Skip to main content

Content quality

Find the thin product pages Every SKU gets a score.

Every product gets a 0–100 content score against your own rules, and the Fix Issues list ranks the gaps so you know where to start.

Start freeBook a demo
  • 0–100 score per product
  • Score per field, not just per product
  • Compare before publishing

Key features

  • Score the catalog against your rules

    Description length, filled specs, image count, SEO fields: you decide what counts and what a passing product looks like.

  • Work a ranked list, not a hunch

    Fix Issues sorts the gaps by how many products carry them, so a single fix can move hundreds of listings at once.

  • Rewrite in your own voice

    Content Logic holds the tone, vocabulary and banned phrases; every generated description follows them, per brand and per channel.

What a low score looks like

Thin content is rarely one bad page

It is the same field missing across a whole category, which is why fixing it one product at a time never ends.

  • A field empty on a whole category

    Nobody mapped the supplier's material column, so 400 products in one category have no material at all and nobody noticed.

    Result: field health shows the gap per attribute, per category.

  • Descriptions of two sentences

    The import brought the supplier's one-liner across and it became the product description, on every SKU from that supplier.

    Result: the score flags length and depth, not just presence.

  • Copy that drifted per writer

    Three people wrote descriptions over two years, with three tones, three spellings of the brand and three ideas of what a spec list is.

    Result: one rule set in Content Logic applies to every rewrite.

How it works

From score to published rewrite

Four steps take a catalog you have never audited to listings you have actually approved.

  1. 01

    Set what quality means here

    Pick the fields that count, the minimum description length and the images a product needs. The score is your definition, not a generic template.

    A score you can defend to the people who own the catalog.

  2. 02

    Read the field health, not the average

    The field view shows how complete each attribute is across the catalog, so you see that one missing column, not a blurry catalog-wide percentage.

    The gap that affects the most products, named.

  3. 03

    Work the Fix Issues list

    Each issue lists the products it affects and hands the whole set straight to bulk enrichment, so one pass clears the issue instead of one product.

    Hundreds of listings improved from a single entry in the list.

  4. 04

    Approve the rewrite, then publish

    Generated copy lands next to the current text with the score it would produce, so you accept, edit or regenerate before anything reaches a channel.

    Nothing goes live that a person has not read.

Content Logic

One voice across the whole catalog

The rules live in one place and apply on every generation, so consistency does not depend on who happened to write that product.

  • Tone and vocabulary once

    Set the register, the words you always use and the phrases you never want to see. Every generated description follows them.

  • Rules per brand and per category

    A premium brand and a value brand under the same roof keep their own wording without maintaining two prompt libraries.

  • Prompts you can read and edit

    Prompt Lab holds the actual prompt behind each generation, so a bad result is something you can inspect and change.

The rules that every generation reads: tone, preferred wording, and what must never appear.

What the score measures

The checks behind the number

Each one is a rule you can switch on, weight, or turn off entirely.

  • Description depth

    Length, structure and whether the text says anything the spec table does not already say.

  • Attribute completeness

    Which required attributes are filled, per family, so a shoe and a drill are not judged by the same list.

  • Media

    How many images a product carries, whether they have alt text, and whether the main image is set.

  • SEO fields

    Whether the SEO title, meta description and slug exist and stay inside the length each channel accepts.

FAQs

Questions about scoring product content

What teams ask before they let a score decide their backlog.

Still have questions?

Can't find the answer you're looking for? Please get in touch with our team.

Contact Support

It does not decide on its own. You do. You pick which fields count, how long a description should be, how many images a product needs and which SEO fields are required. Each product is then scored 0–100 against that definition, per field as well as overall, and the Fix Issues list ranks the gaps by how many products carry them.

Content Logic holds the tone, the vocabulary and the phrases to avoid, per brand and per category, and every generation reads those rules. Prompt Lab holds the prompt itself, so you can see exactly what was asked for. WISEPIM does not infer your voice from your existing copy: you write the rules once.

Both. Rewrite a single product from its editor, or send an entire Fix Issues entry, meaning every product carrying that gap, to bulk enrichment and let it run in the background. Most teams start with their best sellers, check the output, then apply the same prompt wider.

Yes. Generated text is written next to the current value with the score it would produce, and you accept, edit or regenerate per field. Nothing reaches a sales channel until it is saved and exported.

No, and we will not claim it does. The score measures completeness and consistency against your own rules: it tells you which pages cannot answer a buyer's question, not what a fix will earn. Analytics shows what happened to traffic and revenue after you shipped the change; the two are worth reading together.

Next step

Score your catalog before you rewrite it

Import a catalog, set what a good product looks like, and let the Fix Issues list decide your first week of work.

Start free
Book a demo

Quality documentation

How the score is built, which checks are available, and how rules apply per brand and channel.

(opens in a new tab)