I always feel really grotty about evangelising for products, but I switched to Kagi about six months ago and it really is a better experience. In almost all cases, the search results are as good as or better than Google, and I don't have to scroll through an increasing number of misleading ads to see them. I'm a happy customer.
When I first switched, I would often click the button to run the search on google for queries that weren't immediately giving me what I wanted (rather than go through the next few pages of results), but invariably I wouldn't find it there either. I think that's what gave me confidence that Kagi's results were at least as trustworthy as anything else. (to compare, I did the same thing in my multiple abortive attempts to switch to DDG and it always came up wanting).
> I always feel really grotty about evangelising for products
You shouldn't. Word-of-mouth should be the primary way people discover products.
In ye olden days, a region's best bakery or blacksmith didn't become well-known because they put up signposts everywhere, but because the quality of their craft made their name known far and wide.
I feel very comfortable recommending products that are actually good, ran by a UX-first company and reasonably priced.
Exactly. If you don't advertise what is good or bad through word-of-mouth and true reviews then the primary method of learning and evaluating productions is paid marketing. As you may suspect the opinion given by paid marketing is not reflective of product quality. This means that product quality has very little influence on market selection and we end up with tons of crap like we do now.
Information from trusted independent sources is the most useful tool we have to actually incentivize the market to actually create quality products that actually provide value to their users.
Word of mouth is great. In my opinion its the most trustworthy form of product review. However I've noticed more and more companies advertising via 'word of mouth' they call it guerilla marketing and in my opinion its disgusting and a violation of internet users trust. Reddit is absolutely filled to the brime with this kind of marketing.
> In ye olden days, a region's best bakery or blacksmith didn't become well-known because they put up signposts everywhere, but because the quality of their craft made their name known far and wide.
To be fair, advertising has always been a major thing, for example, The romans had a tonne of visual advertising[1]
> You shouldn't. Word-of-mouth should be the primary way people discover products.
I've been a satisfied customer of theirs since 2023.
That said, I've been burned by far too many companies - especially tech companies - who grew big, then proceeded to squeeze every drop of prior good-will out of their success to make a line go up and satisfy investors.
So my support goes as far as opportunistically recommending them for as long as they continue to be good. Which I still do, I use Kagi on every device and love their personal ranking system and translation services, and they've been a cornerstone of untangling my life from a Google login - speaking of being burned.
But going out of my way to evangelize them feels a bit icky, and I can't help but feel like there's another shoe waiting to drop. It kind of stinks to feel like that, because my hesitancy isn't even necessarily their fault.
I wasn’t really sure about paying for Kagi but I was convinced when I couldn’t find some meme video I saw only a year ago using Google, DDG, Bing, etc., but found it almost immediately using Kagi. I hadn’t realized how bad most search providers had gotten.
Pretty much the same boat as far as "a single experience sealed the deal".
Had been having this recurring weird behaviour with the Apple TV. Every time it happened I'd spend a little while trying to search for it and figure out what was going on. The results were consistently completely irrelevant and useless. This had been ongoing for months.
First time it happened after I started my Kagi trial I tried looking for an answer with Kagi. Second result nailed it. Understood what was going on, pressed one button and put it back.
Then knowing the answer I went back and tried the exact same search with Google. I did find the answer... somewhere at result 50+ well past a link to Google Books offering me an article in a 2008 copy of Men's Health Magazine.
Didn't even use up the rest of my free trial. Subscribed right there and haven't looked back.
I don't know that Kagi's always better, but I've yet to run into anything where it's substantially _worse_. I haven't had a single instance where I've exhausted Kagi and tried Google and found a result there instead.
I'm curious because I go through this experience a little more often than I'd like to admit, and typically end up frustrated and without any results (admittedly without using Kagi, yet). Did you just search for a phrase from the video, or what did you do to find what you needed with Kagi?
I've not gone looking for videos specifically, but my experience there is that Kagi seems to focus on what you've explicitly searched for, where Google and others have increasingly leaned into interpreting your intent.
Google's approach works well enough when you're searching for a commodity and you don't care terribly much about the specific source. I get the impression Google, especially post-LLM, wants to divorce satisfying your question from the underlying sources.
I find Kagi is better at finding a specific thing, especially if you're willing to engage with it as a tool, ye olde search engine style. If my query doesn't find what I want, it's usually apparent why and I can reframe it.
Yes and no. Because of aggressive action from IP holders a lot of these sites went underground and deliberately aren't indexed in the US and EU but providers from Russia or Switzerland got shadowbanned.
> When I first switched, I would often click the button to run the search on google for queries that weren't immediately giving me what I wanted
Same, I found it took a while to adjust my searching too. Kagi is much more sensitive to spelling things wrong. Google gets around that by only using the search query as an inspiration but that also introduces a lot of fuzziness in the result IMO. With Kagi, you get as much out of it as you put in is what it feels like to me. It's slightly harder to find things sometimes, sure, but at least we're using a product instead of being the product and that adds enough value for me for this to be the better deal overall.
