Join us at FabCon Atlanta from March 16 - 20, 2026, for the ultimate Fabric, Power BI, AI and SQL community-led event. Save $200 with code FABCOMM.
Register now!Learn from the best! Meet the four finalists headed to the FINALS of the Power BI Dataviz World Championships! Register now
Hi,
I’m currently building a report that uses Translytical Task Flows in Power BI, where I want to use the new Text Input Slicer as a comment field that the end user writes back to the data warehouse.
Here I’m running into a couple of issues. The first one is that when a Text Input Slicer is used as input in a translytical task flow, it becomes mandatory. This means the user must enter a comment in order to trigger the task flow. Ideally, I hope there will be an option in the future to make this input optional.
One workaround that could work is to set a single space as the default value and then treat that space as NULL in the UDF.
The challenge with this approach, however, is that I can’t fully reset the slicer without also removing that space.
To handle this, I’ve tried using a bookmark with selected slicers. This works perfectly in Power BI Desktop, but once the report is published to the Power BI Service, it stops working. The bookmark resets all other slicers correctly, but not the Text Input Slicer 😕
What other options do I have here?
Hi @mariussve1,
Thank you for the detailed update and for outlining the extra testing you performed. Since the bookmark feature now works as expected in both Power BI Desktop and the Service, but the Text Input Slicer value is still not being recognized by the Translytical Task Flow, this suggests there may be a Service-side issue with how parameter binding is handled.
Given that you rebuilt the report, recreated the slicer and task flow, republished, and confirmed bookmarks work while the task flow still fails to detect slicer input, it seems unlikely that report corruption or configuration is the cause. The slicer UI shows the entered text, but the task flow parameter is still empty in the Service, which disables the action button or passes a null value.
It would be helpful to confirm your Power BI Desktop version, capacity type, and workspace region. Testing in a new workspace or on a different capacity, if possible, could help identify if the issue is environment-specific. If the issue happens across environments, it may need to be escalated as a product bug, especially since bookmark handling was recently updated. The combination of Text Input Slicer and Translytical Task Flow should be supported, so this behavior does not seem intentional.
Please update the issue with your Desktop version, capacity type, workspace region, confirmation that bookmarks work, and whether the problem occurs in multiple browsers. This information will help the product team assess if the issue is broader or specific to certain environments.
Thanks again for your thorough validation and documentation. Your detailed input is very helpful for investigating these scenarios.
Thank you.
Hi @mariussve1,
we haven't heard back from you regarding our last response and wanted to check if your issue has been resolved.
Should you have any further questions, feel free to reach out.
Thank you for being a part of the Microsoft Fabric Community Forum!
Hi @v-sgandrathi,
I wrote a new post in this thread:
https://community.fabric.microsoft.com/t5/Service/Text-Input-Slicer-works-with-bookmarks-in-Power-BI...
Still some issues, because now bookmark works, but I cant get the Input Slicer to work with Translytical Taskflows at all.
Hi @mariussve1,
Thank you for your thorough follow-up and for confirming that the bookmark reset now works as expected in both Power BI Desktop and the Service.
Regarding the issue with the Translytical Task Flow not recognizing values entered in the Text Input Slicer, this isn’t expected and doesn’t seem to result from report configuration. Your validation steps rebuilding the report, recreating the slicer and task flow, and republishing, suggest a possible Service-side parameter binding inconsistency. The Text Input Slicer value is injected at execution rather than saved as part of the semantic model filter context. After publishing, the task flow may read a cached or empty state, even if text is visible in the slicer, which could explain why the action button stays disabled.
Since bookmark handling was recently fixed, this could indicate a regression affecting how the Text Input Slicer passes values to Translytical Task Flow parameters in the Service. You can try removing and rebinding the slicer parameter after publishing, doing a hard refresh, or triggering another interaction (like changing a slicer or visual) after typing to force state re-evaluation. It’s also useful to check if the issue varies by Desktop version, browser, region, or capacity type, and to test with a minimal repro model to see if it’s environment-specific.
If the issue continues, please update the existing issue with environment details such as region, capacity type, Desktop version, and confirmation that bookmarks now work but task flow input detection is inconsistent. This will help the product team investigate further.
Thank you.
Hi all,
I just wanted to follow up on this topic, as I have been in contact with Microsoft regarding the related Text Input Slicer and Translytical Task Flow issue.
https://community.fabric.microsoft.com/t5/Issues/Text-Input-Slicer-is-not-reset-by-bookmarks-in-Powe...
I was previously informed that the issue should now be resolved. After further testing, I can confirm that the bookmark behavior is now working correctly. The Text Input Slicer resets as expected when included in a bookmark configured with “Selected slicers,” both in Power BI Desktop and in the Service.
However, I am now experiencing a different issue.
The Translytical Task Flow does not recognize any value entered in the Text Input Slicer. Regardless of what I type, the task flow behaves as if the input is empty, and in some cases the action button remains disabled even though the slicer clearly contains text.
To rule out configuration problems, I have deleted the report entirely, rebuilt it from scratch, recreated the Text Input Slicer, recreated the Translytical Task Flow, and republished to the Service. The behavior remains unchanged. This suggests that the issue is not caused by report corruption or incorrect setup.
