Many talk about the pricing, but for me the value of having a working (unlike DDG) search engine that doesn't secretly think it knows better than me (like Google) is worth everything I pay now.
Internet is I guess, next after a working computer with a working dev environment, the thing that makes most difference for many of us developers.
For me it has value in two different ways:
1. time saved by arriving at the correct answer quicker and not waste a lot of time trying to refine my search to the point where even Google cannot misunderstand it. If your relationship with Google is better, more power to you.
2. I am a happier man as a result of not having to fight my search engines day in and day out.
I also notice that unlike a number of others who like me feel they use search a lot, I don't get close to the quotas.
I guess one thing I might do differently (based on years of frustration with Google and DDG) is I take something like a depth-first approach based on my first search:
- Using Tree Style Tabs I open every result on the first page that looks relevant in a new tree (ctrl-click to open in new tab, nesting is automatic)
- I then look at each of them, ctrl-clicking every link that looks relevant, recursively
This works for me and I have been doing it for years. One nice side effect is it gives me a permanent somewhat organized tree of most of the most relevant docs. In some cases I even select such trees and (with the help of another extension) export them as nested links in markdown in rhe hope that I can use as a more permant table of contents for the topic.
Yeah. This is a very good point. Overall I think the user experience of the web when accessed through google kinda sucks in an overtly annoying way.
I also think it's one of the big selling points of GPT-like search engines. It's not that they necessarily arrive at the answer quicker, they're just so much less annoying to use than Google, not only because you don't have to wrestle with an ambiguous syntax, but you also don't need to wade through bait-and-switch spam results, dismiss newsletter requests, get roped around by filler text, etc. The typewritered text is also a brilliant move to circumvent the sort of skimming-like reading pattern most of us have developed due to exposure to the spammier side of the web.
> but you also don't need to wade through bait-and-switch spam results, dismiss newsletter requests, get roped around by filler text
All that content was created with existing SEO rules in mind. As soon as GPT-likes have established behaviour, we'll get new kind of content that specifically targets them. But at least we can enjoy better results in the meantime.
It's tricky to extrapolate from Google. It's fairly obvious that there are things Google could do to fix this problem such as to penalize ads. They are never going to do as long as their main income is those very ads they would need to penalize to address the search engine spam problem.
In general, a big part of why Google is struggling so much with search engine spam is that they are such a dominant force in directing web traffic that they're effectively re-shaping the web around them to optimize for their algorithm. If you had any sort of diversity in search, this would be much less of a problem because the economic incentives would be much smaller.
I don't think it can be taken for granted that any search ecosystem would struggle in the same way Google has been struggling. I think their SEO-problems is a direct consequence of their monopoly on search and a lack of real competition.
Have you tried asking ChatGPT? That’s where all my technical questions start these days.
In the past few days it’s helped me with a bunch of k8s stuff (“explain this coredns log message”, “give me the kubectl command to show all events sorted by most recent first”, etc…)
I'm trying it a bit and it looks pretty good. A bit worse at re-using context but promising. Why does kagi gives that away for free, but asks a fee for search ? I would argue chat is more expansive operationnally ?
This a Kagi experiment (hence on labs.kagi.com subdomain). We run many experiments to learn about AI and search and we will be soon implementing learnings from FastGPT into Kagi.
I've been using Kagi as my default search engine for four months for both work and personal use, and it's been a great experience.
The only time I find myself re-doing a search in Google is when I want to look something up in Google Maps, but everything else, Kagi is on par with Google or better. I really like that Kagi lets me filter out blogspam results and sites that are just remixing content from StackOverflow.
I have been using Kagi with good results too, but some things it completely shits the bed with. I wanted to find the AWS portal, so I searched for "AWS". First result, American Welding Society. Fine. Second is a medium article about AWS. The third is a reddit post about AWS. On down and down, the AWS website doesn't show up on the first page. So I searched "AWS Portal". Again, the AWS website does not show up. This is still happening for me. Just checked, I haven't blocked their domain. It's honestly such an outlier, I wonder if they intentionally blocked results from the AWS domain from showing up.
I get the same (relevant) results for the query of "aws" as freediver21.
I've never seen Kagi miss something blatantly obvious like that, so I echo sibling's suggestion to reach out to Kagi to investigate what's going wrong in your account.
Perhaps the changes made in the announcement we're all commenting on have improved the situation. I had a few search terms (intentionally niche) that are now showing up as top results where as they didn't in the past. Very happy with the changes.
Now it's just the price holding me back. I just refuse buy metered services. I've had unlimited bandwidth for years despite my usage consistently falling below the cheaper top limited usage download tier. I just don't want to think about usage — it's an unwanted distraction when something is critical to my workflow.
However, $25/mo is just too much for this service. I can see why they need to charge that at the moment due to economies of scale (or lack there-of). However, even $10/mo would be pushing it for this service.
$25/mo is a long way off the mark. Can buy a VPN, support an open source ad blocker, and have money spare to pay for a couple of streaming subscriptions each month.
Oh wow, these are legitimately all good improvements. Some are things I'd use, some are, I'm sure, going to appreciated by others, but what's pointedly not anywhere on this list (except cheekily in a few screenshots) is any sort of AI-powered nonsense. I know they mentioned a while back that they were working on some tooling of that sort, but boy am I happy that (in this release at least), they focused on features that genuinely improve and extend the search experience.
It's possible a form of AI (depending on your definition of the term) is powering some of these features, but it's used as an implementation detail, not a marketing term, which I appreciate.
AI is hot right now. It’s literally a buzzword. When projects add AI to something that doesn’t need it that’s nonsense. So the claim is not that all AI is nonsense. The claim is that Kagi is refreshing for not using AI nonsensically.
It's also a buzzword with no commonly-accepted definition. (If something is commonly accepted, it's that the term escapes common definition.)