The increased fuzzy interpretation is Google's greatest downfall. It takes away any ability to use it as a power user or for super-specific stuff. No Google, you don't know better than me. I need something really specific and you're "smoothing" the results to what some average random searcher might want!!
I am the point in my software engineering career where I simply don't need those dumbed down results. I need some niche research paper or the one guy's extremely in-depth benchmarking blog I found months ago but forgot to bookmark.
It got to the point where Google simply could not help me in my day job so I see the monthly cost as an essential expense similar to my JetBrains sub.
I'm no Google apologist (I use ecosia personally), but did you try using searching in quotes? That should force the search to only find specifically your query directly as spelled. Just curious if you did try it and there was still that "fuzziness."
I switched as well and I actually use the AI assistant since then primarily. It’s awesome to connect search directly with AI, almost always get what I want immediately.
I'm curious, how is the AI assistant experience different from Perplexity or even ChatGPT's search feature? Is it just the convenience of having several models there or are the outputs inherently better because the results are from Kagi's engine instead of google?
To me, the killer feature has been the ability to filter out sites from my search results. I removed all of Pinterest, several tabloids and conspiracy sites, any obvious AI-generated sites that I run into, and just with that my search result quality has increased drastically.
It's a feature that I'd like other search engines to adopt natively.
I had a similar experience. When using DDG my results were never very good and I’d always end up using !g to throw the search to Google. With Kagi, when I checked other engines it would come up empty as well. On more than one occasion I was on an outage call at work where there were many people using Google to find an answer (for hours), and I joined and did a quick search in Kagi and found the answer.
These days it's often the sites in the results themselves that are either ads or highly SEO-optimized low value sites. You can manage a blacklist and a list of sites to promote/demote in the search results. And manage different "lenses" to have different buckets of blacklist/promote/demote settings tailored for what you're researching. And can also be used for their Kagi Assistant when you allow it to perform web searches.
edit: This is detailed in the article, but leaving this here for those like me who first jump to the HN comments
IMO this is really Kagi's killer feature. If I see a poor behaving site (Pinterest, for example), I can easily exclude the domain from every search I ever make. I don't have to carefully craft an enormous search query of all the sites I don't want to be shown. Demoting news sites that have a reputation for promoting bad takes, and prioritizing results from sites that are known to be good.
The quick switch to move to reddit based search, or old web results, are also great, but for me, it's the tailoring of my results to what I actually want, and more importantly, what I don't want is what sold me.
FWIW, it's much better to downrank a page than to outright block it. Blocking is for a page you never, ever want to see.
I personally dislike Pinterest and TikTok, but sometimes it might be the only source of an image or video. Blocking means you don't get to see that result.
Most of the time I agree with you, and prefer downranking, but I have to take issue with your example of Pinterest, because my experience with Pinterest is that if I'm searching for a specific image, I will always get a thumbnail of exactly what I want from Pinterest, but when I click on it, I'm taken to a page of completely unrelated results. It is beyond frustrating, and specifically blocking Pinterest was one of the primary features that sold me on Kagi.
I know I'm probably speaking to the choir but reports like this always make me sad. The internet and the webpages that filled it used to be so cool. Now its just like three websites that are only really judged by how useful they are to us.
> In almost all cases, the search results are as good as or better than Google
Could you (or someone) share some specific search terms that you feel are better than Google? I've tried Kagi a few times and felt no significant difference in the results.
Try anything related to visas.
"us esta", "canada eta".
On Kagi, the official government site is always the first result. On Google, it's buried beneath an avalanche of scammy lookalike services that pay for ads and SEO.
You probably have an adblocker running. I also did this test with ‘Canada eta’ and got the expected result on top. But when I reloaded the page without my content blockers, the official Canadian website was more than 1.5 iPad screen scrolls down.
Yeah, of course. Anyone with a brain does. But "Kagi doesn't have ads" is a different statement than "Kagi's results are better than Google's." I've never seen anyone actually give a search term that Kagi does better at.
Isn’t a lack of adverts a prime reasons ‘Kagi’s results are better than Google’s’?
I hear the odd person say that they like the adverts but it’s hard to believe that’s a plus for the bulk of users.
I switched last year and haven't looked back as well. Only thing I miss for convenience sake is the Shopping tab but obviously the privacy concerns arent worth that convenience.
When I first switched, I would often click the button to run the search on google for queries that weren't immediately giving me what I wanted (rather than go through the next few pages of results), but invariably I wouldn't find it there either. I think that's what gave me confidence that Kagi's results were at least as trustworthy as anything else. (to compare, I did the same thing in my multiple abortive attempts to switch to DDG and it always came up wanting).