How your accounts get linked: the signals that give you away
Reused usernames, profile links, a shared email, a birth year in a handle. How small public details join up, what someone can realistically do with them, and how to check your own trail.
On this page
Most people who get found online were never hacked. Their accounts simply pointed at each other. A username reused on a second site, a link in a bio, an email address that turns up in two places: each is harmless on its own, and together they form a trail that someone can follow.
This guide explains the signals that link accounts, how they add up, what an observer can realistically do with them, and how to check your own trail. You need no special software for any of it.
The signals that link accounts
The same username
Reusing a handle is the most common link of all. Searching a username in quotes, or on one of the sites that check hundreds of platforms at once, takes seconds. Near-matches work too: nightfox_98 and nightfox1998 look like the same person to most readers.
A shared username is not proof, though. Popular names collide all the time, so a careful observer treats a match as a lead until something else confirms it.
Links you placed yourself
The link in your bio, a “find me elsewhere” list, a personal website, a stream panel that points to your other channels. These are the strongest signals of all, because you made them on purpose. To an observer, a link is not a guess. It is a confirmation.
Mentions in bios and posts
“Also on Instagram as …”, “add me on Discord”, a signature under every forum post. Plain text works like a link, just slightly less obviously.
A shared email address
Email addresses leak in quiet ways: on contact pages, in old data breaches, on people-search sites, and in the author details that Git records with every commit in a public repository. If the same address sits behind two accounts and shows up anywhere public, it joins them. Some apps also let other people find you by your email address or phone number unless you switch that off.
Public details that narrow you down
A birth year in a handle, a town in a bio, a school, a team, a first name. Each one matches thousands of people. Combined, they can match one. A well-known study from 2000 estimated that most people in the United States could be singled out by nothing more than their ZIP code, gender and date of birth.
The same picture
Using one profile picture everywhere is convenient, and reverse image search makes it a reliable link. Photos can also carry hidden details of their own; Before you post a photo covers those.
How signals add up along a path
Think of every signal as one step from one account to the next. A path is a chain of those steps, leading from something you keep separate, such as a gaming account, to something that identifies you, such as your real name.
Two rules make paths easy to reason about:
- Strength multiplies. A path is only as convincing as all of its steps together, so a strong link followed by a weak one makes a weaker path.
- Every extra step costs something. Each hop takes the observer more effort and adds room for doubt, so long chains are weaker than short ones.
Kenveil’s model uses exactly this arithmetic: it multiplies the strength of each link and takes 10% off for every hop after the first. Here is a made-up example.
- Steamnightfox_98Gaming
- Same username× 0.9
- Redditnightfox_98Gaming
- Mentioned in bio× 0.7
- Instagrammkeller.photoPersonal
- Link in bio× 1.0
- Personal websiteMaria KellerReal name
0.9 × 0.7 × 1.0 × 0.9 × 0.9 = 0.51High
| Change | Path strength |
|---|---|
| Nothing changes | 0.51 · High |
| Steam renamed to nightfox_99, a similar handle | 0.28 · Medium, weakened |
| Mention removed from the Reddit bio | Broken |
| Steam renamed to something unrelated | Broken |
Two things stand out. First, breaking one link breaks the whole path, so you rarely need to delete everything. Second, a change can weaken a path without breaking it: a new username that still resembles the old one leaves a fainter trail, not no trail. And when one link sits on several paths, such as a handle you reused in four places, changing it closes all of them at once.
What an observer can realistically do
It helps to picture who might actually look, because it changes what matters.
- A casual search. A classmate, a new colleague or an opponent after a heated match, spending a few minutes with a search engine. They follow obvious links, identical usernames and a shared email, and stop after a step or two.
- A persistent observer. Someone willing to spend an evening reading old posts, comparing pictures, trying variations of a name and following long chains of weak clues.
- A data aggregator. People-search sites and data brokers that combine structured records, such as addresses, listings and linked profiles, at scale.
A casual or persistent observer cannot read your private messages or see accounts you have set to private without breaking the law or a platform’s rules. What they work with is public information that was never meant to be connected. That is the good news: public information is also the part you control.
The goal is not invisibility
You do not need to vanish. You need the links between the parts of your life you want kept apart to be weak or gone, starting with the short, strong ones that a casual search would find.
How to check yourself
Set aside an hour and work through these steps in a private browser window, signed out of everything.
- List your accounts and group them. Real name, gaming, creator, work: whatever the parts of your life are. Decide which groups should never meet.
- Search each username in quotes. Note every result you recognise, and try the obvious variations with numbers and underscores.
- Search your email addresses. See where they appear publicly, and check whether they are part of known breaches with a service such as Have I Been Pwned (opens in a new tab).
- Open each profile as a stranger would. Read the bio, the links, the pinned posts and the display name. Look at who you follow and who tags you.
- Reverse-search your profile pictures. Any account that shows up with the same picture is linked, whether you meant it to be or not.
- Draw the paths. For each account you want kept separate, count the steps to your real name. One or two strong steps is the urgent case.
- Fix the strongest links first, then check again in a few weeks. Old profiles resurface, and search engines take time to catch up.
Where Kenveil helps, and where it stops
Kenveil does the mapping for you, on the accounts you prove are yours.
- Graph starts only from accounts you verify, with a short code in your bio or one click in the browser extension, and from your own account email. There is no field for anyone else. It reads official public sources and maps reused usernames, profile links, bio mentions, shared email addresses and public details. A same-username find stays a possible match until you confirm it.
- Paths lists the chains that connect your groups, strongest first, using the arithmetic above, with up to three fixes for each.
- Simulator and Fixes shows what a fix would do to your score, and which paths it breaks or only weakens, before you change anything. The simulation runs on your PC, and the next scan confirms whether a link is gone or still found.
- Split is where you decide which groups stay apart, so the score follows your choices rather than someone else’s idea of privacy.
Kenveil has limits worth knowing, too. It only sees what public sources and your own verified profiles show, so it can miss things. Its score, from 0 to 100, estimates how linkable your accounts look; it is not a measure of anonymity. And it never changes your accounts for you: every fix is yours to make, and the next scan shows whether it worked.