I have also tested using a single space as a default value, but the Translytical Task Flow still does not consistently detect it after publishing. The behavior appears inconsistent between Desktop and Service.
Since there have recently been changes in this area, I am curious whether others are seeing similar behavior. Is anyone successfully using a Text Input Slicer as a mandatory input for a Translytical Task Flow where the action button reliably activates based on user input?
Any feedback on region, capacity type, or Desktop version would also be helpful.
Any insights would be greatly appreciated.
Hi @mariussve1,
Kind reminder to check the proposed solution above. If it resolved your concern, marking it as accepted will help other users navigate similar issues more efficiently. We're here to support you.
Thank you.
Hi @mariussve1,
Here’s a short and clear response you can use in the Microsoft Fabric / Power BI community, reflecting your earlier points:
Thank you for your message. Currently, there is no publicly available roadmap confirming if the Text Input Slicer bookmark behavior will be made consistent between Power BI Desktop and the Power BI Service. This seems to be a limitation on the Service side, not related to report design, but it’s unclear if this will change or if it’s intended.
Also, there is no official support for making Text Input Slicer inputs optional in Translytical Task Flows. The value remains mandatory, and there isn’t a supported way to reset or bypass it while keeping the task flow active. Using a space or NULL can work logically, but this isn’t fully supported, especially in the Service.
For updates or future plans, it’s best to submit feedback through official channels to ensure the product team is aware of your scenario.
Fabric Ideas - Microsoft Fabric Community
Thank you.
Hi @mariussve1,
Your observations are accurate, and the behavior you’re seeing aligns with how the Text Input Slicer currently functions. The value entered is inserted directly into the visual-level DAX query as a literal string (for example, within a SEARCH() condition), which is why table visuals respond to the typed input. However, this value does not become part of the semantic model’s filter context, so DAX measures using SELECTEDVALUE(), VALUES(), or similar cannot access it, nor can it be reused or modified before being sent to a Translytical Task Flow.
This also explains the bookmark inconsistency. Power BI Desktop restores the internal state of the Text Input Slicer when a bookmark is applied, but after publishing to the Power BI Service, the bookmark does not reset the Text Input Slicer, even though other slicers are reset as expected. This points to a limitation or bug on the Service side rather than a report design issue. Currently, there is no supported method to programmatically reset a Text Input Slicer, keep the Translytical Task Flow active, and avoid requiring the user to re-enter a value. Your approach of using a single-space default treated as NULL in the UDF is logically sound and provides a good user experience when it works, but it relies on functionality that is not fully supported in the Service.
Thank you.
Hi @v-sgandrathi,
Thank you for the clear explanation and for confirming the behavior. That helps a lot, even if the answer is not entirely what I was hoping for 🙂
Since the Text Input Slicer has been in preview for quite some time now, close to a year, I am trying to better understand what the expected path forward is.
Regarding the bookmark behavior specifically, do you know whether consistent bookmark support between Power BI Desktop and Power BI Service is on the roadmap? Bookmarks are a very standard and widely used feature, so having the Text Input Slicer state restored reliably in the Service would make a significant difference for real-world scenarios like this one. It would be helpful to understand whether this is considered a known limitation, a Service-side bug that is planned to be addressed, or simply not supported by design.
Separately, I am also curious about the Translytical Task Flow side of this. Is there any plan to allow Text Input Slicer inputs to be marked as optional when used as parameters for a task flow? Being able to configure the input as optional would, in practice, resolve the entire scenario described above and provide a much cleaner user experience in write-back solutions.
If you are not the right person to comment on roadmap or product direction, I would really appreciate being pointed in the right direction.
Thanks again for engaging on this. The feature has a lot of potential, especially in translytical scenarios.
One interesting thing I noticed while testing this further:
If I place a table visual on the page using the same column as the Text Input Slicer, and then type a value into the slicer, that value does show up in the DAX query generated for the visual. For example, when typing Free text testing, the query contains a filter like:
// DAX Query
DEFINE
VAR __FilterFromTextInput =
FILTER(
KEEPFILTERS(VALUES(Comments[Comment])),
SEARCH("Free text testing", Comments[Comment], 1, 0) >= 1
)
VAR __FilteredComments =
CALCULATETABLE(
DISTINCT(Comments[Comment]),
KEEPFILTERS(__FilterFromTextInput)
)
VAR __TopComments =
TOPN(
501,
__FilteredComments,
Comments[Comment],
1
)
EVALUATE
__TopComments
ORDER BY
Comments[Comment]
As you can see:
SEARCH("Free text testing", Comments[Comment], 1, 0) >= 1
So the text entered in the Text Input Slicer is clearly being injected into the visual query as a literal and used as part of the filter logic. This explains why the table visual updates correctly based on the typed input.
However, what I am not seeing is any way for that typed value itself to be exposed to the DAX engine in a reusable way. The value appears to live only inside the visual-level query, and not as something that can be accessed by a measure via SELECTEDVALUE(), VALUES(), or similar patterns.
