dug 
Left a comment on Joe’s govuk prototype kit post and LI said I should post blah blah. Haven’t got time right now, but I’d like to come back to this issue a bit later.
Basically, my generation of designers was taught that prototyping should be as fast and cheap as possible. That meant sketches, lo-fi etc.
When the proto kit appeared it meant we started designing interactions and doing both generative and test-based primary research using an extremely high-fidelity assets, which at first made no sense.
Today, most product designers live in Figmaland, which muddies the waters as well. Why put all that effort into making a prototype so realistic? Is there a way to assess the value of discovered insight versus the time investment in the assets themselves?
Don’t get me wrong, even though I love prototyping in html, I think Figma is awesome (and itegration with design systems, large teams sharing etc etc) but I’m not 100% sure we’ve settled this one.
Maybe when someone figures out how to get an AI tool to do exactly what you want we pivot to that, but for now I think this issue is still live:-)
<original commentt>
Another angle/twist: I first bumped into the kit in 2017 I think it was, when I was at HMRC. Teams were producing a kit version of our service and using it for testing.
My first reaction was: “this is terrible, because you’re wasting so much time with high-fidelity mockups, you’re missing a whole stack of valuable insight you could find with much quicker paper prototyping or sketching/comps etc.”
Buuuuut actually, while ‘best practice’ at the time, that wasn’t entirely true because you know, context…
What we needed to test was a whole series of representations and inputs of data. We wanted to learn if the user understood what was happening, and test their level of confidence that they had taken the right action and/or that the problem was being solved etc. It turns out, that in the context of a govuk-based service, the kit gave us that in spades.
Then, once I got to know it better, and realised it was tightly integrated with the design system, created what was in effect a ‘wireframe specification’ that could be easily developped without hiccups (we had some glitches with Frontend versions but that’s another story) and best of all, gave me a functional mobile app right out of the box I reconsidered my position;-)