Your accessibility scope is probably too big.
That does not mean your ambition is too high. It just means you've collected too many "while we're here" tasks.
Here's an example from recent memory.
The team I was working with started with a pretty clear outcome. They wanted to rework the search so that it works with a keyboard only. That meant, someone can go in activate the search, type something, select some filters, apply them and hit search. Results would show and they could tab to any of them and open them up.
Not bad as far as scope goes. It's a flow, end to end and it clearly helps someone.
Then the scope started to grow. They wanted to redesign the filters. Then they wanted advanced filters and tried to rewrite the error messages and empty states for that. Someone wanted to replace the old dropdown component that was causing issues. Then they got to the forms and things really blew up.
The project had become more noble. And all but impossible to get done in time. Just because they couldn't control scope.
Cut the scope!
How?
- Don't redesign the filters if the main problem is that the checkboxes are not labelled.
- Don't rebuild the search component if the issue is that focus disappears after results update.
- Don't rewrite all error messages if users only need one clear message when a required field is missed.
- Don't fix every colour contrast issue everywhere in the product if the search box has a clear search button that needs some tweaking.
- Don't move to a new design system if the current dropdown only needs focus trapped and a working close button.
- Don't make the table header rows sortable if users only need to order by date and open a row.
You might think this is lowering the bar, but it's not. I'm protecting the outcome by cutting things we don't need right now.
I still want to write down the bigger problems and put them somewhere visible. But it's okay to postpone them and give them owners later. This way, they're not swallowing the things users are desperate for.
So I have two questions for you:
- When was the last time you cut anything out of a feature?
- When was the last time you shipped a feature on time?