All search engines look alike: a box, a question, a list of results. Behind them, though, there are very different ways of building those results, paying for the service and handling your questions. Some build their own catalogue of the web, others use a larger search engine’s, others still combine results from several services. Some choose adverts based on your current search, others based on what they know about you.
This post describes the types of search engines and the technologies that go with them: the index, intermediary search engines, metasearch engines, contextual and behavioural advertising, automatic deletion, private browsing and its limits, and browser opt-out signals. It names no service and makes no ranking: it simply helps you recognise what lies behind the box.
It is the technical side of the recommendation to choose your search engine knowingly. The practical sequence for changing your default on every device is in the guide on how to change your default search engine.
How to read this list
The technologies are split into four groups, like the stages of security:
- prevention: they reduce what is kept or linked to you from the start;
- detection: they show you what a search engine keeps and what it knows about you;
- response: they let you remove what you do not want to keep;
- governance: they shape choice and rights, between people, services and laws.
For each technology you will find the risk it reduces, strengths, limits, an example and complexity. One principle applies to all of them: every search engine has to see your question to answer it; the difference lies in what it does afterwards.
1. Prevention technologies
Search engines with their own index
The index is the huge catalogue of web pages a search engine collects and organises so that it can answer you. Some search engines build their own, visiting the web with automated programs called crawlers.
- Risk reduced: dependence on a single large results supplier, and your searches being passed to third parties to obtain results.
- Strengths: the search engine controls both the results and the rules on data, without having to pass questions on to anyone else.
- Limits: building an index is expensive; smaller search engines may cover fewer pages, especially in some languages.
- Example of use: you want a search engine that does not depend on another service’s results.
- Complexity: low for the user.
Intermediary search engines
These are search engines that obtain most of their results from a larger search engine, under an agreement, and present them under their own data rules. Some say they remove the IP address or do not link searches to the person making them before passing the question on to the partner.
- Risk reduced: your searches being linked to your account or connection at the larger search engine.
- Strengths: results similar to the ones you know, with different privacy rules.
- Limits: the quality of results depends on the partner; some information, such as the question itself and, in some cases, the IP address, may still reach it. Read what the privacy policy says about what is shared.
- Example of use: you want familiar results, but not your searches in your account.
- Complexity: low.
Metasearch engines
A metasearch engine sends your question to several search engines and combines the results into a single list. Some are public services; others are programs that a person or an organisation can install on their own server.
- Risk reduced: all your searches being concentrated at a single service.
- Strengths: results from several sources; in the self-installed version, you decide what is kept.
- Limits: speed and quality vary; a public metasearch engine should be chosen with the same care as an ordinary one, because it receives your questions.
- Example of use: an association or an office that wants to offer its members search without an account.
- Complexity: low for those using a public service, high for those installing one.
Contextual advertising
Many search engines pay their way with adverts. A contextual advert is chosen on the basis of what you are searching for at that moment; a behavioural advert is chosen on the basis of a profile built on what you have done before.
- Risk reduced: an advertising profile being built from your searches.
- Strengths: adverts remain, but they are about the question, not about your past.
- Limits: advertising does not disappear; a contextual advert can still be misleading, like those that impersonate a brand.
- Example of use: you search for “running shoes” and see adverts for running shoes, but you do not find them on other sites for days.
- Complexity: none for you: it depends on the search engine you choose.
Search without an account and without IP retention
Some search engines say they do not require an account, do not keep your IP address, the number that identifies your connection, or make it anonymous after a few days.
- Risk reduced: your searches being linked to your identity and your connection.
- Strengths: the profile left behind is much smaller, or absent.
- Limits: it is a promise made by the service, to be read carefully and with its date; it does not hide your connection from your network provider.
- Example of use: everyday searches, the ones you do not want to see linked to your name.
- Complexity: low.
Approximate location
“Near me” searches work better if the search engine knows roughly where you are. Some services use only an approximate area or the country; your phone itself can give an app only an approximate location.
- Risk reduced: searches carrying your precise whereabouts.
- Strengths: you decide how much location to share.
- Limits: local results may be less precise.
