The benefits of accessibility are largely PR-related: nobody wants to be the company that makes blind peoples' lives more difficult. There may be ADA-compliance issues as well when you get big enough.
That said, accessibility is something you do when you're a big company. When you're a startup, work on making it useful to the easy customers, and then only once you have something that lots of screen-reader people want to use is it worth investing time in making it usable.
Even if you can't directly monetize the golden rule, a lot of accessibility work is super-simple thanks to the toolmakers giving it forethought. You can get to 80% of it in an iOS app by just making sure you've got descriptive text for each UI element in the storyboard. The few blind customers we've gotten as a result (some of whom helped enable it) have been extremely enthusiastic, which made my day.
Accessibility like scalability is far easer to bake in at the start than patch on at the end. However, punting until you start to gain traction is not a terrible idea.
I'm not sure about the US but in the UK what you're suggesting would I think be illegal. You shouldn't make things accessible just when you're a large company, that's a really disappointing approach.
This is just another reason why the US has a healthier startup culture. There are a lot of things that take effort to do that US companies can get away with not doing until forced to by law or the market, or when the opportunity costs diminish that they do it out of the goodness of the individuals' involved hearts.
Probably. My argument for healthier would mostly be on the second-order effects of having a better economy that can generate new sources of wealth. With a step back, until momentum and inertia are all that remain of the US' current dominance in startups that grow into world-changing companies providing value to all (including disabled people who benefit from accessibility), I'm default-skeptical of tinkering with the ecosystem with a top-down legal approach, especially when the approach is something like importing laws from countries without such dominance. In a way this is just "well we've kicked ass this far even though we do [morally reprehensible thing]", which by itself isn't a great argument, but paired with noticing that social systems are complex it's enough to give me a lot of pause. Even something that seems bonehead simple like saying "no more doing [morally reprehensible thing]" can have unwanted second order effects that end up with the consequence of a state worse than the previous one that included [morally reprehensible thing]. Full analysis is required by the people most capable of seeing all the effects, one of the worst outcomes is some group kicking up a stir to get some mandate passed on good intentions without thorough analysis.
Nobody is going to sue the company they've never heard of. If they have heard of you, you're in a good position to start worrying about things like accessibility/scalability/security.
Scalability is something it seems reasonable to leave until you're more sure you actually need it. There's also no legal requirement around scaling, and you're not affecting a specific group of people.
Accessibility is a legal requirement in the UK, and while you're right it's unlikely to come back to bite you as a small company, that's also a pretty crappy attitude to take. I don't like the practice of deliberately breaking the law anyway, but deliberately breaking the law requiring equal access to people with disabilities is awful.
While you have a point I'd say the vast, vast majority of accessibility related things that you can do with mobile apps and web sites are so simple to do it's practically incompetence to not do that and that'll get you 80% of the way there, maybe even more depending on what you're building.
Accessibility benefits everyone. Everyone wants easier to read, easier to understand, websites. How does a harder to use, harder to understand website help a business?
That's not really the comparison for a startup, though. It's "Would you rather have an easy-to-use, easy-to-understand website that does nothing useful, or an easy-to-use, easy-to-understand website that's full of security holes, or a hard-to-use, hard-to-understand website that basically does what it's supposed to?" It's pointless to make a website that nobody wants to use highly usable.
There's an opportunity cost for everything. The most fundamental opportunity that you don't want to miss is building something that lots of people want; if you don't accomplish that (and it turns out to be really hard), it doesn't really matter how well you've built the thing that nobody wants.
"Would you rather have an easy-to-use, easy-to-understand website that does nothing useful, or an easy-to-use, easy-to-understand website that's full of security holes, or a hard-to-use, hard-to-understand website that basically does what it's supposed to?"
There are more options, but they all involve giving up something somewhere. Perhaps you have a secure, easy-to-use, useful site, but charge more money for it. Perhaps it means that people can't use it today but can in a year. Perhaps it means that you've hired gullible underpaid engineers on promises of future wealth and then work them to the bone, but then do you want to be the person who's underpaying your employees?
The point (of both my comment and the original article) is that there are trade-offs, and you get to choose which trade-off you're going to accept, but anything you do is likely to piss off someone, so choose who you piss off wisely. Do you want your top priority to be accessibility? It can be, and that could be a wise choice if you identify a task that really really sucks for disabled people. But if you're looking at that before you've even got the fundamentals of building a useful product down, you risk building something that does something nobody really wants to do but does it really well.
That said, accessibility is something you do when you're a big company. When you're a startup, work on making it useful to the easy customers, and then only once you have something that lots of screen-reader people want to use is it worth investing time in making it usable.