Power Apps or Translytical Task Flows?

I think I have gotten this question at least five or six times in the last few months, and with Translytical Task Flows reaching GA in the March 2026 Power BI update, I expect it to come up even more. So let me write it down once and for all.

The question usually sounds something like: "We want users to be able to add comments or update values in their Power BI report. Should we use Power Apps or this new Translytical Task Flows thing?"

My honest answer is: it depends 😆, but the decision is simpler than you might think.

A quick recap of both options

Power Apps embeds have been the standard answer for write-back in Power BI for years. You build a Canvas App, embed it via the Power Apps visual, and pass filter context from the report into the app. It works, it is flexible, and it can handle complex forms and logic. The cost is that you are maintaining a separate application, in a separate environment, with a separate deployment lifecycle, and you need Power Apps licenses on top of your Fabric capacity.

Translytical Task Flows are new and Fabric-native. Using Fabric User Data Functions, you can wire up report interactions (button clicks, input slicers, filter context) directly to write operations on a Fabric data source. The user stays in the report. The data gets written back. No separate app, no separate license, no context switching.

Where Translytical Task Flows shine

The scenario I keep coming back to is financial commentary. Think of a Finance department that reviews monthly results in a Power BI report. They want to add a written comment to a specific data point and have that comment stored together with its context: the reporting period, the sector, the department. So that next month, when they open the same report filtered to the same dimensions, the comment is right there.

This is a perfect Translytical Task Flows scenario. The filter context from the report passes directly into the User Data Function, the comment gets written to a Fabric SQL Database together with the dimension values, and the report reflects it immediately. Clean, simple, fully within Fabric.

Other scenarios where I think this is the right tool:

• Status updates on records directly in an operational report
• Discount or approval requests that trigger a Teams notification
• Flagging data quality issues from within the report without leaving the context

Where Power Apps still wins

I want to be clear: Power Apps is not going away, and for certain scenarios it remains the better choice.

The main one is UI complexity. Translytical Task Flows currently leans heavily on the input slicer for user input, which is a text field. If your write-back scenario involves multi-step forms, dropdown selections from a separate data source, conditional fields, or any kind of richer interaction model, Canvas Apps give you that flexibility and Translytical Task Flows do not, at least not yet.

The second one is non-Fabric targets. If you need to write back to Dataverse, SharePoint, an external API via a Power Platform connector, or any system outside of Fabric, Power Apps and Power Automate are still the right answer. Translytical Task Flows is built for Fabric-native write-back.

My current mental model is this: Translytical Task Flows for simple, Fabric-native write-back where the report context is the main input. Power Apps for anything that needs a richer UI or writes outside of Fabric.

A few things worth knowing before you start

PBIP and PBIR are fully supported. I initially assumed they were not, carried over from the preview days. I checked with the PM directly and the docs have been updated: full support is there at GA. If you were holding off because of ALM concerns, that blocker is gone.

Governance deserves a conversation upfront. User Data Functions live in Fabric, and someone needs to own them. Who writes them, who tests them, where do they sit in your deployment pipeline, and who gets called when something breaks? This is not a reason not to use the feature, but it is a conversation I would have with a client before the first function gets written, not after.

Power BI Embedded has limitations. Secure embed scenarios are supported, but check your specific scenario against the docs if you are building on top of Embedded.

Wrap-up

Translytical Task Flows is the write-back answer I have wanted to give clients for a long time. For the right scenarios it is leaner, more native, and frankly just less work than the Power Apps embed approach. But it is not a replacement for Power Apps across the board, and I think being honest about that boundary upfront saves a lot of pain later.

If you want to dig in:

• Translytical Task Flows overview
• Power BI March 2026 Feature Summary

Let me know: are you already using this with a client, or still evaluating?
And if you have hit the UI complexity ceiling with the input slicer, I am curious how you handled it.

Quick Tip: Get Ready for Changes in OneLake Operation Reporting

Check Your Capacity Metrics App

Just a quick heads-up for anyone using the Fabric Capacity Metrics app to monitor OneLake usage.

One of the upcoming features from the Fabric Roadmap I'm watching is OneLake Storage Lifecycle Management Policies. The idea is straightforward: rule-based policies that automatically move data between hot, cool, and cold storage tiers based on last access or modification time. So data you haven't touched in months stops costing you the same as data you query every day.

A recent message in the M365 Admin Center got me thinking (you need an Admin account to access this, or check out this Message Center Archive website instead):

Changes to OneLake operation reporting in Microsoft Fabric
Starting April 1, 2026, Microsoft is rolling out changes to how OneLake compute operations are reported in Fabric. This is tied to the (later) introduction of the mentioned OneLake storage tiers (hot, cool, and cold), and while billing rates are not changing, the way operations appear in your dashboards and reports is.

