Kagi: 50,000 people paying for search, and a profitable bet against ads
Kagi launched a paid, ad-free search engine in 2022. It reached 20,000 paying members in January 2024, said it was profitable that May, and celebrated 50,000 members in June 2025.
- Product
- Kagi
- Founder
- Vladimir Prelovac
- Launched
- 2022
- Model
- Subscription search, with a companion browser
Published · 3 min read · 4 sources
Key takeaways
- Kagi launched paid, ad-free search on 1 June 2022. 210 people became customers on the first day, and 121 of them were still customers three years later.
- It reached 20,000 paying members in January 2024 and said it had achieved profitability by 30 May 2024.
- On 9 June 2025 it celebrated 50,000 members, including over 5,000 families, and said it had been break-even profitable for a year.
What it is
Kagi is a paid search engine with no ads and no tracking. Users pay for the product, rather than advertisers paying for their attention (Kagi, What's next). It was founded by Vladimir Prelovac, who is CEO (Kagi, What's next). The company also builds Orion, a WebKit-based browser with the same "pay for the product" model, plus later tools such as Kagi Assistant and Kagi Translate (Kagi, first three years).
Timeline
- 2019: Prelovac writes that Kagi was conceived this year, with the idea of building an "internet company" around search, a browser and eventually email (Kagi, first three years).
- 1 June 2022: Kagi launches search as a paid, ad-free product. 210 people become customers on the first day (Kagi, What's next; Kagi, first three years).
- 2023: a Family plan and unlimited searches for $10 per month help growth; the company later credits that pricing step with reaching 20,000 members by early 2024 (Kagi, first three years).
- 24 January 2024: Kagi celebrates 20,000 paying members, including over 1,500 families, and sets a next milestone at 50,000 (Kagi, 20,000 members).
- 30 May 2024: two years after launch, Kagi "achieved profitability". It also says 93 members had joined as investors through two rounds, the team had grown to 37 people, and it had become a Delaware public benefit corporation (Kagi, What's next).
- 2 June 2025: in a three-year update, Kagi says 121 of those first-day customers are still members, the company is at "almost 50,000 customers", and it has been "break-even profitable since this time last year" (Kagi, first three years).
- 9 June 2025: it reaches 50,000 members, including over 5,000 families and nearly 200 teams, and says it is approaching 1 million daily search queries. Orion has "surpassed 1 million downloads". It also announces a limited free search portal and a Kagi for Libraries program (Kagi, 50K).
Why it worked
- The customer is the user. Kagi's bet is that search quality got worse because the real customer was the advertiser. Charging users lets it rank pages by usefulness instead of ad revenue (Kagi, What's next).
- Pricing that people could actually pay. Unlimited search at $10 a month, plus a family plan, turned a niche paid product into something that could reach tens of thousands of members (Kagi, first three years).
- Word of mouth, not ads. The 50,000-member post says Kagi depends on community recommendations, which fits a product whose whole pitch is rejecting ad-tech (Kagi, 50K).
- Profit as a survival feature. In a market dominated by Google, being break-even profitable means the product does not need a new funding round to stay online (Kagi, first three years).
What to learn
- "Free" search is not free. If you cannot charge, you are usually selling attention. Compare Neeva, another paid-search attempt that raised venture money and shut down.
- A small, loyal base can fund a niche that looks impossible at consumer scale. 210 first-day customers, and 121 still there three years later, is a strong retention signal (Kagi, first three years).
- If you ask people to pay for something they used to get for free, the product has to be noticeably better on day one. Kagi's launch story on Hacker News is about finding a result Google missed (Kagi, first three years).
Risks and what didn't work
- Scale is still small next to Google. 50,000 members is a real business and a tiny share of search. Prelovac has talked about a longer-term aim of 5 million customers; that remains a goal (Kagi, 50K; Kagi, first three years).