This motivates a race-to-the-bottom mixed with a sort of Gresham's law. The worst actors go the furthest. And then they drive the good actors away from the term. "AI-powered X" is quickly approaching red flag territory for me, though I doubt that will be the case for the general population for some time.
"Kagi shopping results have no affiliate links and feature the most helpful, unbiased reviews."
This is a breath of fresh air. Ever since CNN Underscored started being a thing, I've grown to dislike more and more the swamp of affiliate link promoting content.
Some concent creaters on YouTube for example (like randomfrankp), are able pull it off, but deeply affects the credibility of most content.
Can never get into Kagi. You pay for privacy but if it’s ever sold your privacy will be retroactively taken and will be even more valuable since it will be known by the buyer that you pay.
Better off using a VPN and duck or Brave imho.
In any case these improvements do look very good. Public record search is cool. Lenses improvements look interesting as that seems the be the most unique feature.
I was interested in kagi, but it's not private at all. Your search term is right in the url, meaning the nsa can monitor your search terms. As an intellectually curious person, I search for things I shouldn't. I don't care if someone sees my local history, but I don't want to get put on a watch list for searching for production techniques for Compound B.
This is not correct. ISPs cannot see the actual URL being requested. Your DNS provider can see the hostname. The ISP may be able to see the hostname unless encrypted SNI is in place. The ISP can see the IP address you are connecting to.
You seem to know more about this than I do, but my layperson's takeaway from the Snowden revelations is that the NSA records every url we browse. Are you saying that's incorrect?
The only way NSA can record the URL without having infiltrated Kagi's datacenter, is for them to have broken the encryption algorithms behind TLS/HTTPS.
If that's the case, nowhere on the regular Internet is safe.
Snowden revealed that NSA has infiltrated all the major industry players (Apple, Microsoft, Google, etc.), also ISPs. But the only way NSA can know what your plaintext HTTPS URL is either by having access to your PC, or having access to Kagi's servers. Or as I said, that they've cracked encryption schemes everyone assumes to be safe.
You can POST it, and some privacy oriented web tools do that instead of GET, but the reason, as explained elsewhere, is not to hide it from NSA (GET parameters and POST are equally easy/hard to intercept), but to protect against other persons who has access to ones unencrypted computer (spouse, kids, colleagues).
The part after the domain is not visible to anyone between you and kagi as long as you access it over working https/tls.
The reason some privacy oriented sites hide it is because of local snooping on your end point, i.e. if someone has access to your device when you are not there and can look through your logs and by extension search logs.
I’ve been using Kagi for almost a year now and love it. I’m on the early-adopter plan, but just checked and haven’t been anywhere near my included number of searches in the last 6 months. I probably do some searches with DDG on my work machine, so I’d probably being doing less than twice as many as I actually am if I set up my work machine to use Kagi. (But I like to keep work and home stuff separate.) Overall very impressed with it and also using the Orion browser and loving it.
>>down-ranking sites with a lot of ads and trackers more aggressively
This is such an 'of course, why wasn't this done before' move!
Larding up a site with lots of ads and trackers is a blatant signal that it is low quality content attempting to game the search engine site.
It is surprising/not-surprising that Google didn't prioritize this downranking method years ago, as it would have massively improved the usefulness their results; but then it would also reduce the number of pages subsequently displayed using Google's placement engine...
Seems like low hanging fruit for Microsoft to undermine Google, downranking sites with heavy Google ad placements (since MS doesn't have a similarly large ad network that I know of)... and it'd be legit improvement of their results!
I’m currently on a grandfathered plan. When that’s up I’ve thought about whether I should look around elsewhere due to the price increase. Ultimately I don’t think I want search to be a race to the bottom. There’s no guarantee that any service provider will remain aligned with users and privacy long term, but kagi is bootstrapped and I’d rather pay more than push them toward needing to look for other monetization strategies that might be more adversarial to my interests.
I was more talking about their search. It was fine, but it wasn't worth the money for me. The browser didn't really pique my interest on release but I wouldn't expect it to be great on release (they're still a small company, but they're doing some really ambitious stuff which is cool)
Seconded. I gave it a few months, but it just wouldn't improve by itself. To be fair: With a detailed bug report (via Discord, making it excruciatingly hard to find duplicates/follow up), the problem would often be fixed in the next release.
But it wasn't 1 or 2, it was more like 20 easily spotted issues. Come on, I can't do all of your testing.
I think I've never said it this clearly because I don't want to offend (I still very much like the browser's concept), but it was just too painful for regular use for me.
This about a year ago. Your comment seems to indicate that it hasn't changed.
Not sure how orion is related. You can still use kagi search (or neeva) with any browser ...
I've rage quite Orion a couple times. But I keep going back. Some of the features are great. I use multiple browsers anyway so I'm ok with using Orion for the bulk of my browsing, then using other browsers each for certain sites or for certain uses. I'd love to work for them but I don't see a viable business model.
My SO uses multiple browsers as well although I can't make head nor tail what their "strategy" is for choosing. I doubt that having to switch up browsers is acceptable for the public at large.
I imagine there's a /reason/ you were using kagi, and that reason is probably privacy related? Why then would you use neeva as an alternative? Started by ad execs. There is zero chance neeva has your privacy in mind, no matter what they say about it. They are focused on pushing consumer products. This requires user profiles.
One thing that'd be nice is if they saved your selection of temperature scales. Currently they always show Celsius by default, but let you switch to Fahrenheit if you dare. I know that preferring Fahrenheit makes me a reprobate, but is this a hill they want to die on? Are they being paid by Big Metric? I'm logged in, go ahead and save it to my profile.
Big Metric is a group of French scientists from the 18th century going around with a 1 metre long, 1 Kg heavy stick of pure platinum, hitting people's shins whenever they hear nonsense like "it's cold, it's only 50 degrees this morning" or "is that a 39 and 3/8ths inch stick in your hand?"
