Keeping research tabs without keeping the tabs

If you have ever researched something properly — a house purchase, a technical decision, a holiday, a piece of legislation — you know the state. Thirty tabs, all relevant, none finished, and a genuine fear that closing any of them loses something you will need.

The fear is not irrational. Some of those pages took twenty minutes of searching to find and you could not reconstruct the query now. But keeping all thirty open is not solving that; it is deferring it, and the deferral has a cost you pay every day in a browser window you can no longer read.

Three kinds of research tab

Sorting them is most of the work, and there are only three.

Findings. A price, a spec, a fact, a paragraph you will quote. What you need is the information, not the page. Once it is written down somewhere, the tab is finished.

Sources. You will need to go back and read more, or check something, or link to it. Genuine reference. The page itself matters.

Pending. Something happens here later. A viewing on Saturday, an application window that opens on the 1st, a decision you have to make by the 15th, a reply you are waiting for.

Findings and sources have decent answers already. Pending is the one that keeps the window full, and it is the one nobody has a habit for.

Findings go in the document, not the browser

The move that shrank my windows most: keep one document per research project, and write the finding into it with the link the moment you find it.

Not a summary later. Now, while you have it. Two lines:

Roof survey, Company A — £480, 3 week wait
https://example.com/quote/...

Then close the tab. The information is the thing you needed and it is now in a place you can search, next to the other eleven findings, in a form you can compare. Thirty tabs are not comparable with each other; a list is.

This does not need a tool. Any document works. What it needs is doing it at the moment rather than at the end, because at the end you will not remember which tab said what.

Sources go in one bookmark folder, unsorted

One folder per project, no subfolders, no naming discipline. Everything goes in.

The urge to organise is what stops you filing. If saving requires a decision about where, the tab stays open instead. You are not building a library — you are building something you can search from the address bar, and search does not care about folders.

Delete the folder when the project ends. That is the part people skip, and it is why bookmark bars become archaeology.

Pending is where the tabs actually come from

Something happens here later. This is the pile that resists everything, because a tab is the only place most people can store “this page, and a date”.

A viewing on Saturday. An application that opens on the 1st. A pre-order on Thursday. A council decision published on the 15th. A form you can only submit after something else happens.

Each one is two facts — page and time — and no amount of tab organising can hold the second one. So the tab stays open as a proxy for a date, and it stops working the moment it is surrounded by twenty other tabs that look identical.

Put the date somewhere with the page and the tab can close. In Latr that is one key on the page, or — better for this case — highlighting the date the page itself states and right-clicking it. Applications open 1 March; select those words, right-click, done. It reads the time out of the selection and the page comes back on the day.

What this does to a research window

My typical project now runs at four to six tabs instead of thirty, and the difference is not discipline, it is that the pending pile has somewhere else to be.

The other change is that the research becomes reviewable. When findings are in a document and pending items are in a dated queue, you can look at the state of a project in two minutes. Thirty tabs cannot be reviewed at all — you can only re-read them, which is why you never do.

Practical bits

Tag the whole project. Everything you save for the house search gets one tag, so the queue can show you just that. Tags and folders — tags are unlimited, so there is no reason to be sparing.

Use a note. Latr’s reminders take a note, and it is searchable. “Compare with Company A quote” is worth four seconds and saves you re-reading the page to remember why you kept it.

Spread the reading out. If you have fifteen long documents, do not schedule them all for Sunday. One a day for a fortnight is how a pile actually gets read.

Take the thumbnails. Research pages have interchangeable titles — six council PDFs all called Planning Application. The small picture in the queue is what tells them apart.

Close the project properly. When it ends, delete the bookmark folder, archive the document, and clear the tag. A finished project that leaves residue is why the next one starts at thirty tabs.