I want to put an idea in front of you and find out who wants in.
Not as employees. I am not anybody's boss here. What I have in mind is a group of independent people doing the same work with the same tools, and what we call it later we can decide together.
I need those people now, while the thing is still soft and can be bent. If you are the sort who likes to try something early and push on it, this is the part where you would matter most.
Here is where it comes from. I have been building Legato, a Mac browser that repairs pages for VoiceOver. Some of you have already used it. To build the repair layer I had to go through a lot of real sites, and the same handful of problems come back over and over. Buttons with no name. Fields with no label. An error shown only as a red border. Focus that jumps somewhere you cannot follow. Heading levels that skip.
Now the boring part, which is actually the good part. Since the twenty eighth of June last year the European Accessibility Act can be enforced, and it covers online shops, banking, transport and telecoms. The first court cases in France started in November. In the United States more than five thousand digital accessibility lawsuits were filed in 2025, and roughly seven out of ten were against online shops. The money exists now. Companies have a budget line for this and most of them have no idea what to spend it on.
What they buy today is an automated scan. A scan finds contrast problems. Contrast is the one thing that does nothing for us. Nobody sits down, runs the checkout with a screen reader, and tells them the pay button has no name.
If you want to see what I mean, there is a free checker I put up. It reads any page you point it at and tells you what it found. Run your own site through it, or run the site of a company you might want to approach. It all happens in your own browser, nothing is stored anywhere, and there is no account.
Now notice what it will not tell you. It reads the HTML as it arrives, so it can talk about labels, headings and structure, and it is blind to colour and to anything a page paints after it loads. That gap is the whole argument. The machine hands you a list. Somebody still has to walk through the checkout and find out that the pay button says nothing.
That is the part only we can do, and it is worth money.
There are two jobs in this, and I think they want to be two different people.
One person goes out and talks to companies. Finds who has the budget, explains in plain words what is broken on their site, gets the yes and handles the client afterwards.
The other one does the walkthrough. The real route with a screen reader, the whole checkout from the first click to the confirmation, writing down what actually stopped them. Then, after the developer has fixed it, that same person signs the work off.
You can do both. I would not recommend it. A proper walkthrough eats a day, and selling eats the days around it, so one person doing both usually does neither well. Two people working as a pair get much further. The fixing itself is a sighted developer's job, and that can be the client's own developer or one we bring in.
My part is the software. It is the same engine as Legato, turned around. It should take what you found and turn it into a work order a developer can act on, with a check on every item that goes red before the fix and green after. The recheck after the next release is where the steady money is, because sites break again every time somebody changes a theme.
And I want people around me while I am writing it. Not to click through it and say it is fine. To take it to a real site and tell me what is missing, because I am sighted and left alone I will get it wrong.
Two things before anyone asks. I will never sell an overlay. Legato repairs pages on your own machine and touches nobody else's site, and this would work the same way. And I am not going to send threatening letters to companies. The report goes to them for free, in a form their own developers can use.
If any of this sounds like something you would want to do, say so here or write to me. If it sounds naive, I would rather hear that now than in a year.
Comments
I would do the walkthroughs
I would happily do some walkthroughs. Consultancy is a job that many offer, to varying degrees of competence. I wouldn't be at all interested in the sales part. I am extremely preficient with a screenreader and will admit that I tend to just find, or make, a workaround if something doesn't work, even down to inaccessible sites. I know, That isn't how things get fixed, but when time is money it it is quicker for me than to write to a company, figure out who to actually speak to, get in front of them, pitch a problem with a solution, get them to agree, get them to find the money etc etc
If that part was done, and the goal was to review a website and prepare an accessibility report for a dev, then check the dev's work, I would gladly do this. You are welcome to reach out to me via my Applevis profile.
I would be apart of it, it's better than what I could do alone
I was trying to make a business from this concept.
but I only have experience in writing up reports.
I'm from the uk and wrote up a comprehensive report for a organisation I was volunteering for.
the site.
I understand what you mean.
Even an unlabeled button isn't all we're looking for.
Lots of sites have dropdown and list boxes that are very inaccessible with voice over.
others have text fields that just don't open the keyboard to type.
But without voice over they do.
I know because a family member had to do it for me a few times without voice over.
Sometimes in forms they give you visual indicators I believe to show where you haven't filled it out.
but If I'm clicking save and nothing is happening and I don't know what's wrong with what I filled in then How Can I fix it without a struggle trying to find the error?
sometimes it crosses into things that Sited people might even complain about.
Like the site just being downright messy anyway.
And the text fields should always have examples of the formatting and way it should be inputted.
like card numbers.
some want difffferent things inputed in different ways.
These things I doubt are detectable by a scanner,
these things Require an actual person to test and just use the site.
A scanner likely only helps complete a quarter of the job.
I reviewed and tested that organisatioon's website.
then I wrote up a fully detailed report about it.
It explained what worked well,
what didn't work well and Suggestions to improve.
As I said before it's a Thing I wanted to turn into a business.
but I didn't really want to settle for just writing them a report and then they might not actually do anything.
the report was supposed to be the end product.
and that's all I think I could do.
I couldn't help fix it for them because I'm not a Dev.
I had one masive problem
I don't know how to reach out to the right people and what to say to get this started.
Your approach sounds quite good.
because as a collective we can bring skills together and combine them into something that really works.