You're forgetting that the "niche technical audience" has something marketers and influencers and other kinds of pushers don't:
competence and credibility. And (not relevant here, but to plenty other products) control or influence over procurement process in SMBs, corporations, universities and government facilities.
Nearly all of us in the "niche technical audience" are the personal IT support departments of some subset of our family, friends, and their friends. They come to us asking for advice, or to set their computers/phones up, or to fix them after they "caught viruses". However begrudgingly we do that, it puts us in a position of power - they listen to and trust in what we say, and accept uncritically what we do to their machines. And, our interests are mostly aligned with theirs - even if we don't care about particular friend-of-a-friend's happiness, we'll still do them good so they don't have to come back with more issues any time soon.
This is how Google Chrome spread. This is how Firefox still survives. This is how several brands of anti-malware software spread - a mistake that's now difficult to undo. This is how AdBlock Plus became a thing, and how uBlock Origin is now replacing it. All these trends and more, I participated in first-hand. People still remember and follow advice I gave them over a decade ago (which isn't always good - see the anti-malware stuff).
And so is the case with Kagi, to a degree. I'm paid user for 1.5 years now, happy with the service. I recommend it in relevant discussions, I mention it to people who spot it - but since it's a paid product, it is a tough sell with general population. Still, I try to spread the word, like I do with any other good and non-user-abusive tool.
That said, this news makes me somewhat reluctant to recommend Kagi. I'll keep using it because it provides me immediate value for reasonable price, but taking investment often is a Faustian bargain. In this case it's not as clear as with regular VC backing, so I guess we'll see where this goes.