I was an early adopter of Kagi but I cancelled the moment they started messing with the plans. I know I'd have the early adopter flag but I just felt betrayed by the messaging and I also don't want to use a search engine where I'm trying to think about how many searches I'm making and if I'm going over a certain number. The updates in the article look pretty good but I can't get past those two issues.
to anyone doubting: kagi really is that good. it has completely replaced my google usage - i haven't used google.com to search for anything in months. feels so good!!
They build on top of Google and Bing and adds its own index as well so it should be reasonable, but I haven't tested.
However, I think it should be easier to filter down searches to only show these results as Kagis advanced search operators mostly work reliably (and if they don't you can file a bug).
I’m not going to count my searches and pay for tiers. Honestly, I could live without caring about search. It feels like another thing that could go by the wayside, like music albums.
I have Apple Music because it comes with my family plan, and I have YouTube Music because it comes with YouTube Premium. If I had to pay just for a music service — like Spotify — they wouldn’t get my money. Even if it’s the better music streaming option in my car. But my car also plays Apple Music.
The dark pattens and ubiquity of some of these verticals means not having to care too much and still getting most of what I want. I don’t really want to care about search the way some people care about vinyl.
Do I have to be the guy to say that I still listen to music as whole albums and store my music for offline? Artists in a lot of genres still release the album as 'the music', where a song is like one chapter of the book and you wouldn't just read one chapter. I only stream during the discovery phase for new music and even then it's usually: one track good, okay then it's time to try the whole album. There has always been folks that prefer records to radio and it feels absurd to me that this would be going away.
I think individually we all have the right/option to make our choices, but I'm speaking to the market makers. If you want to be a vinyl shop (Kagi, Bandcamp), go for it. But you're not going to be Spotify -- or the Apple Music Store ($0.99 per track, remember?) -- in a world of TikTok and bundled music platforms.
And I have my personal version of vinyl, I listen a lot to soma.fm and Radio Paradise. It's very interesting to me personally how few other services like those really exist out in the world. There are things like DI.FM and Mixcloud, but I don't know anything else like Radio Paradise (music channels, forums, etc). I've scoured all the channels and platforms expose on TuneIn and the like, and there aren't quite as many variants as I would personally expect. So the niche music world isn't even as numerou$ as you'd want a long tail to be.
- How is your search in languages other than English and local to countries other than the US and Europe? In particular, I am interested in Russian and Turkish. I remember that previously (a year or two ago) it was not very good. Also, does your GPT based search work for other languages?
- How is phind.com related to kagi? I remember that at some point, I was searching for something on phind and got an error saying that Kagi is out of capacity or something like that.
- What underlying search providers do you currently use?
User since beta, subscriber since he started charging, and my primary search engine since January. About 500 searches a month. Very happy customer. I do a monthly sweep of subscriptions to make sure we are using all of it, and I never think about dropping Kagi. It keeps getting better, not worse like every other search engine, and the reason is because the user is the customer, not advertisers. Every ad-supported medium turns to shit, no matter where it starts out.
Amusingly, I searched for Kagi's founder and the new public records search section in the main results showed his citation for failing to stop at stop sign and a 2020 patent for "a search interface [that] allows the selection between multiple presentation modes".
I use Kagi exclusively. The results are just as bad as googles, but isn’t covered in spam. Given the sorry state of search quality in the SEO garbage world we live in, that’s high praise. I’m happy to spend the money every month.
Thanks, I felt similar last time I used it. Wasn’t magic like I remember Google being.
I wonder what happened to Google?
My theory is maybe their own systems have become too complex for them to understand and tune now. But they’ve invested so heavily into “AI” they don’t known where else to go in order to fix it ?
I’ve here seen or experienced a search product quite like Google used to be. It truly felt like magic in it’s day.
What happened is that it got gamed, and lost the game. Spammers are better at getting ranked than Google is at detecting them, and the Web has died as a result.
Pre-google internet was way more structured because it had to be (since there were no good organic search engines). Google got in there and had a great algorithm that worked in that world. It got gamed and now content is being produced to make sure you appear on top of searches that might not even be what you were looking for.
Pre-google internet was way more structured because it had to be
I have a feeling many things in the digital space are similar to this and will go the same way, code is a pretty good example. Longer term, I wonder what the generation of massive amounts of code will do to the training data.
For me, i prefer to (attempt to) have clear incentive structures. Kagi seems inline with my goals. Search costs, and i want to do what i can to not be the product.
That's fair. Eventually I want to setup and start using one of the decentralized search products that exist. They seem like a good idea, but more investment than I've been willing to commit so far.
Personally I _want_ to pay for this service. I do not trust free search products anymore. Well... they can't really be free, someone somehow has to pay for the consumed resources.
> Kagi shopping results have no affiliate links and feature the most helpful, unbiased reviews."
Wonder how they determine if a review is unbiased or not. Any attempt at automating this process means someone else will figure a way to game the system over time.
I will throw this on the table (and quickly hide my hand behind my back):
If the price model (IMHO correctly) is connected to the actual costs of the search, why not make a "searches account"?
You pay 5$ per month as a "baseline" and each first of the month 200 searches are credited on your account.
Then, if you use less than 200 searches, at the end of the month you have a credit that goes over to next month.
If your credit goes to 0 within a month, you can buy "packets" of searches, 5$ for 200, 10$ for 500, or something like that, and recharge your account.
The reverse image search is going to be super useful for me as Google recently ruined theirs in attempt to promote their Lens app.
Overall, while I'm still not sure whether Kagi is worth the money I'm paying, I keep doing it so that they can release improvements like this one. Their quick development speed is great to see.