So while the Text Input Slicer clearly influences DAX queries, it seems to do so only as a query-time filter expression, rather than as a value that becomes part of the semantic model or filter context that measures can read.
Happy to hear if others interpret this differently, or if there is a known pattern to bridge this gap.
Hi @mariussve1,
Just thinking out loud here, could you use a measure that grabs the selected value from the text slicer, and if the text slicer is empty the measure handles adding the space? Then you can set the data function to pass that measure instead of the text slicer directly.
Proud to be a Super User! | |
Hi @tayloramy,
Thanks for the suggestion, that would actually be a very elegant approach if it were possible 🙂
I just want to double check one assumption though. As far as I can see, the value typed into a Text Input Slicer is not exposed to the DAX engine in a way that a measure can read it. Unlike standard slicers, the text input value does not seem to be retrievable via SELECTEDVALUE(), VALUES(), or similar DAX patterns, and instead appears to be usable only as a direct input, either as a filter on a column or as input to Translytical Task Flows and UDFs.
Given that, I am not seeing a way for a measure to detect an empty text slicer and inject a default value like a single space, nor a way to pass a measure instead of the slicer itself as the task flow input.
Am I missing something here, or have you seen a way to access the Text Input Slicer value from DAX that I might have overlooked?
Hi @mariussve1,
Darn. I was hoping that Text Slicers exposed their values like other slicers do. I was wrong.
There goes that idea.
I'm not sure I can see a nice solution for you. Bookmarks won't work because they don't bookmark the values inside the text slicer, DAX measures can't access that value so there's no DAX magic we can do.
This is very clunky, but instead of filtering out a space, can you change it to treat something like EMPTY as null, and then train your users to enter specifically that word when they don't want to enter data?
Proud to be a Super User! | |
Hi again @tayloramy ,
Thanks a lot for taking the time to think this through and for testing the idea. I really appreciate it. And yes, I am starting to run out of good options myself as well 🙂
I agree with you that training users to type a specific keyword like EMPTY technically could work, but from a UX perspective it feels quite brittle, especially in scenarios where this pattern would need to be repeated across multiple text inputs. At that point it becomes hard to explain and easy to get wrong.
At the moment, I am trying to push the bookmark approach a bit further, mainly because it actually behaves exactly as I would expect in Power BI Desktop. In Desktop, I can save a bookmark where the Text Input Slicer contains a single space, handle that space as NULL in the UDF, and then reliably reset the slicer back to that state by activating the bookmark.
The frustrating part is that this breaks once the report is published. In the Service, the same bookmark no longer resets the Text Input Slicer, even though it works perfectly for all other slicers. That makes me hopeful that this is simply a Service-side bug or a missing implementation detail rather than a fundamental design limitation. I have created an issue with Microsoft to report this behavior and hopefully get some traction on it.
For now, I will probably keep following the bookmark path, since it is by far the cleanest user experience when it works. Fingers crossed this gets fixed sooner rather than later.
Thanks again for engaging on this, really helpful discussion.
I tried reproducing your scenario, I was able to reproduce it. As you can see in the below image, I entered space as a value in text input slicer. Activated the bookmark, all the slicers got reset however text input slicer is not
After clicking bookmark button
Probably you can report this to Microsoft.
As a workaround, I would suggest you to use "Clear all Slicers" button I tested it and it is working fine. I am able to reset the value entered in text input slicer, As you can see the below images
After clicking on "Clear all slicers"
Hope this helps.
Read my blogs here: techietips.co.in
Connect on LinkedIn
|
Hi @tharunkumarRTK,
Thanks a lot for taking the time to reproduce this, really appreciate the detailed screenshots.
I did test the “Clear all slicers” approach as well, and it does reset the Text Input Slicer correctly. However, the challenge in my scenario is that when a Text Input Slicer is used as an input to a Translytical Task Flow, it becomes mandatory.
This means that once the slicer is fully cleared, the action button remains disabled until the user enters some text again. From a user flow perspective, this unfortunately puts us back in the same situation where the user must manually re enter a value before they can proceed.
The idea behind using a single space as a default value was mainly to keep the action button enabled while still treating that value as NULL in the UDF. Since bookmarks do not seem to reset the Text Input Slicer in the Service, that approach also breaks after publishing.
So unless I am missing something here, it looks like there is currently no way to both reset the Text Input Slicer programmatically and keep the Translytical Task Flow trigger available without requiring new user input.
Happy to hear if you see this differently, or if there is another workaround I might have overlooked.
Thanks again for checking this out!
Are you able to reset text slicer to empty space and enable the UDF button in the desktop? (you mentioned its not working in service)
Because I am able to reset value in text slicer to empty space but not able to enable the UDF though, not sure if I am missing something
before activating the bookmark
after activating the bookmark (if you observe udf is not enabled)
When I enter empty space manually then udf is getting activated
read my blogs here: techietips.co.in
Connect on LinkedIn
|
Share feedback directly with Fabric product managers, participate in targeted research studies and influence the Fabric roadmap.
Check out the February 2026 Power BI update to learn about new features.