Product Specification Sheet Template

A product specification sheet template earns its place only if what comes back can be filtered, compared and pushed to a channel without anyone retyping it. Most cannot. The dimensions arrive as one string, the certifications arrive as a paragraph, and the file becomes a document you read rather than data you use.

This product specification sheet template is built to avoid that. It is a single product sheet in two parts: a Product tab holding identity and commercial basics, and an Attributes tab holding one row per measurement with the unit in its own column. That separation is the whole design.

What is in the product specification sheet template

Three tabs.

How to use this. Written for whoever you send the file to, not for you. Four rules and a note on the attribute set.

Product. Sixteen fields down the page, each with a Value cell and a Notes cell telling the person filling it in what the field means. Required fields are filled red.

Attributes. Nineteen starter attributes with columns for Attribute, Value, Unit, Required and Notes. The units are pre-filled so nobody has to guess whether you wanted millimetres or centimetres.

Filling in the Product tab

Four fields are required, and they are required because everything downstream matches on them.

Supplier part number. The supplier's own code, exactly as they use it. Not your internal SKU, not a shortened version, not one with the pack size appended. This is the only stable key between their system and yours, and the moment it is edited for tidiness the link breaks.

Manufacturer. The brand owner, not the distributor you bought it from. This one is wrong more often than any other field on the sheet. A bracket made by Acme and bought through a wholesaler has Acme in this field. Put the wholesaler in and you end up with two records for the same product, and nobody notices until a customer searches the brand and finds half the range.

Product name. How a buyer would say it out loud. Heavy duty steel shelf bracket, not BRKT-HD-STL-300-GALV.

GTIN, EAN or UPC. Digits only, no spaces, no dashes. Format the column as text before anything is typed into it, because Excel strips the leading zero from a thirteen digit barcode the first time the file is opened and nothing warns you.

The rest are optional but worth chasing. Country of origin and commodity code are the two most commonly skipped, and they are the two that hold up a shipment or a customs declaration later. Unit of measure and pack quantity decide whether the price means anything. Lead time is a whole number of working days, not "usually about a week".

Datasheet URL and Primary image URL want a direct link to the file itself. A link to a product search page, or to a portal that needs a login, is the same as leaving the field blank.

The Attributes tab, and the one rule behind it

One value per row, with its unit in its own column.

A row reading Length, 400, mm can be filtered, sorted, converted between units, compared against a competitor's product and read correctly by an AI assistant answering a question about size. A row reading Dimensions, 400mm x 200mm x 150mm can do none of those things. It is a sentence about the product rather than a property of it.

The same applies to ranges. Operating temperature of minus twenty to sixty degrees is two rows, Operating temperature min and Operating temperature max, which is why the template ships with both. A single cell containing a dash cannot be filtered on, and "show me everything rated below minus ten" is a question the catalogue then cannot answer.

Certifications get one row each. A cell reading "CE, UKCA, RoHS compliant" reads fine and filters on nothing.

What the nineteen starter attributes cover

The set in the file is the one that recurs across technical and industrial catalogues: the three dimensions and both weights, material and finish, colour, the electrical group (operating voltage, current rating, power rating, frequency), IP rating, the two operating temperature limits, thread size, connection type, certification and warranty period.

Five of those are marked required, because a product with no dimensions, no weight and no material cannot be shipped, filtered or listed. The rest are marked optional and most categories will use only some of them.

Treat it as a starting point rather than a schema. A real category usually needs between fifteen and forty attributes to support proper faceted search, and the right fifteen for cable glands are not the right fifteen for safety boots. Add the ones your category actually trades on, delete the ones it does not, and keep the format.

A worked example

The file ships with one filled in row so nobody has to guess at the format. A 300mm galvanised steel shelf bracket, supplier part number ABC-1234, made by Acme, barcode 5012345678900, sold each, five day lead time.

On the Attributes tab that becomes Length 300 mm, Width 200 mm, Height 40 mm, Weight 0.85 kg, Material Galvanised steel, Finish Galvanised, Colour Silver. Seven rows, each one filterable.

Written as prose it would be one line: "300mm galvanised steel bracket, 200 x 40, 850g". That line is shorter and it is worth nothing, because no filter, feed or marketplace listing can be built from it without a person reading it first.

Delete the example row before you send the file out. Suppliers copy it more often than you would expect.

Five things that make a specification sheet unusable

Writing N/A into an empty cell. A blank is a gap you can find with a filter and chase. N/A is indistinguishable from a real value, survives every import, and quietly becomes permanent. The instructions tab says this explicitly because it is the single most common habit to break.

Mixing unit systems in one column. Half the rows in millimetres and half in inches is invisible until something is ordered at the wrong size. One unit system per sheet, stated at the top.

Putting the unit inside the value cell. "400mm" is text. "400" with mm in the Unit column is a number you can sort, filter and convert.

Merged cells. A merged cell has one value and several addresses, and every parser resolves that differently. They look tidy in Excel and break everything downstream.

Using the sheet as a marketing document. Promotional claims belong in the description. A specification sheet carrying "market leading durability" gets distrusted for the measured values too, which is the expensive part.

Sending the specification sheet template to suppliers

Send the specification sheet with the example row deleted and the Attributes tab already cut down to your category. A supplier looking at nineteen attributes when eight apply will fill in the eight and leave the rest, which is fine, but they will also spend the first ten minutes working out which eight. Do that work once, on your side, and completion rates improve.

Say when you need it back and what happens if a required field is missing. Vague deadlines produce vague files.

If the supplier already has the data in a spec sheet PDF, take the PDF. Do not make them retype it into your columns. Retyping is the step that produces both the delay and the errors, and it is the step most worth removing.

When the spreadsheet stops being the right tool

Filling one of these in properly takes twenty to thirty minutes per product, and that is before chasing the supplier for the values that are not on the datasheet. Across a thousand products that is roughly a quarter of a person year, spent transcribing data that already exists.

Because it does already exist. The values are in the supplier's PDF, on their website, in the BMEcat file they send their other customers. They are just not in your structure.

Extracting them is a different job from designing the sheet. SKULaunch reads the supplier PDF, pulls the attribute values into the structure you have defined, flags what it could not find rather than guessing at it, and routes low confidence values to a person to check. The template stays useful as the definition of what good looks like; the filling in stops being manual.

For the wider version of this problem across a whole catalogue, see product data enrichment.