> The reverse image search is going to be super useful for me as Google recently ruined theirs in attempt to promote their Lens app.
I always found Google's to be pretty bad, I like to use Yandex for reverse image search since theirs is crazy good.
> whether Kagi is worth the money I'm paying
Honestly, searches being limited just kills it for me, I don't feel like paying per search nor $25/mo, I do close to 200-250 searches per day, sometimes >500
Sorry to hijack this thread but does anyone know a project that indexes webpages as I visit them in my browser? I revisit a lot of sites often and would like a private search engine that can search the content I have previously visited.
I'm downloading and rendering the web page on my browser anyway, so it seems like a giant waste not to index it while the page is open.
I knew Kagi is paid but I didn't know they had a limit on how many searches you can perform, which is also rather low. But what I really didn't like is how dishonest they are. The lowest tier costs $5 and includes 200 searches per month. That is 6.66 per day! But they say "Includes enough searches for 99% of internet users" because "The average person searches Google three or four times per day". They know very well that someone for pays for Kagi (or even knows about Kagi) doesn't perform three or four searches a day.
FWIW I've been using Kagi exclusively since Nov 2022, and here's what my monthly usage looks like:
Apr 2023 - 895
Mar 2023 - 1047
Feb 2023 - 650
Jan 2023 - 603
Dec 2022 - 1367
Trends much higher when I screw something up on my network!
I have been asking GPT-* a lot of what I would have previously searched for. And there's still a lot of searching `+reddit.com` because they can't seem to solve search.
I pay $10/mo and am grandfathered in as an "early adopter" with 1K searches. I would be willing to pay more because I've been so happy with it. It's simple stuff -- it respects and obeys my double quotes and boolean operators. If my query returns no results, I get no results. No advertisements, period. The ability to pin, raise, lower, or eliminate the rank of domains in your results.
It's just nice to feel a modicum of respect as someone who's been using computers since shortly after they could read.
Yeah. It's just sad that it was what search was, for free, before. This idea that you search for things, and you find them, instead of paid per word shitty articles and ads. Nutty concept in 2023.
At this point, the free options are so bad I'm considering writing my own. It won't do 50% of what the big ones do but it will do it well.
Yea i hate this framing. Nothing is free and we should be wanting to move towards payment ourselves. Yea, i'd prefer "payment" to be the cost of locally running a server, but i'll take what i can get.
The days of believing free is viable for any online service should be long gone. Hell it's difficult to even determine if paying for a service actually protects you from being the product - because often that isn't the case. However free should be a no-brainer these days. There is no free.
Apparently I've been using Kagi a year longer than that but I can't work out how to find my stats though. I signed up to the Team plan for my company as soon as I saw it was available - before Kagi I couldn't imagine paying for search, now though, I couldn't live without it. It's impossible to overstate how useful a functioning search turns out to be.
I started using Kagi exclusively around the same time, I'm almost exactly on the same boat as you. Interesting to see that even though you seem to have a bit more variance than I do we land around the same range of searches/month:
Apr 2023 651
Mar 2023 708
Feb 2023 613
Jan 2023 611
Dec 2022 759
You are correct in saying that almost all current Kagi users are in the other 1% category (tech savvy professionals who search a lot).
But this plan is not meant for then. This plan is literally made for the 99% of all users who search the web 2-3 times a day, as the statistics show. We would like Kagi to have much broader adoption outside the HN crowd, and a low price plan is essential.
How would you change the wording to make this clearer in a way that does not strike you as dishonest?
> How would you change the wording to make this clearer in a way that does not strike you as dishonest?
I don't necessarily agree it's dishonest, but I think it's a bit weird to have a pricing tier than effectively doesn't work for anyone who would pay for the service.
As you said, almost all Kagi users are not in this 99% of users. Having a plan that caters to these users probably doesn't benefit many people in that group because they won't pay for search, but for users who would pay for search they might see that tier and feel that the quotas are stingy and they're being forced up the pricing tiers for behaviour that to them feels normal.
I don't know what your subscriber base looks like, but as an armchair spectator... I'd probably drop the $5/m tier and then emphasise that the "Professional" tier comes with "10x the number of searches that an average internet user makes", or whatever the multiplier actually is.
Also it would probably go some way towards assuaging fears of running out of searches if unused searches rolled over month to month. This tends to feel fairer in general.
> but I think it's a bit weird to have a pricing tier than effectively doesn't work for anyone who would pay for the service.
When we had only the $10 plan, we were getting messages from users that that plan is too pricey for them when they don't search as much.
Hence the $5 month and now about 5% of our users are on this plan. (and Kagi still has zero marketing spend). You have to start somewhere. This gives us an opportunity to onboard almost everyone to Kagi.
> if unused searches rolled over month to month. This tends to feel fairer in general.
Yes, except something still has to pay for all the free trial searches and our salaries. Selling product at cost or below is something VC funded startups can do, which we are not.
> Yes, except something still has to pay for all the free trial searches and our salaries. Selling product at cost or below is something VC funded startups can do, which we are not.
Sidenote, i'd really enjoy the rollover. At least to some degree. Not infinitely of course, but being able to mitigate the cost of some overages would be really nice.
Aside from that i'm a new customer (1st month), having gone through the trial and now on the $5/m plan. I'm quite happy. I expect i'll upgrade to the $10/m, but i hope i don't have to upgrade to the $20/m because i'll be debating if it's worth it. Regardless, a bit of a rollover price would help a ton with any cost aversion. Even if only up to 1 extra month, or w/e, of rollover. Anything would help.
Thanks for Kagi tho, i'm very happy so far! I also really love the continued work and improvements. It feels like i'm not just paying for the current product, but buying into a larger future product. Which also goes to mitigate cost aversion.
