A Review Of The Best Search Engines — Based On Speed & Relevancy

Based on my research, these are the best search engines currently. I’ve rated each search engine based on relevancy and usability.

AI-Based Search Engine / Answer Generator

Perplexity is a compelling search engine that uses AI. Sometimes when you need quick and basic information a search engine like Kagi is faster (see below). Overall, I give Perplexity a score of 9/10.

Subscription-Based Search Engines

  • Kagi (USA) — 8/10 (fast, very good relevancy and organization of results). Sometimes weak when doing simplistic searches, for example, “Franklin Templton Twitter” doesn’t bring up the Franklin Templton Twitter account page.

Independent Search Engines With No Ads

  • Mojeek (UK) — 7/10
  • Gigablast (USA) — 4/10 (slow and search results lack relevancy)

Meta-Search Engines with No Ads

  • SearXNG (Worldwide) — 8/10 (good relevancy — but search results are presented in a basic manner)

Ad-Supported Search Engines

  • DuckDuckGo (USA) — 6/10 (a privacy-oriented search, but results are somewhat cluttered and contain ads)
  • Qwant (France) — 7/10 (good privacy and overall relevancy)
  • Brave (USA) — 7/10 (a cluttered interface obscures good quality results)
  • MetaGer (Germany) — 7/10 (good free meta-search, sometimes defaults to German)
  • Google (USA) — 6/10 (good relevancy but ruined by commercial interests, tracking & clutter)
  • Bing (USA) — 5/10 (good relevancy but has tracking and interface issues)
  • Yandex (Russia) — 6/10 (poor relevancy)
  • SwissCows (Switzerland) — 3/10 (often returns server errors, decent results)

Details On The Search Engines

Kagi

  • Search Engine Type: Subscription-Based
  • Created in: June 2022
  • Headquarters: Palo Alto, CA, USA
  • Search results source: Google, Bing, Wikipedia, DeepL, Teclis, TinyGem, and an AI for instant answers

Kagi results are relevant and concise. I like the way the search results are presented in an elegant, structured manner.

Kagi’s search results include anonymous requests to search indexes like Google and Bing and vertical sources like Wikipedia, DeepL, and other APIs.

Kagi also has its own non-commercial index (Teclis) and news index (TinyGem). Teclis and TinyGem are created using a web crawler to focus primarily on non-commercial content.

SearXNG (fork of Searx)

  • Created in: 2015 (as Searx)
  • Headquarters: Decentralized — various locations worldwide
  • Search results source: User-defined selection from 70 different sources

SearXNG (GitHub page) is an ad-free metasearch engine that can combine results from seventy different search services.

SearXNG offers ad-free Google results, which are fresh, not cached (like the ones found in StartPage). In addition, you can add Bing, Qwant, and Mojeek results into the mix. Overall, this makes the search results very relevant and diverse.

SearXNG has good “Instant Answers” but a weak Image Search.

SearXNG development started in the middle of 2021 as a fork of the Searx project (GitHub page). Users are neither tracked nor profiled. Additionally, SearXNG can be used over Tor for online anonymity.

SearXNG has a better user interface, comes with light and dark versions, and works well on mobile browsers. An administrator can block and/or replace the URLs in the search results. Overall, SearXNG is simple to maintain as an instance administrator.

Regarding privacy, be advised that unknown organizations and individuals run these interfaces.

Of course, you can also run your own instance of SearXNG. Here is an in-depth test of various search engines, including SearXNG.

SearXNG supports the external bangs from DuckDuckGo. To directly jump to an external search page, use the !! prefix.

To set category and/or engine names, use a ! prefix. To give a few examples:

  • Search in Wikipedia for Paris
    • !wp paris
    • !wikipedia paris

DuckDuckGo

  • Created in: 2008
  • Headquarters: Paoli, Pennsylvania, USA
  • Search results source: Bing, Yahoo, Yandex, Wikipedia, DuckDuckGo Bot, and other sources

DuckDuckGo is a privacy-friendly public search engine.

It is the most popular alternative search engine in the United States and worldwide. DuckDuckGo states that it obtains its results from more than 400 sources, but it primarily relies on Bing.

DuckDuckGo uses many sources for its “Instant Answers”.

The search engine licenses map data from Apple Maps and weather from Dark Sky. In both cases, DuckDuckGo draws the information in a privacy-friendly way and does not share information with Apple.

DuckDuckGo generates revenue from ads, but they are not personalized – instead, they are based solely on search queries.

The search engine stores cookies to save settings changes, but this feature is optional. Alternatively, users can use URL parameters to change settings without using cookies.

DuckDuckGo has been recently criticized for downranking state media sites in Russia.

Qwant

  • Created in: 2013
  • Headquarters: France
  • Search results source: Bing results and Qwant‘s index of 20 billion pages.

Qwant is an alternative search engine based in France. The French Government uses Qwant it for web searches.

Qwant states that it has its own web crawlers and is working to build its own index. In the interim, it partners with Bing to buttress its indexing capabilities and provide useful results. It does not share information with Microsoft.

Qwant serves map results from Open Street Maps. Qwant has a weak video search.

