Kagi - Is it worth paying for a search engine?

I am currently seven month into my Kagi subscription, and it is time for a short review of this rather unusual service. If you have not heard of Kagi, it is a search engine with a subscription model. At Kagi they claim to put you first, since your subscription to the service effectively runs the company. There is no data selling, no data usage, no advertisements, no user-tracking. They only generate turnover with customer payments.

Free search is paid for with data

Many people on typical discussion boards (e.g. Reddit) reported that search result quality from Google or Bing declined in recent times, especially for power users (programmers and alike). I am not sure if I could feel this kind of decline as strongly as others. I could and still can find meaningful results on those search engines. Still, it is part of the truth that prevailing search engines rely on a business model in which monetization plays a crucial role towards search results, which requires them to add (or sneak in) commercial content that dilutes quality. Obviously, Googles or Bings result quality can not decline indefinitely. They have to maintain a fine balance between good-enough results and ads, to attract users as well as paying ad customers alike.

Unfortunately, Google and Microsoft have an additional leverage, that is being the default search engine on many OS and browsers. This probably would still secure them more than enough traffic from people who simply do not care, even if the quality of search results would decline a bit further. Now, honestly I think Google delivers a lot of high quality services. I have used many of them for quite some time and still use them to this day. And while they are free-of-charge (to some extent) they are certainly not free. We all know we are paying for them with data. This data collection and user-tracking industry has ultimately grown so big and is so intertwined with our current internet that it is actually a bit scary.

Why testing Kagi?

Personally, my major reason for trying Kagi is the fact that I was somehow growing sick of a SEO-dominated web with extensive user-tracking. I simply dislike the business model, when that is all there is left to discover content on the internet. The goal here was and is to somewhat free some part of my online experience from the data collecting machinery while simultaneously obtaining hopefully ad-, AI- and mostly SEO-trash-free search results. Kagis privacy claims, along with their waiver of ad revenue and user-tracking is a strong selling point in this context.

Kagis value proposition

I am quite aware that Kagi is using the Google search API, which means I am not necessarily escaping Google as a service completely; Similarly Kagi uses Yandex, Bing and some smaller self-made index for the “small web” to compile their results. All-in-all it seems like a reasonable approach to compile the best possible results. Obviously, Kagi is rather a metasearch engine with a meaningful filter and result compilation service than a complete search index-backed search engine by itself.

So what actually is Kagis value proposition from my perspective?

(Not final) verdict on using Kagi

But can they truly live up to these propositions and is it worth paying money for such service? - It depends.

My gut feeling tells me that search results are mostly on a par with Google, very similar in a professional (programming) context. However, sometimes Kagi delivers more original content, smaller blogs, personal, well-curated websites. This fact alone led me multiple times on journeys to aforementioned websites. These adventurous trips felt a lot like using the internet in the early 2000s. And I mean this in the most positive way.

Hence, one could say, Kagi is trying t live up to their promise of a humanized web by throwing interesting “personal” content in the mix. It is not that search engines like Google and Bing could not do this, but they are simply not incentivized to do so. Similarly, more user-centric and data-aware alternatives on the market like DuckDuckGo still operate on ad revenue. They might not collect as much data, but they are still an implementation of the classic search engine business model where users are not the customers. This indeed shows the advantage of Kagis business model that equates users and customers.

In addition, it is a charm to use the Kagi lenses to narrow down search results. Overall, that is a strong feature and differentiator on top of the value I see in Kagis service.

Anyhow, I am not yet sure whether all the aforementioned positive things are enough to justify my current subscription of 10$/month. On the other hand, with everything that Kagi does, it really feels like they put you in the spot and try to live up to their ideals of a humanized web. For now, I will keep on testing it, it is at least a good feeling to use a search engine that is strongly incentivized to get the best possible results to you without annoying you.

At last, it is worth mentioning that Kagi recently hit one of their milestones, that is to have 50k paying customers. These numbers certainly do not seem too impressive. In a market dominated by ad-driven, but free-to-use search engines it is still astonishing that just some years into Kagis journey and without backing by large investors, they have managed to attract such user base and actually reaching profitability. It seems like there really is something to Kagi.