So what is changing?

The following changes are being applied:

• Operation names now include the storage tier. "OneLake Read via Proxy" becomes "OneLake Read (Hot)", for example.
• Proxy and Redirect operations are consolidated under a single operation name. No billing impact, but the names you're used to will disappear.
• Operations are now grouped by workspace in the Capacity Metrics app, under a new OneLake item. So plan for workspace-level reporting. If you rely on item-level operation details in the Capacity Metrics app, you'll have to use OneLake diagnostics going forward.

Units of measure and consumption rates stay the same, so no surprises on the bill.

A preview of how the reporting in the Metrics app will look like:

 

What should you do before April 1?

This change is rolling out in phases:

• Operation name changes start April 1
• Consolidation and Metrics app updates follow in the next weeks. Keep an eye on your tenant.

Steps you should take:

• Check if you have any custom reports, dashboards, or scripts that reference OneLake operation names, update those to match the new naming convention (think FUAM or others)
• If you rely on item-level operation detail in the Capacity Metrics app, set up OneLake diagnostics now so you don't lose visibility
• Share this with your helpdesk or anyone on your team who monitors capacity or cost analytics

You can find more details in the OneLake compute consumption documentation and track the rollout via the Microsoft 365 Message Center (MC1259829).

My Take On The Fabric Conference Updates

I am sure you have seen, there has been a lot of Fabric and Power BI news lately. Not surprisingly, Fabric Conference was also last week! 😀

I won't list all the updates here, you can read Arun's blog, or either of the Fabric or Power BI monthly feature summary blogs to go through the whole list:

• FabCon and SQLCon 2026: Unifying databases and Fabric on a single data platform
• Fabric March 2026 Feature Summary
• Power BI March 2026 Feature Summary

Instead, I will highlight what are the biggest game changers for me, and what you definitely should dive in deeper.

Platform

Database Hub (Early Access)

This is a new control plane for everything database: so Azure SQL, Cosmos DB, PostgreSQL, MySQL and Fabric Database, offering estate-wide observability, delegated governance and Copilot-powered insights. I haven't seen or used it yet, but does this really take away the need to use the Azure Portal and / or SSMS?

Planning in Fabric IQ (New)

A new enterprise planning capability that brings financial and business planning directly into Fabric. So we're talking budgets, forecasts, scenarios and targets. To me it sounds like this is aimed at replacing disconnected planning tools (like Anaplan)? It's a co-operation / acquisition with Lumen, but as far as I got from the sessions and community notes, you don't pay Lumen directly but instead it just costs you CU's from the already present capacity.

OneLake + Databricks Unity Catalog (Preview)

They now added the capability to natively read from OneLake through Azure Databricks Unity Catalog. As I see more and more companies use Databricks combined with Fabric, as either bronze/silver or the whole data platform, I reckon this integration will have advantages for people using both platforms. It still looks like Microsoft is keeping their promise of keep developing an open platform.

Fabric Remote MCP (Preview)

If you haven't looked into MCP servers yet, I highly recommend you do. I also haven't dived deep into it, but I started doing some dabbling around with the Power BI modelling MCP server, for minor tweaks and measure creation, and that was already awesome.

Kurt Buhler and Eugene Meidinger have some good posts about the theory, inner workings and how to get started.

Power BI 

Translytical Taskflows (GA)

Translytical taskflows (TT) a.k.a. writeback for Power BI, although that doesn't paint the whole picture because it can do much more. It uses User Data Functions (Python notebooks) in Fabric to update data in a data store of you choice (e.g. Azure SQL, Fabric SQL Database, Lakehouse) from within certain elements in a Power BI report. Now that TT is GA I reckon much more companies will be able / are allowed to explore this. This will be a game-changer, you don't need seperate licenses for e.g. Power Apps or other 3rd party software that can do writeback.

Be aware though, because there are still some limitations, a.o. PBIR (Power BI enhanced report) and PBIP (Power BI Project) formats cannot be used together with TT.

Direct Lake on OneLake (GA)

Direct Lake now works better with OneLake security and it has more modeling features. The guidance is now clear: use Direct Lake on OneLake by default, unless you have a specific SQL endpiont security dependency. For me, import is still the default, but if you have specific requirements around data freshness or loading very large datasets into memory, you could have benefit by using Direct Lake.

TMDL View in web modeling (Preview)

Finally, we have TMDL view in the web! That makes the web modeling so much more complete, you can copy measures, tables or whole models over. Or adjust sources, parameters, anything! They also implemented a side-by-side diff view before applying changes.

Custom Totals in Tables / Matrices (Preview)

I know a few people that will be very pleased with this... :-) There's been a lot of complaining over the years that "my totals are off!". No, that's just the way DAX works... Well, now you can at least decide for yourself if you want to change it.