- Member-investor rounds are still fundraising. Kagi is not a pure bootstrap. It raised from users and spent a large share of earlier investor funds on 20,000 thank-you T-shirts, which is generous and also a reminder that capital was finite (Kagi, What's next; Kagi, 20,000 members).
- The product surface keeps expanding. Browser, Assistant, Translate, Maps and Mail all have to be staffed. A small team can lose focus if those bets do not stay tied to search quality (Kagi, first three years).
- We did not find a later official member count after June 2025, so the figures above stop at that date (checked 11 November 2026).
FAQ
Short answers drawn from the sections above.
What is Kagi?
Kagi is a paid search engine with no ads and no tracking. Users pay for the product, rather than advertisers paying for their attention (Kagi, What's next). It was founded by Vladimir Prelovac, who is CEO (Kagi, What's next). The company also builds Orion, a WebKit-based browser with the same "pay for the product" model, plus later tools such as Kagi Assistant and Kagi Translate (Kagi, first three years).
Why did Kagi work?
- The customer is the user. Kagi's bet is that search quality got worse because the real customer was the advertiser. Charging users lets it rank pages by usefulness instead of ad revenue (Kagi, What's next).
- Pricing that people could actually pay. Unlimited search at $10 a month, plus a family plan, turned a niche paid product into something that could reach tens of thousands of members (Kagi, first three years).
- Word of mouth, not ads. The 50,000-member post says Kagi depends on community recommendations, which fits a product whose whole pitch is rejecting ad-tech (Kagi, 50K).
- Profit as a survival feature. In a market dominated by Google, being break-even profitable means the product does not need a new funding round to stay online (Kagi, first three years).
What can indie builders learn from Kagi?
- "Free" search is not free. If you cannot charge, you are usually selling attention. Compare Neeva, another paid-search attempt that raised venture money and shut down.
- A small, loyal base can fund a niche that looks impossible at consumer scale. 210 first-day customers, and 121 still there three years later, is a strong retention signal (Kagi, first three years).
- If you ask people to pay for something they used to get for free, the product has to be noticeably better on day one. Kagi's launch story on Hacker News is about finding a result Google missed (Kagi, first three years).
What didn't work for Kagi, and what are the risks?
- Scale is still small next to Google. 50,000 members is a real business and a tiny share of search. Prelovac has talked about a longer-term aim of 5 million customers; that remains a goal (Kagi, 50K; Kagi, first three years).
- Member-investor rounds are still fundraising. Kagi is not a pure bootstrap. It raised from users and spent a large share of earlier investor funds on 20,000 thank-you T-shirts, which is generous and also a reminder that capital was finite (Kagi, What's next; Kagi, 20,000 members).
- The product surface keeps expanding. Browser, Assistant, Translate, Maps and Mail all have to be staffed. A small team can lose focus if those bets do not stay tied to search quality (Kagi, first three years).
- We did not find a later official member count after June 2025, so the figures above stop at that date (checked 11 November 2026).
Sources
Every link cited above, in order of first use.
- 1.Kagi, What's next (blog.kagi.com)
- 2.Kagi, first three years (blog.kagi.com)
- 3.Kagi, 20,000 members (blog.kagi.com)
- 4.Kagi, 50K (blog.kagi.com)
Related case studies
More in Consumer apps →HEY: tens of thousands of paying customers in year one
Jason Fried and David Heinemeier Hansson launched HEY in June 2020 as a $99-a-year email service with no ads and no IMAP. By the first anniversary they said they had tens of thousands of paying customers.
Neeva: users would pay for ad-free search but wouldn't switch
Two ex-Google executives built a subscription search engine in 2019. Users would pay, but few would switch. Neeva shut its consumer product in 2023.
Perplexity: AI answer engine at a reported $20B valuation, with ARR approaching $200M
Perplexity answers search queries with conversational, AI-written responses instead of a list of links. Three years after founding it had raised about $1.5 billion, was reportedly valued at $20 billion and even offered to buy Chrome.