Like DuckDuckGo and Startpage, Qwant generates revenue from non-personalized ads that appear as a result of specific search queries.

Qwant has a solid privacy policy.

Yandex

Yandex is a Russian search engine that is popular in Eastern Europe. Many of the search results are likewise sourced from Eastern Europe.

  • Created in: 1997
  • Headquarters: Moscow, Russia
  • Search results source: Yandex’s independent search index

Yandex has a particularly good Image Search, but its Video Search is weak.

Mojeek

  • Created in: 2004
  • Headquarters: UK
  • Search results source: Mojeek’s independent index of 5 billion pages.

Mojeek is based in the United Kingdom. The search results from Mojeek’s index of 5 billion pages. However, Mojeek’s image results are still served by Bing and Pixabay. Mojeek’s news results seem to draw almost exclusively from UK sources.

Mojeek has a good privacy policy.

Gigablast

  • Created in: 2000
  • Headquarters: USA
  • Search results source: Gigablast’s independent index.

Gigablast is based in the United States. Along with Mojeek, it is the only alternative search engine that relies exclusively on its own index. It contains no ads or third-party trackers.

Gigablast’s search speed is slow, and it presents search results in a 90s retro style.

Gigablast cannot index Linkedin, Facebook, and YouTube because it’s blocked from doing so.

The search engine has a short privacy policy page. It states that no third parties are given access to user information and that query logs are deleted regularly. Gigablast also stated in 2013 that it does not share user IP addresses with third parties.

Gigablast is open-source, making it the only open-source search engine that does its own indexing.

SwissCows

  • Created in: 2014
  • Headquarters: Switzerland
  • Search Results Source: Bing

SwissCows is a privacy-friendly search engine. SwissCows states that it is also a “family-friendly” search engine which means it doesn’t index pornography or other sexual content.

SwissCows often respond to searches with error messages, such as when searching for “latex pillow” (give it a try).

SwissCows has a strong privacy policy, and the company asserts that it does not create personal search profiles and anonymizes all search queries after seven days.

SwissCows generates revenue through non-personalized ads based on individual search queries.

Brave Search

  • Created in: March 2021
  • Headquarters: San Francisco, CA, USA
  • Search results source: Bing with some ranking adjustments from Brave

Brave Search is a search engine created by the same company that makes the Brave Browser.

The search results from Brave are a mixture of Bing results with other results generated by Brave’s index.

The company says they anonymously check Brave’s search results against third-party results and mix them on the results page. Additionally, an option exists to include Google results anonymously. 

For the sake of transparency, Brave provides a “Results Independence” metric. This anonymous calculation shows the percentage of search results from Brave versus these third parties.

Additionally, Brave relies on Bing for most image and video results. However, even when using these external results, the Brave search engine still preserves user privacy, according to the Frequently Asked Questions page.

StartPage

  • Created in: 2006
  • Headquarters: The Hague, Netherlands
  • Search results source: Google

StartPage is majority-owned by System1, a U.S.-based advertising company. This has been the subject of concern, but Startpage’s privacy policy still seems sound.

Like DuckDuckGo, Startpage generates revenue through non-personalized ads.

Startpage is unique among alternative search engines because it licenses its results from Google. Most smaller search engines use Bing results.

Unfortunately, StartPage often produces Captchas for VPN users.

Metager

  • Created in: 1996
  • Headquarters: Hannover, Germany
  • Search results source: Bing, Google, and Yahoo

MetaGer is a non-profit search engine run by SUMA e.V. in Germany. MetaGer states that it draws results from 50 sources, but Bing strongly influences its results. MetaGer has a strong privacy policy.

However, MetaGer does put several ads above the organic search results. The organization has stated that it serves non-personalized ads to cover costs, but it hopes to earn enough donations to drop the ads.

MetaGer is one of the three open-source search engines on the list, along with Gigablast and SearXNG. The company is proud of its ranking process designed to avoid censorship.

MetaGer notes where specific search results come from. It also includes an internal blacklist and allows certain sources to be toggled on and off.

Google

  • Created in: 1997
  • Headquarters: Mountain View, CA, USA
  • Search results source: Google’s index of 60 billion pages

Google is the dominant search engine around the world. It has a huge index, covering somewhere between 60 billion or perhaps hundreds of billions of web pages.

Google’s search results are ad-filled and cluttered.

Currently, Google accounts for 92% of all search queries. This is the search engine market share worldwide:

  • Google: 92%
  • Bing: 3.33%
  • Yahoo: 1.34%
  • Yandex: 0.97%
  • DuckDuckGo: 0.71%

Bing

  • Created in: 2009
  • Headquarters: Seattle, Washington, USA
  • Search results source: Bing’s index of 15 billion pages.

Bing is a search engine with a large index of at least 15 billion pages. Many other search engines use Bing to provide their results (including DuckDuckGo and Qwant).

Like Google, Bing has an ad-filled interface cluttered with videos and other distractions.

4 thoughts on “A Review Of The Best Search Engines — Based On Speed & Relevancy”

  1. Very interesting. I noticed some of your comments are jumbled under the wrong heading. Would be worth cleaning up as this is very valuable content.

Leave a Comment

Your email address will not be published. Required fields are marked *