I ran AuditEase on my own site, and it scored 61 percent. Six tickets. Real issues, on a site I built myself and thought I knew. There is nothing like an automated audit to remind you that the site you are proud of has rough edges you stopped seeing a long time ago.
AuditEase is my AI accessibility tool. The idea is simple: point it at a site, and it finds the accessibility problems, files them as tickets, and tells you what to fix first. Running it on austincodez.dev was the dogfood test. If it could not find real issues on a site I know inside out, it was not a useful tool. It found six.
What we cover
- What a 61 percent score actually means
- Why six tickets is a good result, not a bad one
- How I ordered the fixes
- What running your own tool on your own site teaches you
- What I am not claiming yet
What a 61 percent score actually means
A score is a starting point, not a verdict. Sixty-one percent means the site does a lot right and has real gaps. It is not a failing grade and it is not a pass. It is a measurement of where the site stands today, so the next measurement can show whether the fixes moved it.
The number matters because it is specific. Saying a site has accessibility problems is vague enough to ignore. Saying it scores 61 percent, with six tickets attached, gives you a list and a baseline. That is the whole point of the tool: turn a vague worry into a concrete list.
Why six tickets is a good result
Six tickets sounds like a lot until you think about what the alternative is. An audit that finds nothing on a real site is an audit that is not looking hard enough. Six concrete, fixable issues is a manageable list, and each one is a real improvement for a real user.
The tickets are the record. Each one describes a specific problem, where it is, and what fixing it looks like. That is more useful than a report full of general advice, because a ticket can be picked up, worked on, and closed. A vague recommendation cannot.
How I ordered the fixes
The first fixes are the ones a real user would hit first. That is the ordering rule, and it is the only one that matters. An issue on the homepage outranks an issue on a page nobody visits. An issue that blocks someone from doing something outranks an issue that is merely untidy.
After that, the order is by reach: fix the issues that appear on the most pages first, because one fix in a template can clear the same problem everywhere. Then work down the list until it is empty, and run the audit again to see the new score.
I am not going to list the six fixes here as if they are already done. They are tickets, and tickets get worked in order. The point of this post is the process, not a victory lap.
What running your own tool on your own site teaches you
Dogfooding is the fastest way to learn whether your tool is honest. When AuditEase flagged my own site, I had two choices: argue with the result, or fix the site. Arguing would have told me the tool was useless. Fixing told me it worked.
It also taught me where the tool needs to be clearer. A score without a path forward is just a number. The tickets are what make the number useful, and the ordering is what makes the tickets useful. The tool is only as good as the next action it gives you.
What I am not claiming yet
One audit does not make a site accessible, and a higher score does not mean the job is done. Automated checks catch real problems, but they do not catch everything. Real accessibility work includes manual testing, keyboard navigation, screen reader checks, and listening to the people who actually use assistive technology.
I am also not claiming deep expertise in the 2026 ADA compliance series. I have read the titles and summaries, not the full series, and I am not going to write detailed compliance claims I have not done the reading for. The audit is a starting point, and the learning continues from there.
Final thoughts
Run the audit on your own site before you run it on anyone else's. You will find things you did not expect, and you will be a better advocate for fixing them because you fixed your own first.
The short version: 61 percent, six tickets, and a list that gets shorter every time I work it. That is what an audit is for.
