Key takeaways
- Rewriting every product description around the purchase decision drove a 3x increase in content-driven sales for this D2C shoe brand.
- The bounce problem was not traffic quality. Shoppers arrived, read manufacturer spec copy that answered none of their questions, and left.
- Product content earns its keep when it resolves the three things a shopper is actually deciding: whether it fits, how it compares, and whether it lasts.
- The same content that converts shoppers is the content AI assistants quote when someone asks which product to buy, so the work pays twice.
A D2C shoe brand came to us with a high bounce rate on the pages that should have been converting best: its own product pages.
The descriptions were bulleted lists of materials supplied by the manufacturer. Rewriting all of them around the buyer decision, paired with long tail search targeting, drove a 3x increase in content-driven sales.
This is the method we used, in the order we used it, so you can run it on your own catalogue.
What was actually wrong with the product pages?
Short answer: The copy described the product instead of helping anyone choose it. Manufacturer spec lists answer the question "what is this made of", when shoppers are asking "will this fit me, how does it compare, and will it last". Pages that answer none of the three get read for a few seconds and abandoned.
A high bounce rate on a product page is almost never a traffic problem. The shopper searched for the product, clicked the right result and arrived with intent.
Something on the page then failed to answer the question that brought them. On this catalogue that something was the copy.
The three decisions a product page has to resolve
- Fit and suitability. Will this work for me specifically, with my size, my use case and my constraints? For footwear that means width, arch, break in period and what the shoe is actually for.
- Comparison. How is this different from the other four options I have open in tabs, including your own? A shopper who cannot tell your two models apart buys neither.
- Durability and risk. What happens after month three, and what happens if I am wrong? Returns policy and real wear expectations belong on the page, not buried in a footer link.
Manufacturer copy is written to describe a product to a retailer. It is not written to sell that product to a person, which is why a brand that ships it unedited is competing with every other retailer running the identical paragraph.
Baymard Institute's ongoing cart abandonment research puts average documented abandonment above 70%, with unclear information a recurring cause. Content that removes uncertainty is doing conversion work, not decoration.
How do you rewrite a product catalogue without guessing?
Short answer: Start from the questions buyers already ask, not from a copywriting brief. Pull them from your own support tickets, reviews, returns reasons and on site search, then write each product page to answer the ones that apply to it. The source material for good product copy is already inside your business.
The temptation with a catalogue rewrite is to start writing. That produces prettier copy that converts the same, because it is still guessing at what the shopper wanted to know.
We work from evidence the business already has instead.
- Mine your returns reasons first. Every return with a reason attached is a shopper telling you exactly which question the page failed to answer. Ran narrow, not what I expected, uncomfortable after a day: each one is a missing paragraph.
- Read the support tickets. Pre purchase questions are the highest value content brief you will ever get, because somebody cared enough to ask rather than leaving.
- Mine your own reviews, including the bad ones. Reviewers describe products in buyer language, not brand language. Three star reviews in particular name the tradeoffs an honest page should state up front.
- Check on site search. What people type into your search box is an unfiltered list of what they could not find.
- Then target the long tail. With the real questions in hand, the search demand is easy to map: specific, lower volume queries that convert far better than the head term.
Only at that point is it worth writing. Each page now has a brief that came from buyers rather than from the brand.
What replaced the spec lists
| Manufacturer copy does this | Buyer focused copy does this |
|---|---|
| Lists materials and construction | Explains what those materials mean for comfort, weight and wear |
| Describes the product in isolation | Positions it against the alternatives, including your own other models |
| Uses the brand vocabulary | Uses the words shoppers actually search and ask with |
| Identical across every retailer selling it | Unique to you, which is the only version search can reward |
| Silent on who it is wrong for | States who should buy something else, which is what makes the rest credible |
That last row does more work than it looks like it does. A page that tells a shopper this model runs narrow and points them to a wider one converts better overall, and generates far fewer returns.
Why does this content also win AI recommendations?