- Example of use: you use the privacy-minded search engine for everything, and one with more precise location, on purpose, to find a shop nearby.
- Complexity: low.
2. Detection technologies
The privacy policy, with its date
This is the page where the search engine explains what it keeps, for how long and with whom it shares it.
- Risk reduced: using a service without knowing what it does with your questions.
- Strengths: it is the most direct source; specific statements (“we do not store IP addresses”) can be compared.
- Limits: it is long and changes over time; general reassurances tell you nothing. The Electronic Frontier Foundation suggests preferring statements that a service cannot access data to promises that it will not.
- Example of use: you read two privacy policies with three questions: what does it keep, what does it link to you, what limits does it commit to.
- Complexity: low.
Your account’s history page
If you use a search engine with an account, there is usually a page showing your saved searches, with dates and sometimes place and device.
- Risk reduced: not knowing what is left of your searches.
- Strengths: you see the archive with your own eyes and can act straight away.
- Limits: it shows what is linked to the account; some services say they also collect data not linked to the account, which does not appear there.
- Example of use: once a year you scroll through a month of searches and decide what to keep.
- Complexity: low.
Ad settings
Some accounts show the interests the service assigns to you to choose adverts, and let you correct them or limit personalised adverts.
- Risk reduced: an advertising profile you know nothing about.
- Strengths: it shows you the profile, at least in part.
- Limits: it does not always show everything that is used; limiting personalisation does not remove advertising.
- Example of use: you remove an interest that does not describe you, or that you do not want associated with you.
- Complexity: low.
3. Response technologies
Manual deletion, automatic deletion and pause
From the history page you can delete individual searches or a period, turn on automatic deletion of searches older than a time you choose, or pause your history.
- Risk reduced: an archive that grows without limit.
- Strengths: you keep what is useful to you for as long as you need it.
- Limits: pausing does not usually delete what is already saved; some services say they keep copies in backup systems for a limited time.
- Example of use: you keep the last few months of searches as reminders and let the older ones delete themselves.
- Complexity: low.
Private browsing and its limits
This is a browser mode that, when you close it, deletes the history and cookies of that session from your device.
- Risk reduced: searches visible to the next person who uses the device.
- Strengths: useful on a shared computer or for a search you do not want to find again among the suggestions.
- Limits: the US Federal Trade Commission points out that it does not stop websites from seeing your activity; if you sign in to an account, the service knows it is you.
- Example of use: you search for a present on the family computer.
- Complexity: low.
Resetting browser settings
This is the function that returns the browser to its initial state, usually turning off extensions and restoring the search engine and home page.
- Risk reduced: a search engine or home page changed by unwanted software.
- Strengths: you start again from a clean base, without reinstalling everything.
- Limits: if the cause is a program installed on the computer, it needs removing first, or the problem comes back.
- Example of use: your search engine keeps changing after you installed a free program.
- Complexity: medium.
4. Governance technologies
Browser opt-out signals
Some browsers and some extensions can send sites an automatic signal asking them not to sell or share your data.
- Risk reduced: your data being passed to third parties for advertising.
- Strengths: you set it once and it applies to every site that respects it.
- Limits: its value depends on local law; in California, for example, the state’s privacy authority says businesses must honour it as a request to opt out of sale and sharing. Elsewhere many sites may ignore it.
- Example of use: you turn it on in your browser settings, alongside your choice of search engine.
- Complexity: low.
Choice screens
In the European Union, the Digital Markets Act has required large platform operators, since 7 March 2024, to let users easily choose an alternative default service, including the search engine. In the UK, the competition authority (CMA) began work on similar measures in 2025 and consulted on them in 2026: as of November 2026 no decision had been taken.
- Risk reduced: a default decided by someone else.
- Strengths: the choice is offered to you at the right moment, during set-up.
- Limits: it appears only on some devices and in some countries; choosing in a hurry makes it pointless.
- Example of use: you set up a new phone and spend a minute on the search engine screen.
- Complexity: low.
