CYBER WELFARE

Protect your Digital Privacy

Search history privacy risks: what it means when a search engine remembers and links

When people talk about online privacy, the focus is nearly always on what you publish: photos, posts, comments. Searches feel different. Nobody sees them, you type them in a few seconds, you often forget them straight away. Yet they are some of the most personal information you produce: you write them before you have talked to anyone, in the words you would use with yourself.

Every question you ask a search engine can stay somewhere. Some search engines keep searches in your account, together with the result you tapped, the time and an estimate of where you were. Others say they do not keep them, or do not link them to the person who made them. This is what we mean by search history: an archive of questions that, once linked to you, tells the story of your interests, your plans and your worries.

So the useful question is not only “is this search harmless?”, but: what does it mean that my searches are kept and linked to me, for months or for years?

This post explains search history privacy risks by looking at the three things against which the security of any information is measured: that it stays private, that it stays accurate, that it stays available. They are the everyday version of three technical words, confidentiality, integrity and availability, and they help show that the risks of a long search history are not of a single kind.

It takes a closer look at the recommendation to choose your search engine knowingly. How to review everything your accounts keep, from email to maps, is the subject of the recommendation on reviewing what your accounts keep.

Three questions to measure an impact

Faced with your search history, these are the right questions:

  • What does this archive say about me, and to whom? — this is the confidentiality question.
  • Is what I see when I search the full picture, or a picture chosen for me? — this is the integrity question.
  • What would I lose, or no longer be able to do, if the account it is kept in were locked or broken into? — this is the availability question.

Applied to a search engine that keeps nothing, they produce modest answers. Applied to ten years of searches saved in the same account, the answer changes a great deal. The rule of thumb is this: a search weighs little if no one keeps it, and a lot if it stays linked to you for years.

1. Confidentiality: your questions stay yours

Confidentiality is the guarantee that information can be read only by those who are authorised to read it. A search history linked to your account is information about you, and often very precise information: it says what you are curious about, what worries you, what you are planning. The longer it is, the more it tells, to whoever keeps it and to anyone who manages to read it.

Some searches touch on subjects the law treats as especially sensitive, which the General Data Protection Regulation (GDPR) protects with stricter rules. There is no need for examples: it is enough to know that a history can contain this too, and that a search engine that does not keep it cannot lose it.

A practical example

Sarah’s history holds a year of searches: a photography course, gym opening times, a trip to Portugal, the price of a used car, then, in a single week in March, “how to ask for a pay rise”, “permanent contract rights”, “graphic design jobs Bologna”. None of these searches is secret. Put in a row, they say that in March Sarah was thinking of changing job and city.

Incident scenario

Sarah signs in to her account on a colleague’s computer to print a ticket, and forgets to sign out. The next day, when the colleague opens the browser, the search box offers suggestions based on Sarah’s recent searches. He means no harm, but he now knows Sarah is looking at jobs in another city. The US Federal Trade Commission points out that companies may track people in order to “remember the things you searched for online”.

Warning signs

  • the search box suggests searches you made weeks or months ago;
  • you use the same account on several devices, including shared ones;
  • you do not know whether your search engine keeps your searches, or for how long;
  • you have never opened your account’s history page.

Recommended behaviour

Find out what your search engine keeps by reading its privacy page. If it keeps a history, decide for yourself how long: many services offer a pause or automatic deletion after a set period. Sign out of your account on devices that are not yours.

2. Integrity: is what you see the full picture?

Integrity is the guarantee that information is accurate and complete. Here the question is about results: when a search engine uses your history to choose what to show you, is what you see the most useful picture, or the picture that matches the idea the search engine has formed of you?

Personalisation is not a problem in itself: a result near your home or in your language is often exactly what you need. Some search engines also point out that not all results are personalised, and that people searching for the same thing may get the same or similar results. The risk arises when personalisation is invisible: you do not know what is driving it, and you cannot compare.

A practical example

Sarah searches for “best car insurance”. The search engine, which knows she has visited a particular insurer’s site many times over the past few weeks, shows her adverts and suggestions linked to that insurer at the top. Nothing is false, but the picture is tilted towards what Sarah had already looked at.

Incident scenario

After searching several times for information about a car loan, Sarah starts seeing ever more insistent finance offers in her results and adverts, some from brokers she has never heard of. One evening she opens one that looks like a bank’s official page. Before entering her details, she notices that the web address has one letter different. In an alert dated 21 December 2022, the US FBI reported exactly this: search engine adverts that impersonate well-known brands and lead to fake sites almost identical to the real ones.

Warning signs

  • you keep seeing the same adverts, linked to searches from days ago;
  • the suggestions in the search box “guess” too well what you are about to type;
  • the top results are often adverts that are hard to tell apart from ordinary results;
  • searching for the same thing on another device, or signed out, gives different results.

Recommended behaviour

Tell adverts apart from results: they are usually marked with a word such as “Sponsored” or “Ad”. For important sites such as your bank, email or public services, type the address you know instead of searching for it. If your search engine allows it, check the personalisation settings and decide for yourself whether to keep them. The signs to watch most closely are collected in the post on signs your searches are being profiled.