Short answer: Because an assistant answering "which running shoe should I buy" needs extractable, specific, comparative facts, and that is exactly what buyer focused product content contains. Spec lists give an assistant nothing to quote. Content that explains fit, comparison and durability gives it a reason to name you.
This is the part that was not obvious when the work was done, and is now the larger half of the opportunity.
When someone asks an AI assistant for a product recommendation, the answer is assembled from retailer data, reviews, roundups and discussion rather than from brand slogans.
Google is explicit that its AI features draw on core ranking and quality systems, so the content that ranks is the content in contention to be cited.
- Specific claims are quotable. "Runs about a half size small, with a two week break in" can be cited. "Premium craftsmanship" cannot.
- Comparisons answer the real prompt. Most shopping prompts are comparative. A page that honestly compares your models is eligible for far more of them.
- Structured product data makes the facts machine readable. Google's product structured data guidance covers price, availability and review markup, which is how an assistant states facts about your product with confidence.
- Consistency across retailers matters. If your specifications disagree between your own site and the marketplaces, an assistant has no reliable version to repeat.
Our guide to showing up in ChatGPT shopping results covers the retail specifics, and agentic commerce covers where this is heading as assistants start completing purchases.
How do you run this on your own catalogue?
Short answer: Do not rewrite everything at once. Rank products by revenue, take the twenty pages that matter most, rewrite those against real buyer questions, and measure before going further. A tested template beats a catalogue of untested new copy.
Catalogue rewrites fail when they are run as one large project. By the time it ships nobody knows which change did what.
- Rank by revenue, not by how bad the copy is. Your worst page is irrelevant if nobody buys that product.
- Take the top twenty. Enough to read a signal, small enough to finish in weeks.
- Write to the three decisions. Fit, comparison, durability, every time, using the language from your tickets and reviews.
- Measure the right things. Add to cart rate and return rate per page, not sessions. A rewrite that lifts conversion and cuts returns has paid for the whole project.
- Then template what worked. Roll the pattern out, do not roll out a style.
Our guide to ranking product and category pages covers the technical side, and bottom of funnel keyword research covers finding the long tail queries that convert.
About the numbers on this page
These figures come from this one engagement and describe what happened for this client. They are not a forecast for another business, because the starting position decides most of the outcome.
What does transfer is the method below, which is why it is set out in full.
Frequently asked questions
What kind of content increases e-commerce sales?
Content tied to the purchase decision: fit and suitability detail, honest comparisons against alternatives, and durability or risk information. It outperforms generic blogging because it resolves the specific uncertainty that stops a shopper buying, on the page where they are deciding.
Should I rewrite manufacturer product descriptions?
Yes, for the products that drive your revenue. Manufacturer copy is identical across every retailer selling the item, so it gives search engines nothing to prefer about your version and gives shoppers nothing they cannot read elsewhere.
Where do I find out what my product pages should say?
Inside your own business. Returns reasons name the question the page failed to answer, pre purchase support tickets are a ready made content brief, three star reviews name the real tradeoffs, and on site search shows what people could not find.
How many product pages should I rewrite first?
Around twenty, chosen by revenue rather than by how poor the current copy is. That is enough to produce a readable signal on conversion and return rate, and small enough to complete before the business loses patience with the project.
Does product content affect AI shopping recommendations?
Substantially. Assistants assemble product answers from extractable, specific facts and comparisons. Spec lists offer nothing quotable, while content explaining fit, comparison and durability gives an assistant a concrete reason to name your product.
What should I measure on a rewritten product page?
Add to cart rate and return rate for that specific page, before and after. Sessions and bounce rate move for many reasons, but a rewrite that lifts add to cart while reducing returns has demonstrably done its job.
Sources and further reading
- Baymard Institute: documented cart abandonment rate statistics
- Google Search Central: product structured data
- Google Search Central: AI features and your website
- Google Search Central: creating helpful, reliable, people-first content
- US Federal Trade Commission: endorsement and testimonial guides
Put this method on your product pages
We will show you which of your product pages lose shoppers and why, before you commit to anything. Start with our e-commerce AEO service or book a free visibility and content audit.