On another note: the most common prompt a webdesigner nowadays gets is "make it look like x". Finding a own visual identity is just not in these days (this does not only affect the web but actual real worlds design as well).
It's easier for a programmer without design chops to cobble together something using boostrap than actually design a site from scratch
Regarding eCommerce, you really don't want your visitors to do anything other than add products to the cart and checkout so a simple, universal design is preferable to something that's going to win awards for aesthetic appeal.
Websites need to collect and convert users, somehow. You can use styles that are known to work, or you can go off and create something new that might fail. Unless your company's product is user interfaces, the choice is obvious.
With the rise of design systems and pre-built UI kits, it's easier than ever for developers to skip designers and build something fast. This lack of customization of UI kits results in similar design elements across different projects.
The ones that are trying to use language to sound they are changing the world and you will infinitely satisfied if you buy, when you are getting I dunno an API key or some such tiny feature.
That is more for the MVP consumer side.
Otherwise for B2B it is large bullshit bingo wording where it is really really hard to figure out even what they do.
What is Linear?
Oh right: https://linear.app/
Our current website advertises a desktop packaging tool, and is modern looking with a slightly unusual visual design (https://hydraulic.software/). It's an adaptation of a SaaS landing page template appropriately called Basic, and I chose it somewhat arbitrarily because it had animated bubbles (https://preview.uideck.com/items/basic/) which fit with the whole water-power theme of the company name, plus I liked the organic wavy look. Later I decided the bubbles looked unprofessional and turned them off.
To be clear, the template is fine for what I paid for it (very little) and the support I got is fine too. I have no complaints with UIdeck. But we've grown out of it.
The big problem is that it's too hard to add content. The template is really just a hand crafted landing page, it's not based on any sort of component system like React. Extending the template to support non-landing pages was an exercise in CSS frustration. Modern product website designs are oriented around a stack of similar sized blocks arranged vertically. The Basic template does this too but blocks are all individually unique and none of them are generic, which makes it hard to extend a page to advertise the whole feature set. This means I'm always torn between linking to the landing page or linking direct to the developer docs, because the latter explains all the features whereas the first looks nicer, but that's a stupid issue to have. We should be able to easily add content to reflect new features without hitting visual design issues.
Second problem - changing anything without breaking layout is too hard. The background images are SVG so if the browser window resizes the wave image rescales too, and this can mean white text ends up getting pushed out of the way and becoming invisible. I've wasted way too much time on layout bugs that only appear at certain screen sizes on certain pages. The more standard type of generic SaaS website doesn't have this issue because they avoid organic shapes that can move around in complex ways. It's all straight lines and 90 degree gradients.
Final problem - adding content involves writing markdown and HTML manually. I use a little home grown template system to help but that's unjustifiable (it was an experiment that didn't pan out). For the new site I'm not quite sure whether to go for something that's a totally visual CMS like a hosted WordPress, or do a bit of RDD and use it as an excuse to brush up my JS framework skills. Whichever I pick it's going to be focused on finding a template that gets all the awkward web design details out of the way (in particular for mobile support), and then tweaking it to fit our needs.
Why not hire someone to do a totally bespoke design? Maybe when we're bigger but my guess is that it'll have the same general structure as other sites simply because otherwise we'd need to go back and ask for more bespoke design work every time we launched a new feature and wanted to advertise it. You can do that but it's expensive and not justifiable at the current size.