Modern Visual Defaults (Preview)

I wanted to call out this one, because I'm not sure if I'm actually a fan of this feature. The new default will be smoothed lines, which can sometimes give false data points in a graph. I'd have to check wether this will turn out good.

Developer

Git integration now has selective branching, where you can branch out for a specific feature and only pull items you need.

Furthermore, two new open-source projects have been added:

• Agent skills for Fabric: these are natural language plugins for GitHub Copilot terminal. This sounds like plugging into the Matrix and learning a new skill :) It also means you can use and share skills from and for the community.

• Fabric Jumpstart: these are reference architectures and single-click sample deployments. These look familiar to the earlier industry solutions, but now completely set up with data inside your own tenant.

Closing

With everything taken into account, Microsoft is betting hard on Fabric as the AI data operating sysem. I'm curious what the coming weeks and months will bring.

As I am typing this blog, I'm now actually in the plane to the MVP Summit in Redmond, Washington at the Microsoft HQ! Hopefully I can dive even deeper into a lot of these topics while I'm there!

Power BI Gebruikersdagen 2026

Today, it's that time of year again! The Dutch Power BI Gebruikersdagen are here :-)

Last year's keynote, photo credits by PBIG

Today is Master class Thursday, all day workshops, tomorrow is Deep Dive Friday, with 90 minute deep dive sessions, and Saturday is Community Saturday

This conference is a must see for me and my colleagues at Powerdobs. Although last year I had to miss it, because it coincided with our annual family ski-ing trip. I was really sad I had to miss out.


There are workshops on Thursday, those cover a whole day deep dive on one topic.
Then there are deep dive sessions on Friday, 90 minute (mostly technical) sessions on one topic.
And last, we have community Saturday, with sessions of 60 minutes, with also a lot of less technical  and more business focused sessions. This year, there are 11 (!) simultaneous sessions, so it will be very hard to pick one session per slot!

You can find the agenda with the full list of sessions here.


I'll be heading to the conference tomorrow. This year, I will present a 90-minute session on Friday, where I'll deep dive into the Direct Lake (pun intended).

It will be fun to see so many familiar faces again. And it's always good to meet new ones too.

Come say hello or ask a question in, before or after my session on Friday. I'll also be present the whole day on Saturday!

Fabric Workspace Settings Update: "License Type" Renamed to "Workspace Type"

Just a quick post because I noticed a change in the Fabric UI, specifically in the Workspace settings.

I am working on a demo for my Power BI Gebruikersdagen session, and wanted to switch a workspace to Fabric capacity. I noticed that the setting License type has changed, and is now called Workspace type.

You can see the new layout in the screenshot below.

The new setting called Workspace type

As you can see in the screenshot, there are now 2 groups of workspaces (= licenses):

• Power BI only workspace types
• Fabric and Power BI workspace types

The 1st category is for a Pro or PPU user license or an embedded SKU.

The 2nd type is for capacity licenes only, with an F or FT (trial) SKU, or an old P-SKU

There's no change to the licensing itself (at least as far as I know), it is just a slight rewording of the already available types of user and capacity licenses. The documentation on Workspace types is already updated to reflect these changes.

I've checked 3 different tenants, and in all of these I saw the same new setting being displayed. All these tenants are in the West-Europe region, and because it might not be available in all regions worldwide your tenant may still be showing the old License type.

I like the new categories and more clear distinction between Power BI and Fabric workspace types.

Let me know: Do you like this change?

Transitioning to the New Power BI Enhanced Report Format (PBIR) - What You Need to Know

Let me be clear: I really like and support the updates that Rui Romano has been pushing the last years!
In short, it brings:

• Better support for CI / CD and source control 
• Better integration for programmatic report updates, e.g. with LLM's
• More reliable merge outcomes with the PBIP and PBIR structures

With that being said, I do think some customers do not want to have preview features in production, so they will be cautious with the recent developments. Since the end of January, the PBIR format will be the default if you don't take action.
If you don't want to enable PBIR yet, or just want to know more about the transition, read on.

A little bit of background

Let's start with the basics, what is PBIP, TMDL and PBIR?

PBIP
PBIP turns Power BI into a real project instead of a single magic PBIX file. Files on disk, folders, source control that actually works. If you’ve ever tried to diff a PBIX and felt pain: PBIP is the fix.

TMDL
TMDL is the semantic model as text. Tables, measures, relationships: readable, reviewable, automatable. This is Microsoft finally saying “models are code”, not artifacts you click together and hope for the best.

PBIR
PBIR does the same, but for reports. Visuals, layout and interactions stored in a format that Git can understand. PBIX still exists, but the report inside it stops being a black box. That’s why PBIR becoming the default is a big deal.

