At some point the favicons get so narrow you cannot tell them apart, and you are using Cmd+Shift+A to search your own open tabs. That is the point at which the tab strip has stopped being navigation and become storage.
Open tabs are a bet, not a memory
Every tab you leave open is a wager that Future You will remember what it was for. The wager is fine at four tabs. At forty it is a certain loss, because the thing you kept the tab open for — the specific paragraph, the one component on the page, the argument you wanted to check — is not stored in the tab. Only the URL is.
This is the whole problem in a sentence: a tab preserves the location and discards the intent, and the intent was the valuable part.
What tabs actually cost
It is worth being precise, because the usual "tabs eat your RAM" framing is mostly outdated and it makes the real cost easier to dismiss.
| Cost | Reality |
|---|---|
| Memory | Less than folklore suggests. Chrome freezes background tabs and discards them under pressure, so a hundred idle tabs may hold a few hundred MB total. Heavy app tabs are the exception and can each hold 300 MB or more. |
| CPU and battery | Real but concentrated. A handful of tabs with timers, video, or a polling websocket account for nearly all of it. The other ninety are asleep. |
| Findability | The actual cost. Past roughly 15 tabs the strip is unreadable and you are searching, which means the tabs are now a worse database than an actual database. |
| Attention | The expensive one. A strip of unfinished intentions is ambient, unresolvable pressure, and it is on screen for eight hours a day. |
The one-session cleanup
This takes 20 to 40 minutes and you do it once. The goal is not zero tabs. It is that every remaining tab is something you are working on right now.
Snapshot everything first
Bookmark all tabs into one dated folder: Cmd+Shift+D on Mac, Ctrl+Shift+D on Windows. Name it "Tabs 2026-08-03". This takes two seconds and makes every decision after it reversible, which is what lets you move fast.
Close the duplicates blind
You have the same three docs pages open four times each. Close every duplicate without reading it. No decision required, and it typically removes a quarter of the strip.
Sort what remains into four piles
Doing now, keeping forever, reading later, and honestly never. Do not deliberate — first instinct, one second per tab. Deliberation is what turns this into a two-hour job you abandon halfway.
Save the keepers with a sentence each
This is the only slow step and the only one that matters. For each "keeping forever" tab, write the one line that explains why it is worth keeping. Not the title — the reason. Then close it.
Give "read later" a deadline
Put them in a single reading folder with the date. Two weeks from now, anything you have not opened gets deleted unread. This is not defeatist; it is the same outcome as leaving them open, just without the tabs.
Close the rest
The snapshot folder from step one has them all. You will not go looking, and that fact is itself the evidence you needed.
Why it comes back, and what stops it
Two weeks later you are at sixty tabs again, because the cleanup fixed the symptom. The cause is that closing a tab is a loss and saving it is work, so the cheapest move in the moment is to leave it open. Any lasting fix has to make saving cheaper than leaving.
- Make saving one action. If saving a page costs a click, a decision and a form, you will leave the tab open instead. A browser extension that captures the title, URL, and icon from the page you are on removes the entire data-entry step.
- Capture the reason, not just the link. Speaking a sentence takes three seconds; typing it takes fifteen. That difference is exactly the margin that decides whether you do it. Inspistor puts a voice note on the save form for this reason.
- Have an obvious destination. "Where does this go?" is the decision that stalls the save. A structure you can answer in two seconds is the difference between saving and deferring.
- Use a hard tab ceiling. Pick a number — 12 is a good one — and when you exceed it, the oldest tab must be saved or closed before a new one opens. Constraints work where intentions do not.
Tab groups are a staging area, not storage
Chrome, Edge, Arc and Safari all have tab groups, and they are genuinely good for what they are for: holding the eight tabs belonging to one task while the task is live. Collapse the group, the strip stays readable, and the context stays together.
Where they fail is as long-term storage. A collapsed group named "research" from three weeks ago is a folder with no note, no search, and no way to tell from the outside whether it holds anything. Use groups for the day, and move anything that outlives the task into a real library.
The rule that actually holds
One line, and it removes almost all of the decision-making:
A tab is open because you are working on it right now. Anything else is either saved with a reason, or closed.
There is no third category. "I might need this" is the category that produced 137 tabs, and it is the one to delete — not because it is wrong, but because it has no exit condition. Saving it with a reason attached is the same bet, made somewhere that will still be legible in six months.
Frequently asked questions
How much memory does a Chrome tab use?
A discarded background tab costs almost nothing. An active one typically runs 50–150 MB, and a heavy app tab (a design tool, a mail client, a docs editor) can hold 300 MB to over 1 GB. Modern Chrome freezes and discards background tabs aggressively, so the real cost of 100 tabs is usually not RAM — it is that the tab strip stops being navigable at around 15.
Is it better to bookmark tabs or use a tab manager extension?
Bookmark them if you want them in six months; use a session manager if you want them back tomorrow. The distinction is duration. Session managers accumulate untitled snapshots that nobody reopens, so anything genuinely worth keeping should graduate into a real library with a note attached.
Why do I feel anxious about closing tabs?
Because an open tab is an unfinished intention, and closing it feels like abandoning it. That feeling is accurate — the fix is to move the intention somewhere that keeps it, so closing the tab settles it rather than losing it.