Hacker News new | ask | show | jobs
by hosteur 1205 days ago
I am a paid user of Kagi. I absolutely love the search. It is fantastic compared to Google which I have almost completely stopped using due to declining quality.

I think that while the pricing is probably quite reasonable when considering the costs involved in search, Kagi is up against a mentality that web search is free. Because that has been the way it has been for all people's memory. I know it has never really been "free" because our data has been exploited by Google, and the likes. However, most people don't think that far - as is evident by the fact that Google, et al grew to the behemoths that they have become.

I think they will probably need some way to ease people in to the idea of search being a paid service.

Now, I live in one of the richest countries in the world, and when I pitch Kagi Seach to my friends, family and co-workers, they balk at the idea of it being a paid service.

Some of these people are software engineers or IT professionals who rely heavily on search and who stands to benefit the most from the increased quality. And these are the same people who often complains about Google's declining search quality and SEO spam, etc. But still, Google is "Good Enough" and it is "free". So this is what Kagi is up against.

So I think they probably need some better free tier to get people to actually experience the difference in quality. I know that the free tier is likely a huge loss leader and there is some limit to how much Kagi can provide as free.

I have thought about this quite a lot since started using Kagi initially with a free tier account. My free tier account is limited to N searches per month. I run out of that number of searches in a typical workday. Problem is that I set my default search engine to Kagi on all my devices. This means that when I just type some single word just to get to a website, say, "reddit", "amazon", etc. That counts towards my limit. So that when I actually need a quality search result because I need to search for a solution to some rare error condition or similar, then I will have run out of free searches.

So one idea I have thought a bit about is: not all searches are equal. This goes both for the value of the search result for the user as well as the cost of executing the search query. Some common terms like one word queries such as "facebook", "amazon", "apple", etc. would likely be trivial to cache and thus not cost as complex queries that are also likely unique.

I think they should consider making non-unique, one word searches does not count towards the limit. And I would also strongly consider not having such queries require a login. The reason being that this would let many more people get started with Kagi.

And also, if I am to pay even a single cent for a search that is trivial, I would go out of my way to ensure that Kagi is not a default search engine on any devices and I would try my search on one of the other engines first. Even if the cost is trivial - but it is still there and it is something I am conscious about. And what is worse: I now have a variable cost I need to worry about...

Also, I think they really need to get companies on board for this. But most companies have very strict policies regarding recurring fees.

That is, until the mentality around search being free changes... And solving that will likely require investing a lot in giving free searches away. It's a marketing cost. And people really need to experience how the quality and experience of search can be better than Google. Because even though Google sucks most people have a hard time envisioning what an alternative would look like.