How do you find your item in a table with thousands of rows?

Search box behaviour on a long table

The search box is the fastest tool available and the easiest one to misjudge. On a long table it usually matches text anywhere inside a row, which means a common word returns hundreds of results and a misspelling returns none. Begin with the shortest string that is still unusual: a model code, or a brand plus one distinctive word.

Results are not saved anywhere. If a search found the row you wanted, keep the string somewhere else, because a reload clears it and you will be searching from scratch a second time.

Filters that cut thousands to dozens

Filters work by category rather than by text, and they are the second tool to reach for. Brand, category and date are the common ones. Applying two filters narrows a list faster than refining a search term, and it has the advantage of showing you a neighbourhood rather than a single row.

Filter on the thing you are most certain about. If you are sure of the brand and unsure of the model, filter by brand and read the list. Guessing a model name into a search box is how people conclude that an item is missing when it is sitting there under a different name entirely.

Brand and model naming that breaks search

Naming is the main reason a row cannot be found. The same shoe appears as a full model name, as an abbreviation, and as a string of seller-chosen characters that means nothing outside that listing. A table written by several people over several months will contain all three, and none of them is wrong.

Search for the fragment that survives renaming: a number, a silhouette name or a material word. Numbers are the most stable part of any listing title, which is why a model code is the best search input a table can offer you.

Sorting by date to find the newest rows

Sorting by date answers a different question from searching. A search asks where a particular item is; a date sort asks what has been added lately. If you are browsing rather than hunting, sorting newest first shows you the rows somebody actually touched, which is a better use of a large table than scrolling through its middle.

Watch for a date column that was clearly filled in all at once. A long block of identical dates means the table was stamped rather than maintained, and the newest rows in it may not be new at all.

Browser find on a virtualised table

The browser’s own find function sees only the rows that have already been rendered. Long tables frequently render a window of rows and load more as you scroll, so find can report nothing for an item that is genuinely present further down the list.

A quick test settles it: search for a word you can see on screen right now. If find cannot locate a visible word, the table is rendering in windows and you should stop relying on that tool for the rest of the session.

Bookmarks and saved searches

A bookmark to the table is not a bookmark to your row, because filters are usually not part of the address. Save the search string and the item name together in a note you control, and treat the bookmark as a door rather than as a place.

Where a table does keep filters in the address, that address is worth keeping for the length of a working session, since it restores the exact view. That is the closest thing to a saved position a public table can offer.

When the row you want is not in the table

Absence is a result. A table with thousands of rows is still a selection, and whoever wrote it chose what to include. Before concluding that an item cannot be found, check whether the table covers that marketplace or that category at all.

Our own tables state their scope rather than implying it. The measured footwear rows carry a method as well as a figure: 46 rows measured on three pairs of each model, with a step column recording how far the next labelled size actually moves, which comes out at a few millimetres on footwear and a few centimetres on garments. A table that names its scope can be trusted inside it and nowhere else.

1. Check the table’s stated scope before searching any harder.
2. Search a number before a name.
3. Filter on the field you are certain about.
4. Keep the search string, not just the bookmark.

Where the numbers in this entry come from

Own measurement run one, filed as KB-2026-03: the step between adjacent labelled sizes was measured on the same models, and it comes out at a few millimetres on footwear, which is why a labelled size alone cannot be used to guess a fit.

◇KB-2026-03· 2026-09-30 See the batch ledger · Challenge a figure

Entries filed next to this one

• How is the link table laid out, column by column?

  Most tables run left to right: a row number, the link, a short title, the variant, then a date. Past that the order varies by author. Treat the first four columns as the identity of the row and everything further right as commentary, because authorship and maintenance differ from table to table.

• What is a CNFans spreadsheet, exactly?

  It is a table of links, not a shop and not a file you download. Each row points at one listing on a domestic marketplace, with the few details that row needed to be useful. The buying itself happens on the agent platform, which is a separate object from the table you are reading.

• What does each column actually tell you?

  Treat every header as a claim rather than a fact. Numeric columns with a stated method are worth the most, seller wording is worth the least, and a status column is only as good as its date. Where two columns disagree, the one with a method and a date wins.

Gauges this entry points at

• Slip cross-check bench

  Compare what the slip claims against what you measured.