I don't know your unit economics (but they sound fascinating), but generally free trials are considered a marketing expense and would be worked into the CAC (customer acquisition cost). Then you can trade-off CAC with lifetime value (LTV) and a target payback period – i.e. the break-even period.
Let's say a search costs 1c, the free plan therefore costs $1. With a 5% 30d conversion rate that's a $20 CAC. On $10/m for 700 searches, that's just under 7 months to pay back which is quite long, but drops to just over 3 months if users average ~400 searches, so I can see why you don't want to roll over.
That said, if you believe you have a big LTV/long retention, retained users past their payback period should be more than enough to pay salaries/R&D/etc, so maybe there is more flexibility.
> Selling product at cost or below is something VC funded startups can do, which we are not.
I know it often feels like this, but I think the answer is more complicated. It's often hard to know what cost actually is – when you're hiring, growing, and selling a service running on tech that is hard to price.
I'm interested in how you know your cost per search at such a level of precision. It suggests to me that either you've done _way more_ work measuring it and optimising your infrastructure than I expect for a company of Kagi's age, or that it's a fairly naive number (understandably so!) based on dividing infra cost by number of searches. If it's the latter, are there economies of scale that significantly change the number if you have, say, 10x or 100x the user base?
The answer to the questions is long, nuanced and interactive. I'd be open to explaining it in a more interactive environment, for example our Discord server. kagi.com/discord - feel free to ping me there @VladP
Don't change anything. Your marketing is just fine. The commenter you're responding to is just voicing their surprise at discovering that most people don't actually use the internet as much as they do, or they're realizing they don't user it as much as they thought they did month over month, and, either way, consequently feeling insecure. They don't understand the difference between current user base and target market. You don't market to your current user base. You market to your target demographic. Like you said, if Kagi wants to attract average internet users, then having a plan that suites them is pretty important and in no way dishonest. I experience the same knee-jerk reaction when reviewing your recent pricing plan changes/updates. Then I thought about it for two seconds and obviously understood.
I currently subscribe to the $5 plan, and I'm happy with that.
Most searches I do are simple, any search engine will do. So I don't use Kagi for those. When it's not trivial, I just start my search with "k " to use Kagi. I easily stay below the 200 searches limit then.
For sure it would be more convenient to have unlimited searches, but I'm not willing to pay for a higher plan.
Having a low entry is very important to get new users on board. But the included searches of 200 in the standard plan is extremely low. Even a normal user will hit this limit quickly (if not, they are not your target group - or why should they consider paying for a search engine anyway). This makes the standard plan absolutely unattractive even for non-tech-savvy professionals. But the professional plan also only offers 700 searches for $10.
Personally, I think Kagi needs to reconsider their pricing plans and either lower the monthly price or significantly raise the limit for included searches. Otherwise I don't know if I can judge $10 per month for a premium search engine whose service will not even last me for a whole month.
Don't get me wrong. I love the product and would be happy to pay for it, but your current pricing plans are just not very convincing to me and do not fit most peoples needs. Nowadays, a search engine is something you use on a daily basis. And you don't want to care about it.
> But the included searches of 200 in the standard plan is extremely low. Even a normal user will hit this limit quickly
They will not, a 'normal' user searches only 100 times a month.
> (if not, they are not your target group - or why should they consider paying for a search engine anyway
Because they want higher quality search experience, have their privacy respected and/or do not like the entire order of things on the web where they are constantly being the product.
> I think Kagi needs to reconsider their pricing plans and either lower the monthly price or significantly raise the limit for included searches
We will. Pricing is not a matter of our mood though, but of economics, and once you remove advertisers from the equation the true cost of search surfaces. We were able to bring it down to just 1.5 cents per search.
> Nowadays, a search engine is something you use on a daily basis. And you don't want to care about it.
Isn't something you use every day, tens if not hundreds times a day, and helps you accomplish important stuff, in a more productive way worth the $25/mo then?
> Because they want higher quality search experience, have their privacy respected and/or do not like the entire order of things on the web where they are constantly being the product.
I really wish it would be that way. But from my personal experience non-tech affine people don’t care about privacy or ads that much (as long as it’s free). Just take my wife and father-in-law as an example: They have Google as their startpage in the browser. And instead of directly typing the URL into the browser bar, they will always do a search for it. I told them so many times, but they literally don’t care.
And they are not completely wrong since it works for them. 99% of the time Google will return what they are looking for within the first result. No chance I can convince them to suddenly pay $5-10 for the same thing.
I'm currently in Early Adopter ($10) plan. If I move to $5 tier plan and move back to $10 plan in the future, which plan would I be at that time? Is it still Early Adopter?
BTW, I'm a happy Kagi early adopter currently subscribed at $10/mo. I do have concerns about even my limit, though. After I switched all my browsers (on all my devices) to use Kagi as the default search engine, I regularly exceed 1k searches per month.
I'm very curious how you measure the "average" user's usage. It seems hard for me to believe that even an "average" user (which I grant I am not) who has switched their browsers' default search engine to Kagi on their (one) computer and (one) phone -- assuming an average user has a computer and a smartphone -- uses less than 200 queries per month.
We've all been trained for years now (by Google) to type searches (not urls) into our browser's address bar. For so many people, the first action they take to "access the web" is to type something that is not a URL in their browser's address bar. I suspect this behavior is even _more common_ among "average" users ("power" users seem more likely to actually type "mail.google.com" instead of "gmail", for example).
Yes potentially, we found out that incorporating billing system is painful and takes away resources from our (small) team that we'd rather spend on building better search experience (like this update).
~7 per day is on the low side for me. I will normally rerun the same search with different terms multiple times, especially if I'm looking for an answer to something highly technical, like easily 5-6 different searches for the same thing. And this happens multiple times per day.
There's something perverse with the incentives here: they make more money if you have to perform more searches.
It doesn't quite sit right in the same way github actions charges per minute: The slower their runners are, the more money they make.
