Why search engines disappear
Running a general-purpose web search engine is one of the more capital-intensive undertakings in software: it requires crawling and re-crawling a meaningful fraction of the public web on an ongoing basis, storing and ranking an index measured in the billions of pages, and doing it fast enough that users don't notice the cost. That combination of infrastructure spend and engineering depth, paired with the strong network effects that come from a single default search engine most people already use, has made general search one of the most consolidated categories on the internet. Once Google's index and ranking quality opened a durable lead in the early 2000s, most competitors that couldn't differentiate on something other than "better general search" eventually ran out of reasons for users, or investors, to keep them alive.
The engines below shut down for different immediate reasons — an acquisition that folded the brand into something else, a business model that never worked, or simply running out of runway — but nearly all of them share that underlying cause.
AltaVista (1995–2013)
Launched by Digital Equipment Corporation in December 1995, AltaVista was for several years the technically strongest search engine on the web: a fast, large-scale index built to showcase DEC's server hardware, at a time when most competitors returned results slowly and covered a smaller share of the web. Ownership changed hands repeatedly through the late 1990s and early 2000s — through Compaq, a dot-com-era spinoff, and eventually Overture Services — before Yahoo acquired Overture, and with it AltaVista, in 2003. Yahoo kept the AltaVista brand running as a secondary search property for a decade before formally shutting it down in July 2013 and folding its remaining traffic into Yahoo Search.
Infoseek (1994–2001)
One of the original 1994-vintage search engines alongside Lycos and WebCrawler, Infoseek built an early reputation for fast, relevant results and was briefly a candidate to become Netscape's default search provider. The Walt Disney Company acquired a controlling stake in 1998 and merged it into the Go.com portal, betting on search as a component of a broader entertainment and media destination rather than as a standalone product. That bet didn't pay off: Go.com was shut down in 2001, and Infoseek's search technology was discontinued along with it.
GoTo.com / Overture (1998–2003)
GoTo.com, founded by Bill Gross in 1998, didn't compete on index size or relevance at all — its innovation was the pay-per-click auction model, where advertisers bid for placement and paid only when a user clicked, which became the financial foundation of nearly every search engine's business model that followed, including Google's own AdWords. GoTo.com renamed itself Overture Services in 2001 and licensed its paid-search technology to Yahoo and, for a time, to Google itself. Yahoo acquired Overture outright in 2003 for roughly $1.63 billion, absorbed its advertising technology into Yahoo Search Marketing, and retired the Overture brand. GoTo.com/Overture is a useful reminder that a search engine can disappear as a consumer brand while its core invention quietly becomes permanent industry infrastructure.
Northern Light (1997–2002)
Northern Light launched in 1997 with a feature few competitors had: "Custom Search Folders" that automatically clustered results into topic groups, plus a paid library of full-text journal and publication content layered alongside free web results. It built a loyal following among researchers and information professionals but never reached mainstream consumer scale. Northern Light discontinued its public consumer search engine in 2002 and pivoted entirely to enterprise search software for corporate clients, a business it continued in a much smaller, business-to-business form.
Cuil (2008–2010)
Cuil launched in 2008 with a founding team of former Google engineers and a specific, quantifiable claim: a larger index than Google's. Index size alone turned out not to be the thing that mattered to users — early reviews and hands-on testing found Cuil's relevance and result quality well behind Google's, and its distinctive magazine-style results layout didn't offset that gap. Traffic never recovered from the negative early reviews, and Cuil shut down in 2010, less than two years after launch. It's frequently cited in the industry as a case study in why raw index size is a poor substitute for ranking quality.
Ask Jeeves / Ask.com (1996–2026)
Ask Jeeves launched in 1996 from Berkeley, California, built around a genuinely novel idea for its time: letting users type full natural-language questions rather than keywords, answered by a butler mascot pulling from a curated knowledge base. It went public in 1999 at the height of dot-com enthusiasm, acquired the search technology firm Teoma in 2001 to strengthen its underlying ranking, and was acquired by IAC in 2005, which rebranded it as Ask.com. By 2010, CEO Barry Diller had publicly acknowledged that Ask.com could not compete with Google on general web search, and the company shut down its own web crawler, outsourcing its underlying search results to other providers while repositioning Ask.com around Q&A content. That arrangement persisted for roughly a decade and a half before IAC discontinued the Ask.com search business entirely, shutting it down in 2026 — nearly thirty years after it launched. IAC's stated reasoning was straightforward: as the company "continues to sharpen its focus," it made the decision to discontinue the search business rather than continue subsidizing it. Ask Jeeves' real legacy isn't its market share, which was always modest next to Google's; it's the question-and-answer interaction model it pioneered, which is now central to how AI-native search and answer engines work.
What their disappearance teaches us
Read as a set rather than individually, these shutdowns follow a small number of repeating patterns: a parent company decided search wasn't core to its business and folded the technology into something else (Infoseek, GoTo.com/Overture, AltaVista); a differentiator that sounded compelling on paper didn't translate into better results users could feel (Cuil's larger index, Northern Light's folders); or a business simply ran for decades on a shrinking niche until its owner concluded the niche was no longer worth subsidizing (Ask.com). None of them lost simply for lack of trying, and several — GoTo.com's pay-per-click auction most of all — left a permanent mark on how every surviving search engine makes money today, even after the original brand was gone.