Product Catalogue Template

Most catalogue files are a list. This product catalogue template is a measurement. The last column counts how many of the eight fields a product genuinely needs are actually populated and turns it into a percentage, down to row 500. Sort that column ascending and you have your enrichment backlog, in priority order, without running a project to find it.

That is the only unusual thing about the file. Everything else is a well ordered set of columns, which matters less than people think, because the columns are rarely the problem. Knowing which rows are broken is the problem.

What is in the product catalogue template

Two tabs: a How to use this tab, and a Catalogue tab with twenty eight columns across five groups.

Identity. SKU, parent SKU, product name, brand, GTIN, MPN. What every downstream system matches on.

Classification. Category level 1, 2 and 3. Three levels is usually enough. Catalogues that insist on six normally have a taxonomy problem rather than a depth problem, and adding levels hides it rather than fixing it.

Commercial. Cost price, sell price, currency, stock, lead time in days.

Content. Short description, long description, feature bullets, primary image URL, image count, datasheet URL.

Attributes. Colour, size, material, weight and the three dimensions. This is the group that decides whether filters work, and it is almost always the emptiest.

The twenty eighth column is Completeness %, which is a formula rather than something you fill in.

How the completeness score works

The formula counts eight fields per row: SKU, product name, brand, GTIN, category level 1, sell price, long description and primary image URL. Eight out of eight is 100%. Six is 75%. It runs from row 2 to row 500 and returns blank on empty rows, so an unfinished file does not fill the column with zeros.

Those eight were picked because each one independently blocks something. No GTIN blocks marketplace listing. No image blocks the product page. No category blocks navigation. No long description blocks organic search. A product missing any of them is not a weak record, it is a record that cannot do a specific job.

Extending it is deliberately easy. If your category cannot be sold without a material value, add material to the count and change the divisor. The point is that the score reflects what you actually need, not a generic notion of completeness.

What the number usually comes out at

Run it against a real export and it tends to land lower than expected. Catalogues that feel broadly fine often come out between forty and sixty per cent on a measure like this, and the missing part is concentrated in attributes and identifiers rather than in descriptions.

That distribution is the useful finding, and it has a simple cause. Thin copy is visible: somebody reads the product page and notices. Missing attributes are invisible from the front end. The page looks complete. The filter behind it is empty.

Which is how a catalogue ends up looking populated and performing empty. A customer filters by material, gets four results out of nine hundred, decides you do not stock what they want, and leaves. Nothing in the analytics says why.

Three ways to use the file

Sort ascending and work down. This is the whole point. It beats working alphabetically, by supplier, or by whatever arrived most recently, which are the three orders enrichment work normally happens in.

Filter to your best sellers first. The commercial cost of a gap is not spread evenly. A missing barcode on a line that sells four a year does not matter this quarter. The same gap on a top fifty line is money. Sort by completeness within your top sellers and the priority order is obvious.

Re-run it monthly and watch the number move. This is the only way to know whether enrichment effort is outpacing new product arrivals. Plenty of teams work hard on enrichment for six months and end up at the same percentage they started at, because three thousand new SKUs came in behind them. The absolute number matters less than the direction.

A worked example

The example row is a 300mm galvanised steel shelf bracket. SKU ABC-1234-300, parent SKU ABC-1234, brand Acme, GTIN 5012345678900, MPN ABC-1234, categorised Hardware then Brackets then Shelf brackets. Cost 12.50, sell 24.99, GBP, 25 in stock, five day lead time.

Short description, long description, three feature bullets, a primary image URL, an image count of three, a datasheet URL. Colour silver, size 300mm, material galvanised steel, weight 0.85kg, 300 by 200 by 40mm.

That row scores 100%. It is also, deliberately, more complete than most rows in most catalogues, which is the comparison worth making when you open your own export next to it.

Delete the example row before you use the file.

Two columns worth getting right early

Parent SKU. This is what groups variants into one product on the website. Filling it in retrospectively across a catalogue is a long job, and getting it wrong produces either nine separate product pages for one bracket or one page with a size selector that lists nothing.

Image count. A single number that tells you more than it should. Products with one image convert worse than products with four, and a catalogue where the median is one is a catalogue with an image sourcing problem rather than a copywriting problem. It is also the fastest gap to quantify, which makes it a good first argument when asking for budget.

When the spreadsheet becomes the constraint

Past roughly five thousand products a catalogue file like this starts to creak. The formulas slow down, two people cannot work in it at once, and the version you are looking at is a snapshot of an export taken on a date you cannot quite remember.

The measurement is still the right measurement. It just needs to run against the live catalogue rather than against a copy of it, continuously rather than whenever somebody remembers to re-export.

That is what completeness scoring inside a product data platform does. SKULaunch applies the same logic against the governed record, flags the specific missing attributes per product, and fills what it can from supplier PDFs, spec sheets and product URLs rather than leaving the gap for a person to type. The template is a good way to find out whether that is a problem you have. For most catalogues over a few thousand lines, the answer it gives is uncomfortable and useful.

For the wider version of this, see product catalogue enrichment software.