Your data rights
In the European Union the GDPR, and in the UK its British version, provide for the right of access (Article 15), the right to erasure in the cases set out (Article 17) and the right to object (Article 21), which cannot be refused for direct marketing.
- Risk reduced: data being kept or used without your knowledge.
- Strengths: you can ask what a service knows about you, and object to profiling for marketing.
- Limits: it takes a formal request and a little patience; for a specific case it is worth contacting the ICO or a professional adviser.
- Example of use: you ask a service what search data it keeps on your account.
- Complexity: medium.
Comparison table
| Technology | Group | Risk reduced | Complexity | When to use it |
|---|---|---|---|---|
| Own index | Prevention | Dependence on a single supplier | Low | If you want an independent search engine |
| Intermediary search engine | Prevention | Searches linked at the large search engine | Low | If you want familiar results under other rules |
| Metasearch engine | Prevention | Concentration at a single service | Low / High | For searches across several sources |
| Contextual advertising | Prevention | Advertising profile | None | Always, if the search engine offers it |
| No account, no IP | Prevention | Searches linked to you | Low | For everyday searches |
| Approximate location | Prevention | Precise location in searches | Low | Except for deliberate local searches |
| Dated privacy policy | Detection | Using a service without knowing | Low | Before choosing, and once a year |
| Account history | Detection | Unknown archive | Low | Once a year |
| Ad settings | Detection | Unknown advertising profile | Low | Once a year |
| Automatic deletion and pause | Response | Archive without limits | Low | If the search engine keeps searches |
| Private browsing | Response | Searches visible on the device | Low | On shared devices |
| Browser reset | Response | Search engine changed by software | Medium | If the search engine changes by itself |
| Opt-out signals | Governance | Sale and sharing of data | Low | Turn them on once |
| Choice screens | Governance | Default decided by others | Low | During set-up |
| GDPR rights | Governance | Data used without your knowledge | Medium | When you want to know or object |
How to choose
- If you mostly search for general information: a search engine without an account that says it does not keep your IP address, with contextual advertising, is often enough.
- If you do a lot of local searching: consider a privacy-minded search engine for everything and one with more precise location, on purpose, for shops and addresses.
- If you want to stay with your usual search engine: turn on automatic deletion, check your advertising interests and read the privacy policy once a year.
- If you share a device with your family: separate profiles, or searching without an account.
- If you use a work device: follow your organisation’s choices and ask your IT team which search engine is intended.
The rule of thumb: choose the search engine that best answers the three questions, what does it keep, what does it link to you, what limits does it commit to. Everything else, from the index to the adverts, helps you understand the answers.
How this connects to the Cyber Welfare Framework
| Pillar | What this content contributes |
|---|---|
| Skills | Knowing the types of search engine and the technologies that decide what is left of your searches, and being able to recognise them |
| Awareness | Understanding that rules on data and the source of results are two different things, and that private browsing has clear limits |
| Secure Behaviour | Choosing the search engine and settings that suit the way you search, and reviewing them when they change |
Reference level: FL3 — Autonomous. This is the level at which you choose your search engine with clear criteria, knowing what lies behind the search box, rather than accepting the default. To measure your other digital habits, take the digital resilience self-assessment.
Conclusion
Search engines look alike on the surface and differ underneath: in the index they use, in how they pay their way, in what they do with your questions. No technology solves everything on its own, and some, like private browsing, protect less than they seem to.
Knowing them serves one purpose: reading a search engine’s promise and working out whether it convinces you. The rest is a choice, and you can make it again whenever you like.
Related content
- Choose your search engine — the recommendation this post builds on
- How to change your default search engine — how to use these technologies, in the right order
- Search history privacy risks — what these technologies protect
- A browser that limits tracking by default — the neighbouring topic of the browser, which protects what happens on every site
Related resources
Short reads from the Resources section, for anyone who wants to focus on a single aspect:
- Alternative Search Engines: What Changes for Your Privacy
- Privacy Focused Browser: The Setting You Use All Day
- AI Search Privacy: What Changes When You Ask Instead of Search
Start with the first step: the Cyber Welfare Programme guides you free of charge, one recommendation at a time.




