Google Chrome new profile: what to set up first

There are two very different reasons this search happens. One is deliberate: a second work account has arrived, and the sessions need somewhere to live that is not the personal one. The other is not deliberate at all: Chrome opened one morning with an empty window, a stranger's name on the avatar, and no bookmarks. The creation flow is short in both cases. What matters is what happens in the ten minutes afterwards, because a profile that is not set up properly gets abandoned within a week and leaves a stale folder behind.

Where the option lives and what pressing it does

The avatar circle at the top right of the Chrome window opens a small panel. Below the current identity there is a list of other profiles on the machine and an entry for adding one. Choosing it opens a short setup screen that asks whether to continue without an account or to sign in, then asks for a name and a colour.

Underneath, Chrome creates a new directory in its support folder and starts it empty. The first profile on a machine lives in a folder called Default and every later one gets a sequential name such as Profile 4. That folder holds its own cookie store, its own history, its own password file, and its own extensions directory. None of it is linked to the profile the button was pressed from.

The colour and the name are stored separately from the folder, in a small index file Chrome keeps for the whole browser. Renaming later is easy and changes nothing on disk. This matters when a profile is opened directly by folder name from a script or a shortcut, because the label seen on screen is not the string the command needs.

The sign in question is the one that actually decides things

The setup screen presents signing in as optional, and it genuinely is. A profile with no Google account attached works completely: it holds cookies, saves passwords locally, installs extensions, and keeps its own history. Nothing syncs anywhere and nothing is uploaded.

Signing in adds two effects at once. The first is sync, described by Google as follows:

If you choose to sign in to Chrome with your Google Account in this new profile, you'll be able to save your bookmarks, passwords, and other settings in your account and access them on your other devices where you're signed in. Source: support.google.com

The second is naming. Once an account is signed in, the profile label becomes the account name automatically, overwriting whatever was typed during setup. Anyone who wants a short label such as Work or Client needs to rename the profile after signing in rather than before.

The practical rule is simple. A profile that will be used on one Mac only, for one narrow purpose, is fine without an account. A profile carrying bookmarks and passwords that need to exist on a phone or a second machine has to be signed in, and there is no local way to achieve the same thing.

Nothing comes across, and each item moves differently

The most common surprise is how empty the new profile is. There is no partial copy and no import from the profile it was created from. Everything has to be brought over deliberately, and the method differs by item.

What	Comes across automatically	How to move it
Bookmarks	No	Export to an HTML file from the old profile, import into the new one
Saved passwords	No	Export to CSV from the old profile, import in the password settings
Extensions	No	Reinstall from the store, then configure each again
Open tabs	No	Copy the URLs, or use a reading list
Site logins	No	Sign in again on each site, by design
History	No	Not transferable between profiles

The password export is worth planning rather than improvising. The exported file is plain text and readable by anything, so it should be deleted from the disk as soon as the import has finished. Some managed accounts have the export option turned off by policy, in which case the passwords have to be re-entered by hand or reached through sync on the account itself.

Extensions are the slowest part. Reinstalling takes a few minutes, but the settings inside each one do not travel, and extensions that hold their own login state need signing in again individually.

Three settings to fix before the first day is over

A new profile fails for boring reasons. These three cover most of them.

Launch it directly instead of switching into it

Switching through the avatar menu means opening the wrong profile first, then correcting. Launching the right one from the start removes that step. On a Mac, a specific profile can be opened by folder name from the terminal:

open -na "/Applications/Google Chrome.app" --args --profile-directory="Profile 4"

The string in quotes is the folder name, not the label shown on screen. Wrapping that command in a small application or a shortcut gives each profile its own launcher, which is what makes a multi profile setup survive.

Make the window recognisable at a glance

The colour chosen during setup tints the top of the window, and it is the only visual difference between two profiles. Picking two colours that are far apart, and giving each profile a name that shows in the toolbar, is the difference between knowing which account is on screen and finding out after sending the message.

Decide where links open

Links clicked outside the browser go to whichever Chrome window the operating system considered active. Chrome has no way to know that a calendar invitation belongs to the work account. This is the single most persistent annoyance of a multi profile setup and it has no complete fix inside Chrome, so the practical answer is to keep the profile used for external links in front and open the others deliberately.