3. Availability: an archive that depends on a single account

Availability is the guarantee that you can use your data and your services when you need them. For search history the point is less obvious but real: when searches, email, maps and documents all sit in the same account, their availability depends on that account. And a long archive makes every problem with the account more serious.

A practical example

Sarah uses her history as a reminder: it is where she finds a shop’s address, the title of a book, a recipe. It is handy. But it means that if the account becomes inaccessible one day, she loses this too, along with the email and documents linked to it.

Incident scenario

Someone gets into Sarah’s account with a password she had reused from an old service. Before she notices, they read her email, look through years of searches and change the recovery number. Sarah is locked out of her own account for days. Meanwhile, whoever got in has been able to piece together from her searches where she works, what she was planning and which services she uses. The longer the history, the more material they found.

Warning signs

  • your search engine account is the same one you use for email, documents and your phone;
  • the account has no two-factor authentication;
  • the history has been on for years, with no automatic deletion;
  • you do not know which devices the account is signed in on.

Recommended behaviour

Protect the account with a unique password and a second factor. Keep your history for as long as it is useful to you, not for ever. If you use it as a reminder, save what you really want to keep separately, for instance as bookmarks in your browser.

One scenario that brings them together

During a difficult period at work, Sarah searches a lot: jobs in other cities, employment rights, rents, language schools. The searches stay in her account, linked to her name. A few weeks later, someone who has had access to her laptop sees the suggestions in the search box and works out her plans (confidentiality). Meanwhile, Sarah receives ever more targeted adverts for removal firms and loans, some from dubious sites (integrity). And when, months later, her account is broken into, whoever gets in finds a tidy account of her life, while she loses access to her email and documents for days (availability).

None of the three effects is serious on its own. But they all start from the same cause: searches kept for a long time and linked to a single identity, without anyone really having decided it.

The impacts that show up later

Some effects do not show up straight away:

  • the profile that builds up — every search adds a detail; after years the profile describes habits you have forgotten yourself;
  • searches that outlive who you were — a question asked during a particular period stays there even when your life has changed;
  • data that changes hands — a service may be sold, change its rules, or share data with partners that supply results or adverts;
  • anonymity that does not hold — a list of searches “without a name” can be enough to recognise a person, because questions contain places, surnames and habits. The 2008 opinion of Europe’s data protection authorities already asked for anonymisation to be done in a genuinely effective way, precisely for this reason.

Not every search weighs the same

The table describes the typical weight of each kind of search, when it is kept and linked to you. It is not a rigid rule: it depends on who you are and what you search for.

Kind of searchWeightWhy
News, weather, sports resultsLowSays little about you, and everyone searches for it
Recipes, hobbies, filmsLowDescribes tastes, rarely anything private
Shopping and pricesMediumReveals budget, plans and stages of life
“Near me” places and addressesMedium-highTogether with location, tells where you live and where you go
Work, contracts, legal questionsHighReveals plans you often have not yet shared
Names of people you knowHighLinks to you people who have chosen nothing
Very personal or sensitive subjectsVery highThis is the information the law protects with stricter rules

Why even a “careful” search engine is not enough on its own

Choosing a search engine that keeps little removes much of the risk, but not all of it, because your searches also pass through other places.

SituationWhy it still matters
Careful search engine, but signed in to another serviceSearches made from that account may end up in another history
Careful search engine on the computer, not on the phoneEach device has its own default
Careful search engine, but your connection is visibleYour network provider and the sites you open can still see your connection
Careful search engine, but precise location turned onYour phone’s location may travel with local searches
Private browsingClears the session on your device, but does not hide searches from the search engine

The rule of thumb stays the same: know who receives your questions and what they do with them. The differences between kinds of search engine are described in the post on types of search engines.

How this connects to the Cyber Welfare Framework

PillarWhat this content contributes
AwarenessRecognising that searches, kept and linked to you, describe interests, plans and worries
SkillsKnowing how to read confidentiality, integrity and availability as three questions to ask when looking at your own history
Secure BehaviourChoosing a search engine knowing what it keeps, and keeping your history only for as long as you need it

Reference level: FL2 — Beginner. This is the level at which you move from “I just search” to “I understand what is left of my searches and I decide”. Awareness of the impacts is what keeps the choice steady over time. To see where you stand on your other digital habits, take the digital resilience self-assessment.

Summary

  • Confidentiality is about what your history reveals: interests, plans, worries, and people you have searched for.
  • Integrity is about what you see: results and adverts chosen on the basis of your profile, sometimes including sites that imitate real ones.
  • Availability is about the account where everything is kept: if it is breached or locked, you lose access and someone else finds your archive.

A long history linked to you does not produce a single impact: it opens the door to all three, especially if the same account is used for everything.

One thing to do today. Open the privacy page of the search engine you use and look for a specific sentence on what it keeps and for how long. If you cannot find one, that is information too. Five minutes, and you will know who remembers your questions.

Related content

Related resources

Short reads from the Resources section, for anyone who wants to focus on a single aspect:

Start with the first step: the Cyber Welfare Programme guides you free of charge, one recommendation at a time.