And both of these scenarios there is no user agency to assist with that past a certain point.
> they make more money if you have to perform more searches.
I believe the reason for the pricing is Kagi's cost structure. Accessing for example Google index is not free. Since they don't show ads or monetize the user in other ways, they need to pass the costs to customers.
Pay-for-what-you-use models are not common for consumer services, but I think it is a healthy pricing model. Better than fixed price packages, where you assume 80% of low volume users will cover the costs for the 20% heavy users
It’s tricky because there are subtle psychological effects when marginal use directly incurs marginal cost. It makes each use a decision.
When you assemble tools for cognitive work, it’s important that they have low overhead. Thinking about the financial cost of using a tool is a small context switch that slows you down. Thus a bundle of prepaid stuff increases the utility of the service beyond what you’d get with pure pay per use, even though the latter is more economically efficient.
> they make more money if you have to perform more searches.
Their model pushes people who make a large # of searches towards the unlimited plan, where incentives are always aligned. I personally find that fair because I expect to get at least as much value from my search engine as I do from my IDE. $20/month seems reasonable.
Their model also allows them to have lower expenses against the lower tier plans if users of those plans make fewer searches than their quota. That's the incentive you're looking for. As long as users are not reaching their quotas, or are on unlimited plans, the user incentive is aligned with the business' incentives. And shortly after exceeding their quotas, users will probably upgrade their plan.
The perverse incentive only exists within a few "holes" between the plans, and serves to encourage the user to upgrade. I believe that by limiting the window of the perverse incentive, it should discourage goal-seeking to fit customers within that window. The more optimal outcome month-to-month should likely be to improve customer experience and get more signups, rather than juicing the current customers for limited additional gains inside those constrained windows.
It's completely honest. You don't have a sense of typical use. I'd argue search power users make fewer searches a day, too, but like you, and unlike Kagi, I can't back that up with sources.
I was vaguely aware of Kagi from their posts here in the past and thought it sounded interesting but hadn't bothered trying it out as I'm fairly happy with DDG.
I just gave it a quick try there. While some of the features seem promising, the website ranking for example, what they consider a search doesn't feel great from a user perspective. Each step below resulted in part of my quota being used up:
1. Making the initial search.
2. Switching to image results.
3. Waiting a few minutes before switching back to the web results.
4. Viewing page 2 of the web results.
5. Ordering results by time.
6. Changing the sort order to descending. The previous step executed another search without allowing me to modify this.
7. Limiting the search to results from the past 24 hours. This returns a totally different set of results from when we merely sorted by most recent.
8. (Sometimes) Clearing the filters.
9. Opening the Lens > Edit menu, which has a big cross at the top right making it look like an overlay. Instead, it actually sends a GET request for the search again, using up your quota if you've exceeded whatever the grace period is.
This, along with nearly tripling the price for existing subscribers that want the same functionality does not inspire confidence or goodwill. The change might've been more palatable if they'd kept a single subscription and put a monthly price cap on additional search fees to match the $25 unlimited tier.
Maybe I'm not the target user but the 100 lifetime trial searches just means I probably won't return once I've ran out. 25/100 are already gone to evaluating what they charge for and I don't think I'll get a good idea of whether they're actually providing value over existing providers with the remaining 75. The amount charged isn't low enough that it's not a decision.
The main justification for all this seems to be that adding AI to the search experience is expensive. The obvious rebuttal is that if AI is such a big value-add why don't they just charge for that service in a higher tier or as an add-on.
The first 8 sound about right, (9 is probably a bug) and all of them perform a new search which costs us about 1.5 cents to do. There are only three ways for the cost of that search to be paid today:
- Advertisers paying it for you
- VC money paying it for you
- User pays for it
Kagi is built for people who want a search engine built for them. There is no way around it but to pay for search like you would pay for donuts you eat in a coffee shop.
AI is no justification, search is just expensive. Google is making about 4 cents worth for every user search. We are selling search at 1.5 cents while building a completely user-centric product. If you know of other ways to do this, we are all ears.
Most of the above examples of unnecessary searches could be avoided if the UI was designed to help the user avoid unnecessary searches instead of somewhat cloning Google's UX, which is designed around searches being effectively free for them.
Very much this. Probably 75% of my searches are “Just looking something up on Wikipedia” or “Just looking something up on StackOverflow”. If Kagi were to let me search these sites without the cost of a full Google lookup, its plan would work for me once again and I’d return to being a customer.
In the interests of fairness, they do support, and do not charge for, "bangs" that defer the search to that site. They even support creating your own custom ones which is really nice. Wikipedia is one of the default ones. According to their docs you can search it with "w! query", "!w query", "query w!", "query !w" or even add the bang to a "quick" list and use something like "w query".
Unfortunately, you do give up consistency and whatever filtering/ranking/customisation Kagi offer.
I appreciate the response and I'd like to reiterate that I might just not be the right user with the right expectations. Still, I hope it's useful to hear why someone might be on the fence about the product.
> (9 is probably a bug)
To clarify 9, the request is sent when clicking the cross on the Lens > Edit page rather on than visiting it. This also occurs when returning from maps. In both cases a new search is only billed when the cache has expired. Interestingly, from a totally naive user's perspective, maps don't appear to use up a search.
> If you know of other ways to do this, we are all ears.
I don't have an issue with charging for individual searches. On the whole I think it's a reasonable idea. I do think that some aspects of the UX encourage billing superfluous searches and that, on a purely financial level, Kagi is dis-incentivised from improving this.
Taking steps 5 and 6 as an example: there's no reason for these to be separate billable events. When viewing search results the user interface simply doesn't allow the user to select the sort order in the same operation they select the field. I sincerely doubt that this is intentional but it's easy to see the cynical take.
