This time we’re going bigger than ever. Fabric, Power BI, SQL, AI and more. We're covering it all. You won't want to miss it.
Learn moreDid you hear? There's a new SQL AI Developer certification (DP-800). Start preparing now and be one of the first to get certified. Register now
A few years ago, I had an experience that completely reshaped how I think about accessibility. I was working with a stakeholder who, for six months, kept asking me for “quick updates” on a dashboard I’d built. It was a classic RAG‑status report, everything red, amber and green. I remember feeling increasingly frustrated because the visuals seemed so clear to me, yet he always needed extra explanation. When he finally told me that he was colour blind and couldn’t perceive any of the colours I had used… it hit me harder than almost anything else in my 15+ year career. I realised how many times I had unintentionally excluded him. That moment stayed with me. It pushed me to learn about WCAG properly, and for the past two years I’ve been deep‑diving into digital accessibility, especially in the context of Power BI.
Whenever I speak about accessibility now, the same question always comes up: is there a centralised tool that can run accessibility checks for Power BI? For a long time, the answer was no. And I hated that answer, because I knew exactly what it felt like to get accessibility wrong without even realising it. That’s why I built PBIX A11y, a free, browser‑based accessibility checker designed specifically for Power BI reports. It’s built on my understanding of WCAG and powered by Lovable, which made it possible to turn that knowledge into something practical, scalable and genuinely helpful. In many ways, it feels like a small example of “AI for Good”: using the technology not to replace people, but to support them in building more inclusive experiences.
One of the first decisions I made was that the tool had to run entirely in the browser. Power BI reports often contain sensitive commercial data, internal naming conventions, or client information. I didn’t want anyone to hesitate because they were worried about uploading a .pbix file to a server. With PBIX A11y, your file never leaves your machine. You sign up only so your audit history can be saved, something you can revisit, compare, or delete at any time. The .pbix itself is never stored, only the summaries you choose to keep. You also have the option to audit your reports as a guest.
PBIX A11y now focuses on five core checks that consistently surface the issues that matter most. Colour contrast is the first: every text colour is compared against its background using the WCAG luminance formula, ensuring normal and large text meet the required ratios. Non‑text contrast comes next, checking bars, lines, KPI fills and other graphical elements that carry meaning. Titles and labelling are also reviewed, because missing page titles, visual titles or axis titles can make a report almost unusable for screen‑reader users. Finally, font scaling is checked using proportional minimums based on canvas size, something I researched extensively and published as guidance for the community.
Initially I tried to build a check on colour blindness safety within the audit, but I could not get quite there, so I decided to add a standalone tool where you can upload a screenshot, brand palette or chart and instantly see how it appears under seven types of colour‑vision impairments. It runs entirely in the browser and helps us rethink our colour choices.
The second tool I added is the Alt Text Generator, which is currently in testing. Alt text in Power BI is often overlooked or added as an afterthought, and I wanted to make it easier and more consistent. The generator identifies which visuals need alt text, ignores decorative shapes, understands the fields used in each visual, and even detects slicers on the page. It then asks a few clarifying questions about the purpose of the insight before generating both static WCAG‑aligned alt text and dynamic DAX alt text for conditional formatting. The dynamic version updates automatically with slicers and data refresh, making it far more meaningful for assistive technologies.
It’s important to me that PBIX A11y never becomes a replacement for human judgement. The goal is simply to catch the obvious issues earlier, so we can spend our time on the parts of accessibility that genuinely require thought, context and nuance.
Every issue the tool flags represents a real person who might struggle to read a label, distinguish two bars, or understand what a slicer filters. PBIX A11y exists to make Power BI more inclusive and to help prevent the kind of unintentional exclusion I once caused without even realising it.
If you’re curious, give it a go:
https://pbiaudits.com/
The application is currently being tested, so if you have any feedback or suggestions, please reach out, I’d really value it.
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.