Browser extensions that actually save time, and how to spot the ones that don't

There is a particular kind of disappointment that comes from installing a browser extension. You find it, the screenshots look good, you add it, and for two days you use it constantly. By the second week it is a small icon you no longer notice, sitting in a toolbar with eleven others, quietly asking your browser for permission to read every page you visit.

Most extensions fail this way. Not dramatically — they just stop being worth the toolbar space. After a few years of installing, keeping, and mostly uninstalling them, the pattern of which ones survive is fairly consistent, and it is not the pattern the store listings suggest.

The only question that predicts whether you'll keep it

Before installing anything, it is worth answering one question honestly: what specific action does this replace, and how many times a day do I do that action?

That sounds obvious. It is not how most people choose extensions, including me. The usual process is closer to "this looks like it could be useful," which is how you end up with four screenshot tools and a colour picker you have used twice.

The reason the question works is that it forces you to name a real, countable behaviour. "Helps me stay organised" is not a behaviour. "Copies the current tab's title and URL as a markdown link, which I do about fifteen times a day" is a behaviour. The second one you will still be using in six months, because the friction it removes is friction you actually encounter.

A rough threshold: if the action happens fewer than about three times a day, an extension is probably the wrong shape for it. Something you do weekly is better served by a bookmark or a web page you visit when needed. Extensions earn their permanent place in a toolbar through repetition, not through capability.

This is the test TabVault had to pass before it was worth building. Saving a tab and returning to it is not a weekly action for anyone who keeps a lot of tabs open — it is a dozen times a day, which is exactly the frequency that justifies living in the toolbar.

Count the seconds, then multiply

The maths is what makes small tools worth caring about, and it is worth doing properly rather than hand-waving at it.

Say a task takes 20 seconds. That is nothing. But if you do it 12 times a working day, that is 4 minutes a day, roughly 20 minutes a week, about 16 hours a year. Sixteen hours is two working days spent doing something a piece of software could have done instantly.

Now the other side of that ledger, which almost nobody calculates: an extension that adds a 400-millisecond delay to every page load, on 200 page loads a day, costs you 80 seconds a day. That is 6 hours a year — spent waiting, in fragments too small to notice individually.

This is the trade that matters. A well-built extension that saves 20 seconds twelve times a day and costs nothing on pages where it has no work to do is enormously net positive. A heavy extension that saves you a click occasionally while slowing every page is net negative, and it will never feel that way, because the cost is distributed and the benefit is concentrated.

How to check the cost side

You do not need to guess. Every Chromium browser has a task manager — on Windows and Linux it is Shift + Esc, on macOS it is under the Window menu. It lists every extension as its own process, with live memory and CPU columns.

Open it, use your browser normally for a few minutes, and look at what your extensions are doing while you are not asking them to do anything. Most should sit at essentially zero CPU. Any extension using measurable CPU on an idle tab is running code on pages continuously, and you should be able to explain to yourself why it needs to.

Read the permissions like they mean something

Permission requests are the closest thing to a specification an extension listing has. They are also skimmed past by almost everyone, which is a shame, because they are unusually informative.

The one worth pausing on is "Read and change all your data on all websites." It is not automatically a red flag. Plenty of legitimate extensions genuinely need it: an ad blocker, a password manager, a dark-mode tool, anything that modifies pages you visit. The question is whether the described feature could plausibly require it.

Some useful contrasts:

• A tool that reformats text you select and copy needs access to the current tab when you invoke it — not to every page continuously.
• A tool that changes how one specific site behaves should be scoped to that site's domain, not to all sites.
• A tool that works only when you click its toolbar button can usually use the narrow activeTab permission, which grants access to one tab only at the moment you click.

When a simple, single-purpose extension asks for access to everything you browse, forever, the mismatch is the signal. It might be laziness during development rather than malice — using the broad permission is easier than scoping properly — but laziness about permissions is not a quality you want in software with that much reach.

The other thing worth checking: whether the extension has changed hands. Extensions with real user counts get bought, and a well-behaved tool can become something else entirely in a single update that installs automatically. The reviews, sorted by newest, usually tell that story before anything else does.

The categories that consistently earn their place

Across the extensions that survive long-term, most fall into a few groups:

Things that remove a fixed sequence of clicks. Anything where you currently do "open menu, click item, confirm, close" and could do it in one click. The gain is small each time and constant.

Things that put information where you already are. Not a new dashboard to visit, but a fact surfaced on the page you were already looking at.

Things that undo an annoyance the web imposes on you. Autoplay, popups, layout that fights you on mobile. These are pure removal of friction, and they never stop paying.

Things that make one site you use daily behave better. Narrow in scope, scoped in permissions, high in value if it is a site you live in.

And the ones that consistently do not last: dashboards you have to remember to open, anything requiring an account before it will do its job, anything whose main feature is a nicer new-tab page, and anything with a habit-tracking layer bolted onto a simple utility.

A short checklist

Before installing:

1. Name the exact action it replaces and how often you do it. Under three times a day, reconsider.
2. Check whether the permissions match the described feature.
3. Look at the newest reviews, not the top-rated ones, for signs of ownership changes.
4. Prefer things that work locally over things that send pages to a server.

After installing, at the two-week mark:

5. Open the task manager and check what it costs while idle.
6. Ask whether you have used it in the last three days. If not, remove it. It costs something every day whether you use it or not.

That last step is the one people skip, and it is the one that keeps a browser fast. An extension you are not using is not neutral; it is a permanent tax you agreed to pay for a benefit you stopped receiving.

Why this is how CraftlyKit builds

These are the rules the studio's own extensions are held to, and they are easier to check than to promise. TabVault is scoped to the narrowest permission its feature genuinely needs, does its work in the browser, and stays inert on pages where it has nothing to do.

On the permissions question specifically: TabVault stores everything — saved tabs, categories, scroll positions, video timestamps — in your own browser on your own device. There is no account and no sync server, which means there is nothing to log into and nothing held anywhere for someone else to lose. In a category where a great many extensions sync your entire tab history to a server by default, that is the part worth checking before you install anything, mine included.

The projects page tracks what exists and what state it is in, and the about page covers why this studio exists in the first place. If you disagree with any of the above — particularly the three-times-a-day threshold, which is a rule of thumb rather than a law — tell me. It is the sort of thing worth arguing about.