I think there's also a decent case for longer caching of results, giving users the option to refresh results if it's important they aren't stale. This avoids the current situation where a user might spend some time visiting the firsts webpage in the result set, return wishing to continue perusing subsequent results but end up triggering a billable event. The same applies should someone want to swap between web and images.
A quirk I noticed when trying to trigger shopping results was that searching for "best hair dryer" will display the shopping widget whereas searching "best hairdryer" does not, despite "showing results for best hair dryer" appearing as part of the search.
Maybe I'm misunderstanding the feature but searching for variations on "cordless drill", "best cordless drill", "best cordless combi drill", "best cordless percussion drill" never returned any shopping results, should I have expected it to?
To expand on my prior point as to why moving to a total number of trial searches might be harmful to conversions. In evaluating the billing rules and shopping results, I'm now on search 46/100. I haven't yet got into a flow with the product. That might change over the next 50 searches but I still haven't explored: redirects, lenses, personalised results, fine tuning search settings, or just the broader search quality.
> I do think that some aspects of the UX encourage billing superfluous searches and that, on a purely financial level, Kagi is dis-incentivised from improving this.
Kagi is selling searches at cost, and we make money on subscriptions. Making a suboptimal UX id definitely not what we are doing, as people pay us for UX to be optimal.
You point out some valid use-cases there and the reason we haven't addressed them so far is probably because they are rare and users didn't care enough to post and upvote them on kagifeedback.org for it to make a difference.
> Hanlon's razor in general applies well when thinking about Kagi, expect it is not (always) stupidity but lack of resources.
Please don't feel that I was making any hard judgements about Kagi or yourself based on the points I mentioned. My only intention has been to learn a bit more about the product and provide some insight into what someone evaluating the product for the first time might feel. I appreciate that the points may come across as hostile/challenging but I'd be happiest if they were rebutted easily.
Your present response does raise some concerns however.
> Kagi is selling searches at cost
This is a different claim from your blog post on 08/03/23 where the cost of each search was 1.25 cents. If this isn't the case, you're losing money on any annual unlimited subscribers that make more than 1,416 searches each month.
> and we make money on subscriptions
If this is the case, why not differentiate the subscriptions based on their value-add features and operate an upper bound of $25/mo on the amount charged for searches for all tiers?
If you're making money off of the subscription, why don't the search quotas roll over?
Together, these three points give the impression that the aim is for subscription search quotas to be under-utilised rather than to provide the searches at cost.
> You point out some valid use-cases there and the reason we haven't addressed them so far is probably because they are rare and users didn't care enough to post and upvote them on kagifeedback.org for it to make a difference.
That's a fair point regarding general feedback. I can't speak as your average user but as someone that wasn't already invested in the product they were areas that introduced friction in the decision of whether I'd like to pay for the product or not. A user evaluating the product is unlikely to post why they haven't converted to kagifeedback.org, even if they're aware of the site. It's worth bearing in mind that issues for potential users may not be a problem for existing users, apportioning time to the desires of both groups is a difficult balance.
> and is planned on our roadmap, just not prioritized (just one upvote) and we had bigger fish to fry.
My personal perspective is that it's absolutely unacceptable for a company to double-charge due to their own UI decisions. Spurious billing in general is something I would expect to be treated as top priority on an ongoing basis. Treating it's occurrence as a feature request raises serious concerns, especially regarding how similar/more impactful situations might be handled.
I'm disappointed, there's a lot to like about the product itself.
> This is a different claim from your blog post on 08/03/23 where the cost of each search was 1.25 cents.
Yes, cost of search has significantly increased since. Microsoft raised prices 6x and it is 2.5 cents to do a search with the Bing API alone. We are trying to absorb much of that through creative ways so that users do not see it.
> If this is the case, why not differentiate the subscriptions based on their value-add features and operate an upper bound of $25/mo on the amount charged for searches for all tiers?
Because the cost of all other features pales in comparison to the cost of search. In general if a feature does not costs us anything we do not charge the user for it (example: bangs are free).
> If you're making money off of the subscription, why don't the search quotas roll over?
Two main reasons:
- It means more billing systems to build and we are eager to work on search features like this update
- Something still has to pay for all our additional costs like free trial account searches and salaries
> It's worth bearing in mind that issues for potential users may not be a problem for existing users, apportioning time to the desires of both groups is a difficult balance.
Agreed and it is a matter of product roadmap prioritization. While the issue was previously raised, it had only one upvote. Now that we got more alarming feedback it was prioritized internally and 5 and 9 from your list should be addressed asap (others do not really apply as we do run a full search for those).
> My personal perspective is that it's absolutely unacceptable for a company to double-charge due to their own UI decisions.
I agree with this perspective and as I hopefully explained that was not the intent, but a bug.
Mmh, so your excuse for using dark patterns that are designed to overcharge users is that not enough of your users have noticed and complained? Are your users supposed to be UX experts?
Our users notice and (rightfully) complain about everything that is not right - because they pay for the service.
My "excuse" for some of these things not being fixed is that they have low perceived impact in practice and/or lack of resources to address all issues at once (hence roadmap).
Thanks for posting this, I've been curious about what counts as a search to Kagi. On one hand, I get it: making requests obviously costs them money, and they certainly don't owe me anything as a non-user anyway. But I'm even less likely to switch to it with this knowledge, because this just doesn't fit well with how I search for things. It doesn't jive with the mental model of what I would think of as a search in my head as an end user, where changing the actual search terms is obviously a new search, but filtering, reloading, viewing the next page, etc, are just part of the same search.