Naming it so the label still makes sense in six months

Profile names age badly, and the usual culprits are predictable. Client names change when the contract ends. Company names change after an acquisition. An email address as a label is accurate and completely unreadable in a switcher list, because the useful part is buried after the at sign and the first characters of two addresses often match.

Names that survive describe the role rather than the party: Work, Client, Admin, Testing. They stay true when the underlying account changes, they fit in the narrow space the profile menu allows, and they read at a glance in a list of five.

The same logic applies to the picture. Chrome offers a set of illustrations, and picking visibly different ones matters more than picking attractive ones. Two profiles with similar avatars and similar colours are two profiles that will be confused, and the confusion always surfaces at the worst moment, which is halfway through sending something.

What to do with the profile that was already overloaded

A new profile is often created because an existing one has become a mixture of two working lives. Creating the new one is only half of that job. The old profile still holds both sets of bookmarks, both sets of saved passwords, both sets of extensions, and a history that spans both.

Splitting it afterwards is manual work, and doing it in the right order keeps the mess contained. Move the bookmarks that belong to the new role first, into a single folder, then export that folder and import it on the other side. Sign out of the moved accounts in the old profile rather than leaving both sessions live, because a stale session is exactly what causes a message to go out from the wrong identity. Remove the extensions that only served the moved role. Leave the history alone, since it cannot be split and is rarely worth the effort.

The step people skip is the last one: deciding which of the two profiles keeps the default position. That is the profile that receives every link opened from outside the browser, and leaving the decision unmade means it defaults to whichever window happened to be in front.

When Chrome creates a profile without being asked

A new profile appearing on its own usually has one of three causes.

Signing a managed account into an existing profile is the most frequent. Accounts administered by a company or a school often trigger an offer to keep that account in a separate profile, and accepting it creates one immediately. The new profile is real, is subject to the organisation's policies, and cannot be merged back into the personal one afterwards.

A blank profile named Person 1 with no bookmarks usually means Chrome could not read the existing profile folder rather than that anything was deleted. The old data is generally still on disk. Quitting Chrome fully and reopening it, then checking the profile list before creating anything new, avoids piling a second empty profile on top of the problem.

The third cause is a shortcut or a launcher that carries a --profile-directory value pointing at a folder that does not exist. Chrome creates it rather than refusing, and the result is an empty profile that appears whenever that shortcut is used.

Check that it actually holds before relying on it

A new profile can look correct and still be sharing something it should not. Two checks take a minute and catch most of it.

Open the same service in both profiles side by side and confirm that each shows a different account without either logging the other out. If signing in on one knocks the other out, the two windows belong to the same profile and the separation was never made.

Then check what the extensions can see. Extensions are installed per profile, but a password manager signed into one vault in both profiles will happily offer credentials from either side, which reintroduces the mistake the profile was created to prevent. Restricting each installation to the vault that belongs with that profile is a setting inside the manager, not inside Chrome.

Removing a profile is permanent

Deleting a profile removes the folder and everything in it from the machine, without a trip through the trash. Bookmarks, history, saved passwords, and settings held only in that profile are gone.

The Google account survives, and anything that had already synced is still in the account and returns when the account is signed into a new profile. A profile that was used without signing in has no such copy anywhere, which is the case where deletion is genuinely destructive.

When another profile is not the answer

Adding a profile solves exactly one problem: two live sessions on the same site. It solves nothing about window count, notification origin, or link routing, and each new profile makes those three slightly worse.

If the count is already at three or four and the friction is finding the right window rather than logging in, the unit of separation is the issue. Keeping each web application in a window of its own is a different arrangement of the same machine, set out in Workspaces and in the capability list on the Features page. For readers weighing that against a paid alternative built on the same idea, Compared with Wavebox lays out the differences, and Pricing states what the change costs.

What to change first

Create the profile, then spend ten minutes on the launcher, the colour, and the password import before using it for anything real. If this is the fourth profile rather than the second, stop and count the windows instead, because that is the problem SpaceDeck is shaped around.