Writing Two Hundred Product Descriptions

The comparison that matters

Most advice about machine-written content compares it against good writing, and concludes correctly that it is worse.

For two hundred product descriptions, that comparison is not the real one.

The realistic alternatives are the supplier's text, which every competitor also has, or an empty field, which is what most small shops actually have.

Against those, a drafted description that is specific and readable is a straightforward improvement.

Why it works here

• The facts exist in your product data.
• The task is arranging, not inventing.
• Each output is short.
• The format repeats, which suits it.
• Nobody was writing them anyway.

The first is what separates this from a service page. You already hold the dimensions, materials, compatibility, and specifications, so the tool is working from real information rather than filling a gap with plausibility.

Feed it the specifics

Which is the whole difference between usable and generic.

Give it the actual attributes for each product: measurements, materials, what it fits, what is included, weight, origin.

Add anything you know that is not in the supplier data: what customers use it for, what they ask before buying, what it is commonly confused with.

Those last three are what make your description better than the supplier's, and they usually apply to a whole category rather than to each item.

So writing them once per category and supplying them alongside each product is a small amount of work with a large effect.

Without the specifics, you get two hundred descriptions that could describe anything.

Vary the structure deliberately

Since the repetition that helps also hurts.

Left alone, the tool produces two hundred descriptions with the same rhythm, the same opening move, and the same closing sentence.

A customer browsing a category sees that immediately, even without knowing why.

Ask for three or four different structures and rotate them, or vary the length by product importance.

Your best sellers deserve a longer, hand-finished description, and the long tail can have the standard treatment.

That split is the practical answer: fifteen written properly, the rest drafted.

A worked example

A shop had about three hundred products, of which forty had the supplier's description and the rest had a single line.

They exported the product data, wrote a short note per category covering what customers ask and what the items are confused with, and drafted descriptions in batches of twenty.

Each batch was read before importing, which caught several claims about compatibility that were not correct.

The work took about two days spread over a fortnight.

Traffic to product pages rose over the following months, mostly from longer, more specific searches.

Nobody in the business would ever have written three hundred descriptions by hand, which was the actual choice being made rather than the one usually discussed.

Check the claims, not the prose

Since the risk here is specific rather than general.

Product descriptions contain compatibility statements, measurements, materials, and safety claims, all of which are checkable and all of which the tool will confidently invent when the data is thin.

Read for those rather than for style, which is the opposite of how people edit.

Anything asserting that a product fits, contains, complies with, or is suitable for something needs to be true.

A wrong compatibility claim produces a return, a refund, and a review, so this pays for itself in avoided problems rather than in traffic.

Work in batches small enough to actually read, which is around twenty, since a batch of a hundred gets skimmed and a batch of twenty gets checked.

Where it stops working

Worth knowing before starting.

Products where the appeal is subjective rather than specified, such as anything about taste, feel, or style.

Anything with regulatory or safety language, where the wording matters legally.

And your handful of genuinely important products, where the description is doing real selling and deserves a person.

For those, use the tool to draft questions or a structure and write the words yourself.

The dividing line is roughly whether a customer could verify the description against the product, since anything they cannot check is a claim rather than a specification.

Do the categories first

A sequencing point that improves the return.

Category pages usually receive more traffic than individual products and almost always have less written on them.

A few hundred words per category explaining what the range covers, how to choose, and what the differences mean is more valuable than the equivalent effort spread across items.

It also gives you the per-category material that improves every product description underneath it.

So write the categories, then draft the products using what you wrote.

Publish them in stages

A practical protection against the volume risk, and it costs nothing.

Adding two hundred pages on a Tuesday is a pattern that looks nothing like how a business ordinarily grows.

Release them over several weeks in batches, starting with your best-selling categories, which is also the order that returns soonest.

That gives you a chance to see whether the first batch is doing anything before committing the rest, and to change the approach if it is not.

It also spreads the checking, which is what actually determines whether the descriptions are worth having.

The counter-case

Volume has its own risk.

Two hundred thin pages added at once is the pattern search engines have spent the last year discouraging, and quantity without specifics is exactly what that targets.

Descriptions that merely restate the specifications add nothing a customer could not see from the table above them.

And a shop with thirty products should write those thirty properly rather than drafting them.

Write the categories first, supply real specifics, draft in batches of twenty, and check every compatibility claim.

The method

1. Write the category pages yourself.
2. Note per category what customers ask.
3. Export the product attributes.
4. Draft in batches of twenty.
5. Read each batch before importing.
6. Check every compatibility claim.
7. Hand-write your best sellers.

Step two is what makes the output yours rather than generic, and it is a paragraph per category rather than per product, which is why it is achievable.

The problem this replaces is covered in writing descriptions that are not the supplier's.

Frequently asked questions

West Coast Media Solutions Inc. provides web design, web development, hosting, digital marketing, and business consulting to organisations across Canada, drawing on more than twenty-five years in the field.