Great, I think we deserve an honest discussion about this. When you say "hey my product is $10/month, please adopt it", users are going to factor that pricing information into their decision of whether to adopt the product. That's the bait. Later, once you have persuaded the consumer to adopt your product, changing the terms in a way that are significantly less favorable to the consumer (the switch) has the appearance of fraud. You further misrepresent Kagi as some kind of expert in the search space, while presenting a bait and switch defense basically stating that you're only now learning the "realities of the search market". So, which is it? Isn't "we didn't know the realities of _____ market" just a cop out that anyone committing fraud can claim? Why not grandfather in your early adopters? I think if you could answer some of these questions in good faith it would go a long way -- as it stands you're just embarrassing your organization.
I like duckduckgo but now I prefer bing, mostly for the bing chat inclusion. Yet I miss the ! shortcuts from DDG, and after trying it at a friend, I'd like to use Kagi too (so yes, I want everything and the kitchensync)
Would you have a product that would also use ! shortcuts and do some bing chat like search on top of Kagi?
By bing chat, I mean not just openai, but doing queries to augment the answer. I'd be happy to pay for it, especially if you offered an API to specify a few basics (ex: what is XX? Please augment your answer using results using this search query:" ")
Kagi doesn’t have a chatbot, but they have recently added some thoughtful Generative AI features. They tend to start with the search results and then summarize.
> They know very well that someone for pays for Kagi (or even knows about Kagi) doesn't perform three or four searches a day.
Fwiw, i pay for Kagi and i am totally inline with their estimates. Maybe a pinch more, 7 instead of 4 atm. But at $5/m i can't complain, and so far (it's new to me) i've found myself not falling back to Google.
I was on DDG for ~2 years and i constantly opened up Startpage (aka Google) and sometimes even Google. I've been very happy with Kagi, so far.
Not sure i'd pay $20/m, but $5/m is a very nice value to me. I hope things work well for them.
edit: Sidenote, but the first thing i noticed is just how damn fast Kagi is. Not really a selling point to me, but wow.
They literally provide sources for those claims. I have no idea how you can say with a straight face that they're dishonest. They're very clear and transparent about how their plans are limited.
It's also extremely easy to check what your usage is for the current month and get a monthly overview. If you happen to use more than the 200 searches for the standard plan, they charge 1.5 cents per search. If you blow past it and end up using another 500 searches, which matches the $10 plan, you're still only paying $12.5.
They also provide soft and hard limits to how many searches you can use to avoid unexpected charges.
Like, I honestly don't know what else you expect them to do.
edit: Also, for reference, as a heavy user who basically lives in their search engine, my total searches for previous months are: 786, 782, 610, 519, 717.
The GP didn't say they're lying about the number, the claim was that the number isn't representative of the amount of Google searches Kagi users are likely to perform.
Kagi pricing page does not talk about Kagi users but Internet users, which Kagi is trying to attract with the $5 plan.
Almost all of current Kagi users are in the other 1% category, but that doesn’t change the fact that vast majority of internet users search 2-3 times a day and that Kagi would like to attract them with this plan.
The obvious implication of putting that on a pricing page is that the average internet user accurately represents the average Kagi user.
I have no idea if that's true or not, but if it isn't, then there isn't really a way to "word it differently" - including the information at all is misleading.
I’m trying to follow the complaint but not getting it.
Are you saying that the average Kagi user, who is an internet power user doing far more than 4 searches a day, is going to be misled into believing they are an average internet user and that they really only do 4 searches a day?
Why is it not representative? I was surprised by this recent Kagi pricing change but after reviewing my search stats I realized that their estimation is pretty on point.
I don't know, it seems low, but it works for me, but I still just use Google for most things. Like most of my searches are simple things where Google is fine. Recipes, weather, easy queries. For more advanced stuff where Google doesn't work I use kagi with fair success. Like today I was looking for an obituary from 1843 and Google just kept giving me shit about the Ukraine War. Kagi worked! But also gave me some Ukraine War results, but somehow the most relevant page was near the top of Kagi but nowhere to be found on Google even with multiple searches.
So you essentially have to budget your queries to make sure you have enough. Which is unproductive. The "simple queries" should already be cached aon the server and should not even count to the quota.
To clarify, there is no limit on Kagi. Searches are 1.5 cents per search (after those included with subscription) and you can do as many as you’d like. Or get an unlimited plan. A lot of things don’t cost a search (bangs, reloads etc..)
i hit my 1000 search limit yesterday, and my cycle is up in a little over a week. I was a little miffed, but there is a helpful button that just redos your search in google
it was my first time using google search in a few months since i started kagi, but.... have google's results become better?
maybe voting with our wallets worked, and google is rolling back some of their less popular recent ranking changes
I was a customer for a few months until they decide to raise it. I do, on average, 30-40 searches per day. I'll need the $25 plan and while they are (subjectively) better than Google, I couldn't really justify the price bump.
It also doesn't say what currency it's in anywhere on that page. Is that my local currency or (presumably) USD? That makes a very large difference to my price.
Internet is I guess, next after a working computer with a working dev environment, the thing that makes most difference for many of us developers.
For me it has value in two different ways:
1. time saved by arriving at the correct answer quicker and not waste a lot of time trying to refine my search to the point where even Google cannot misunderstand it. If your relationship with Google is better, more power to you.
2. I am a happier man as a result of not having to fight my search engines day in and day out.
I also notice that unlike a number of others who like me feel they use search a lot, I don't get close to the quotas.
I guess one thing I might do differently (based on years of frustration with Google and DDG) is I take something like a depth-first approach based on my first search:
- Using Tree Style Tabs I open every result on the first page that looks relevant in a new tree (ctrl-click to open in new tab, nesting is automatic)
- I then look at each of them, ctrl-clicking every link that looks relevant, recursively
This works for me and I have been doing it for years. One nice side effect is it gives me a permanent somewhat organized tree of most of the most relevant docs. In some cases I even select such trees and (with the help of another extension) export them as nested links in markdown in rhe hope that I can use as a more permant table of contents for the topic.