So, PBIP is the structure, TMDL is the model, PBIR is the report.
Together, they move Power BI away from file-based BI and closer to proper software development, whether you asked for it or not. 😊
And to be clear, PBIX is still here and not moving away either, I will expand on that later.

Why This Matters Now

PBIR will become the default report format starting January 25th, 2026 in the Power BI Service, with Desktop adopting it in March 2026, but it will still be in preview.

This will affect developers using PBIP workflows, admins managing enterprise tenants, and teams relying on legacy formats.

Because a lot of clients don't want to use preview features, you can opt out during preview: let me explain how and why you might want to...

M365 Message center PBIR announcement

Legacy Format vs. Enhanced Report Format

The Legacy format is the older JSON-based metadata structure inside reports, which contains 1 big, unreadable JSON file for all things reports in your Power BI Desktop file, whether it's PBIX or PBIP.
This legacy format poses limitations on readability, CI/CD, automation, source-control and working with AI / LLM's. 

As I mentioned earlier, and Microsoft is also explicit about that, PBIX isn't going away. It will still be the primary developer format.

What will change is the metadata inside a PBIX file: from PBIR-Legacy to the new PBIR structure.

Official Timeline

Changes will hit the service first, and Power BI Desktop later.

Power BI Service, starting January 25:

• New reports in the service will default to PBIR
• Existing reports will auto-convert to PBIR when edited and saved in the service

Power BI Desktop, starting with the March 2026 release:

• Desktop switches to PBIR as the default for both PBIX and PBIP-files. Until then, PBIR must be explicitly enabled in the preview features

General Availability (GA) & Beyond:

• PBIR remains in Preview during the transition and becomes mandatory at GA (right now: expected in Q3 of 2026, check the roadmap for latest info)
• At GA, PBIR-Legacy will be removed and PBIR will be the only supported report format

What You Need To Do Today For Opting Out

In Power BI Desktop:

• Disable the PBIR preview features (under Options > Preview features).

In Power BI Service (Tenant Setting):

• Turn off Automatically convert and store reports using Power BI enhanced metadata format (PBIR).
• Important: this tenant setting already existed for quite a few weeks but only took effect in the end of January 2026

PBIR tenant setting to opt out before GA

What about rollback?

If a report is already converted to the new format, you can still go back, but you have to do it with a backup of your report. Backups are automatically created for you, both in Desktop and in the service, but they will only be available for a limited time window.

Desktop

In Desktop, a backup is automatically written to disk and kept for 30 days. Depending on which version you have, the location differs a little bit:

• Microsoft Store version:
  %USERPROFILE%\Microsoft\Power BI Desktop Store App\TempSaves\Backups
• Executable installer version:
  %USERPROFILE%\AppData\Local\Microsoft\Power BI Desktop\TempSaves\Backups

Service

When a report is automatically converted in the service, a backup in legacy format is retained for 28 days. You can restore it from the workspace in the report settings.

For both the desktop and service reverted versions, it won't prompt again to upgrade for that specific file.

Wrap-Up & Call-to-Action

While the transition to PBIP and PBIR in general is a good thing, and I'm already using it for most of my projects, there might be some clients that rather do not want to, or just aren't allowed to use preview features in production.

If you are in the last group, I suggest you:

• Check the Power BI Desktop preview settings
• Check the Tenant settings in the Fabric Admin portal

Quick Tip: Analyze the Fabric Capacity Metrics App Easily

Just a quick tip for anyone using the Fabric Capacity Metrics App.

If you find yourself checking the Metrics app and see a spike in usage you might want to analyze that. How many times did you have to click to get exactly the column you needed? Or before you were able to click any column at all? 😁

You can of course filter on the dates, but sometimes that's not enough detail.

Let's say I want to analyze the spike around November 19th, but I can't click that one red bar that has the highest peak?

When you hover over the visual you can select Focus mode.

This brings up the visual on the entire screen, making it much easier to click on individual bars.

When you select the single bar you are looking for, and click Back to report on the top left you go back to the normal page view, but with that bar still selected.

You can then click Explore and go to the timepoint details.

Making sense in case of huge spikes

Sometimes you may hit a big spike in usage, either by high usage of the capacity, or because the capacity was paused.

In the above case I had a severe throttling issue, where the capacity would have been unavailable for multiple hours and the client decided to pause / unpause the capacity. This effectively takes all the future CU-usage and makes you pay it off at once, hence the enormous spike in usage on November 12th.

Note

Pausing a capacity will almost always increase your Azure bill, next to the Fabric capacities you already have.

The downside is that this immediately renders this visual useless, because now you can't see anything other than that 65000% spike anymore.. 😁

If you want to make this visual meaningful again, you can filter out the date with the very high %, in the filter pane on Date.

Change the selection to Basic filtering and select only the dates without the spike!