R and Python visuals retire in May 2027: how to find them and what to replace them with
Microsoft is retiring R and Python visuals in Power BI, including R-powered custom visuals, in May 2027. What stops working, what does not, and a practical plan to find and replace them.
Microsoft has announced that R and Python visuals in Power BI retire in May 2027. That includes R-powered custom visuals from AppSource. After retirement they stop working in published reports in the Power BI service, and that also covers reports published years ago.
There is no in-place migration. Every affected visual has to be rebuilt in a different way. Six months sounds like a lot, but if script visuals are spread across many workspaces, the inventory alone can take a while. So this is a good week to start.
What actually changes
The announcement on the Power BI Updates Blog and the Microsoft Learn page say the same thing, and the scope is clear:
- Power BI service: at retirement, all R and Python visuals stop working in published reports, including reports published before the date.
- R-powered custom visuals: custom visuals that use R also stop working.
- Future versions of Power BI Desktop: R and Python visuals are removed, so you cannot add them to reports.
- Older versions of Power BI Desktop: they keep rendering the visuals locally with the R or Python installation on your computer.
That last point is not a way out. A visual that only works on one analyst's laptop is not part of a report that other people use.
What is not affected
The retirement applies only to R and Python visuals in interactive Power BI reports. Python scripts that you use in Power Query to populate tables are not affected.
That distinction matters. If your team uses a Python step in Power Query to clean or enrich data, you do not need to change that because of this announcement. The scope is the visual on the report canvas.
It is also worth remembering that this is the second step. Microsoft already ended support for R and Python visuals in the "Embed for your customers" and Publish to web scenarios from May 2026, where they now show as blank. If you are an ISV, you have probably dealt with that already. Everyone else is now in scope.
Why this one needs more care than a normal visual swap
A script visual is often more than a chart. In my experience it is where someone put the analysis that did not fit anywhere else: a forecast, a clustering, a statistical test, or a chart type Power BI could not draw at the time.
That logic usually lives only in the script. It is not in the semantic model, it is not tested, and the person who wrote it may have moved on. So when the visual stops rendering, you do not just lose a picture. You lose the only place that calculation existed.
So for each visual, ask: which part of this is analysis, and which part is presentation? The answer decides where it goes next.
A practical plan to find and replace them
Here is how I would approach it.
- Build the inventory. List every report that uses an R visual, a Python visual or an R-powered custom visual. Note the owner, the audience and how often the report is used.
- Search source control if you can. If your reports are saved in Power BI Project (PBIP) format in Git-integrated workspaces, search the report definitions for the
visualTypevaluespythonVisual(Python) andscriptVisual(R). These values do not identify R-powered custom visuals, so review the custom visuals in each report separately. - Save the scripts now. Copy each script and the list of packages it depends on into source control. Microsoft recommends this too. You do not want to reverse-engineer a visual after it has stopped rendering.
- Classify each visual. Is it presentation (a chart type) or analysis (a calculation people rely on)?
- Pick a replacement path. Microsoft lists these options:
- A built-in Power BI visual. The built-in set covers more today than when many of these scripts were written.
- An existing custom visual from AppSource or your organisation.
- A new custom visual built with the Power BI custom visuals SDK. Microsoft notes that AI coding tools can help here.
- A Fabric notebook, if the analysis does not belong in a report. You might need to update scripts for supported languages, libraries and data access.
- A Fabric app (currently in preview), for visualisations you cannot recreate in a report. Both notebooks and Fabric apps can be added to an org app and shared with your audience.
- Move logic upstream. If a script calculates something people rely on, compute it in a notebook or in the model, store the result, and let the report display it with a normal visual.
- Tell people early. Let report owners and users know which reports are affected and when the replacement is planned.
What this means for you
If you are a Power BI admin or a BI lead, the first job is the inventory, not the rebuild. You cannot plan work you have not counted.
If you are a report developer, start with the visuals that are used the most and the ones that carry real analysis. A decorative word cloud can wait. A forecast that a sales director looks at every Monday cannot.
And set an internal deadline well before May 2027. Rebuilding takes longer than expected, testing takes longer still, and you want time to compare old and new results side by side while the old visual still works.
The bigger lesson
After two decades in data, I have seen many retirements like this. They rarely hurt because of the feature itself. They hurt because of what the feature was quietly carrying.
The good habit here goes beyond this deadline: put calculations where other people can see, test and reuse them. A visual should display the result, not be the only place it exists.
How many script visuals do you think you would find in your tenant if you searched today?
Sources
- Announcing the retirement of R and Python visuals in Power BI (Power BI Updates Blog, 9 November 2026)
- What are script visuals in Power BI? Retirement of R and Python visuals (Microsoft Learn)
- Previous monthly updates to Power BI: November 2025, deprecation in Embed for your customers (Microsoft Learn)
- Power BI Desktop projects (PBIP) (Microsoft Learn)
- Create a Fabric app (Microsoft Learn)
Enjoyed this? Get the next one by email
Occasional emails about Microsoft Fabric, SQL Server, Power BI and Synapse.