
https://learn.microsoft.com/en-us/copilot/finance/variance/create-ai-action
The Fabric Data Agent
https://learn.microsoft.com/en-us/fabric/data-science/concept-data-agent
Multimodal input with AI functions, pandas on pyspark with pyarrow
https://learn.microsoft.com/en-us/fabric/data-science/ai-functions/multimodal-overview?tabs=pandas
DAX parameters in Power BI Desktop and paginated reports
https://learn.microsoft.com/en-us/power-bi/transform-model/desktop-what-if
Visual calculations
https://learn.microsoft.com/en-us/power-bi/transform-model/desktop-visual-calculations-overview
DAX UDFs
https://learn.microsoft.com/en-us/power-bi/transform-model/desktop-user-defined-functions-overview
Foundry Tools in Fabric
https://learn.microsoft.com/en-us/fabric/data-science/ai-services/ai-services-overview
Semantic link
https://learn.microsoft.com/en-us/fabric/data-science/semantic-link-overview
And the list goes on...
In 2026, we're going to see a proliferation of features as agentic coding takes over.
Now I'm just waiting to see if MS relaunches a supported version of Silverlight Flash. On a Mac.
A reddit post from 20 days ago.
No comments:
Post a Comment