[Discussion] Built-in Reporting for Playwright Workspaces – Feedback and Feature Exploration

[johnsta]

Summary
Playwright Workspaces in Azure App Testing does not currently include built-in reporting like the previous Microsoft Playwright Testing (MPT) offering. We chose not to make reporting a paid feature at this stage, since lower-cost alternatives already exist. That said, we’ve heard feedback from some users requesting a more integrated reporting experience.

This issue is intended to:

• Explain what’s currently possible
• Outline current limitations and potential roadmap items
• Collect feedback from users to help shape future investments

---

What’s Possible Today
You can publish Playwright’s native HTML reports using Azure Storage static website hosting. While this requires some manual setup, it offers:

• Full HTML reports per test run (including results, trace viewer, filters, screenshots, videos, and logs)
• Shareable URLs emitted automatically in CI logs
• Low-cost, scalable hosting under your own Azure subscription

Advantages over MPT Reporting:

• Lower cost
• Similar reporting capabilities
• Greater flexibility and control over retention

---

Known Gaps

• No built-in authentication or access controls
• No built-in consolidated view across test runs

---

On the Backlog
We are exploring potential improvements, including:

• Authentication and access control for reports
• Integration with the Azure Portal (e.g., surfacing report links in the Playwright Workspaces activity log)

---

Feedback Requested
We’d appreciate your input on the following:

• Is the current Azure Storage + HTML reports setup sufficient for your needs?
• How important are the backlog features (e.g., authentication, portal integration)?
• Are there other reporting requirements we haven’t addressed?

Your feedback will help us prioritize what’s most valuable. Feel free to share your thoughts in the comments.

[waltergalvao]

Is Microsoft Playwright Testing getting sunset?

[johnsta]

Playwright service capabilities for running highly-parallelized tests at scale have moved to Playwright Workspaces in Azure App Testing, which is now Generally Available. Azure App Testing unifies the experience for managing both load and functional tests in one place.

The standalone Microsoft Playwright Testing service will be retired on March 8, 2026, and folks can find more details on how to move in our migration guide.

[bojingo]

Centralized reporting in Microsoft Playwright Testing is a core reason my client has been using it. In fact, for our CI builds we run the Playwright tests on the GitHub runner (or even locally) and push results to Azure, even without using the cloud-hosted test runners (via setting useCloudHostedBrowsers to false). That way, whether we are running at scale or just running one-off runs, all the results are in the one place for everyone to view and collaborate around. Doing away with reporting in Azure App Testing and making it only about running the tests at scale would undermine a major motivation for using the service.

And besides, why would I have to roll my own reporting solution outside of the service that runs the tests? I can think of no example where there is a tool, service, or product that will run your tests but then you have to figure out how to publish and view the results from the test on your own. Even Azure Load Testing, which seemingly will also be under the Azure App Testing umbrella has results viewers built right into the Azure Portal. Sure, to deep-dive into results we have to download the result CSV and do some analysis on our own, but the reporting built into the tool is plenty sufficient to quickly assess results. It would seem silly to me not to have parity with some reporting between Azure Load Testing and Playwright Workspaces if they are both part of the same Azure App Testing platform. You already have a fine solution in Microsoft Playwright Testing - please find a way to port that capability over for your customers too.

[amayfoxen]

We've been using MPT and the build-in reporting has been one of the most useful features (we're also using the cloud hosted browsers). We need an access controlled solution for viewing reports, so having to build something out for that is more challenging than uploading to Azure Storage and making available as a website.

[krystianYoyo]

I’m working with .NET Playwright.

My company recently moved from Azure DevOps to GitHub, and I started using MPT mainly because of the reporting features it provides. However, it’s frustrating that there’s no straightforward way to view test results along with screenshots and traces directly in GitHub Action results.

For now, I’ve managed to get a link to MPT reports, which works, but it’s still inconvenient to have to log in every time I want to view them. That’s a limitation I could live with.

What makes things more difficult is that since you’re planning to turn that off, and there’s no HTML reporting available for .NET Playwright, I now need to look for a completely new solution again.

Does anyone know of a good solution for reporting test results with .NET / Playwright / NUnit in GitHub Actions?

I’m not just looking for a simple pass/fail summary. What I need is a way to view reports together with test attachments, without having to manually download them, find the right files, and then re-upload them to https://trace.playwright.dev.

I’ve already tried using GitHub Pages and Allure Reports, but when working with multiple branches and a larger number of traces, the generating reports are laging and become less reliable.

[puagarwa]

HI @krystianYoyo I can understand the lack of test reporting in github can be an issue.

Just a question, is this an option for you to keep using ado for CI pipeline while your code can still in github ? lot of customers follow this path and github has ado pipeline extension/actions already ?

[UserR00T]

I have to fully agree with what @bojingo said. We also started using the Microsoft Playwright Testing environment without cloud hosted runners as a way to manage our reports from a singular place. Previously we had about 500 Selenium tests that we wanted to modernize and make more robust (codebase expanded over the years to an unmaintainable size) so we started looking for alternatives.

Microsoft Playwright Testing basically had everything we were looking for; Entra ID for authentication, great integration with the .NET Playwright library, and all-in-all a pleasant interface to collaborate in and view reports. Unfortunately the primary usecase for us is the reporting feature, as we had the cloud runners fully disabled and done a local run. (currently no added benefit for running the tests concurrently/parallel due to the nature of the tests.)

The closest alternative that is suggested are the HTML files along with a static file host within azure storage. Ironically as @krystianYoyo mentioned this is not even easily accomplished within the .NET library of Playwright. Having our CI/agent produce these files as artifacts for example just adds extra complexity & isn't the security we envisioned initially as we previously had Entra ID authentication; and would go back to an unauthenticated, http, storage account. Sure, one could place Azure CDN or Azure Frontdoor to secure and add tls/https to the webpages, but at that point we're just reinventing the wheel we originally had with MPT. At that point we might as well go back to Selenium with our custom-made reporting functionality which we so desperately are trying to get rid of and upgrade to a more modern approach.

Like many others we would really wish for Microsoft to reconsider retiring the reporting functionality, and finding a way to integrate it in Microsoft Azure Playwright Workspaces. Unfortunately, without the reporting functionality we might have to look at services outside of Azure that can bring this functionality to our team.

[krystianYoyo]

@puagarwa yes we had this setup for a few months but the reason why I moved to github completly is our company wants to get rid of ADO

[Kranael]

We as a company waited for the release and now its there azure app testing cool... not so much because we realized that a already working functionalty (Reporting of Traces) a really good feature so all reports are shareable is stripped out? Really?

Why do you remove something that worked already and was 1 of 2 features of the whole service? So like 50% reduction of service? Why?

We want to pay for that reporting maybe we opt in for cloud usage but for now no reason to use azure app testing.

[apis3445]

Hi! then there isn't plans to include the html reporting soon? I can only see general test results but that isn't useful

[johnsta]

Thanks everyone for the feedback – really appreciate you taking the time to share your thoughts.

Just a quick update: we’re actively looking into adding reporting features to Playwright Workspaces. Keep the feedback coming – it’s super helpful, and we’ll post more updates here as things move forward.