Blog / Shopify & e-commerce /

How to use product data without turning copy into a spreadsheet

A product feed can hold hundreds of attributes. Paste them all into a description and you get a spreadsheet with full stops. Nobody reads it, and nobody buys from it.

Product data should help the writer decide what matters, then become clear prose that explains a choice. Start with verified facts and the customer’s decision, and include only the details that improve understanding or prevent a bad fit.

Sort facts by decision value

Separate identity, compatibility, dimensions, performance, included components, care and constraints. Some details are essential for choosing. Others belong in a specification section, and some are internal only. Confirm units and variant scope before you write. A value attached to the parent product may not apply to every child option.

Turn fields into explanation

Use prose to say why a fact matters: which device a connector fits, what a pack contains, how one size differs from another. Keep the exact values in a scannable specification list. Don’t paraphrase numbers in a way that changes their meaning, and don’t add benefits the source data doesn’t support. Let the page structure carry the detail, and don’t force every field into a paragraph.

Review for gaps and clutter

Compare the finished copy with the source record and check no important qualification disappeared. Then cut repeated attributes and generic filler. Ask a colleague to find one critical fact quickly.

Write for the question, not the field

Fields are how the database thinks. Shoppers think in questions. Will it fit? What’s in the box? Is it the same as the other one? Walk through the product’s fields and turn each important one into the question it answers. If you can’t name the question, the field probably belongs in the specification list, not the description.

I’d rather ship a short description that answers three real questions than a long one that recites twelve attributes. If a detail only matters to a small group of buyers, link to it or put it in the specifications.

Handle variants and exceptions

Variants are where product data tends to bite. A dimension that is true for the large size can be wrong for the small one, and a “waterproof” attribute may apply to one colour’s material only. Check which variants each statement covers and say so on the page when the answer differs.

When I read a draft against its feed, I look for sentences that quietly generalise from one variant to all of them. If the data isn’t clear, ask the product owner. A short “ask first” is cheaper than a returned order.

Apply the review to one real case

Define what a finished page changes for the shopper, not just what work the team did. A product with tidy data can make a catalogue pattern look more universal than it is.

Follow one real product from its source record to its published page, and check what a visitor can understand or do without an internal explanation. A copied export may describe an earlier model or configuration.

Separate products that share a template from products that share a shopper need, because a bulk rule may suit one and not the other.

Check the page where customers see it. Imports, apps, caching or theme behaviour can alter approved copy. If the data can’t answer a question, say so and assign a follow-up. When fields conflict, compare timestamps, definitions and ownership before choosing a value.

Keep sources in an approved system and link to them instead of pasting confidential data into shared notes. If the information is incomplete, publish a smaller accurate description instead of a bigger guess. Allow exceptions only for real problems and write down their limits. When you close the work, check feeds, support answers and linked pages that use the same facts, and know how to restore the previous copy.

A worked example: from fields to sentences

Say a feed lists a backpack with capacity, laptop sleeve size, weight, material, water resistance and strap length. Reciting all six gives you a spreadsheet. Ask what the buyer is deciding instead. A commuter wants to know whether their laptop fits and whether the bag copes with rain. So the opening sentences say which laptop sizes the sleeve takes and what the water resistance does and doesn’t cover, in the source’s own wording. Weight and strap length go in the specification list. Material gets a line only if it changes how you care for the bag. Two sentences, one scannable list, and nothing that contradicts the feed.

Which product-data field needs explanation rather than another line in the description?