On a marketplace, when someone searched a category term, they landed on the platform's category page. Your listing was one tile on it. You never owned that page and you could not have.
On your own store, that page is yours. Most sellers who move do not realize it, leave it as an empty grid of products, and then wonder why they only ever rank for exact product names.
This is the highest-leverage page type on a small store and it is usually the emptiest.
Why the category page is the one that ranks
Think about the queries people actually type. Most of them are not a specific product. They are a type of thing, with a modifier: a material, a color, a use, an occasion, a recipient, a size.
Those queries want a page that shows options, not a page that shows one product. A general search engine, choosing what to show, will prefer a page that lets the searcher compare.
A product page cannot be that page. It shows one thing. It can rank for its own specific description and it is the wrong shape for a browse query.
So the broad and mid-breadth queries, which is most of them, want a collection page. If your store does not have one for a category, you have nothing to offer that query.
What most collection pages look like, and why it fails
The default is a title, a grid of products, and nothing else.
That page has almost no text, nothing that distinguishes it from every other store's identical grid, and nothing that answers the question the searcher actually has, which is usually "which of these is right for me".
It is not a bad page. It is an unfinished one, and finishing it is one of the few pieces of work on a new store that reliably pays.
What a collection page should contain
Not a template to copy. A set of jobs the page has to do.
A title that names the category the way buyers say it. Not your internal name for the range. The words in your buyer messages and your old marketplace search terms report.
Introductory text that helps someone choose. Two or three short paragraphs above or near the grid, written for a person who has arrived not knowing which option they want. What are the meaningful differences between these products? What decides it? What do people usually get wrong?
This is the part that does the ranking work and it is the part everybody skips.
Filtering that reflects how buyers actually decide. If people choose by size, filter by size. If they choose by occasion, filter by occasion. The filters are a merchandising decision informed by your own order history.
Links to the related collections and to the guides. A collection page is a hub, and it should send people onward rather than being a dead end.
Supporting content below the grid, where a longer explanation belongs. Sizing, materials, care, how to choose. This is where the question queries in your list can be answered without cluttering the top of the page.
And genuinely useful images. Category-level imagery, not only product thumbnails.
How to decide which collections to build
This comes straight out of the keyword sorting exercise, and it should be a short list rather than a long one.
One collection per category query you identified as reachable. If nobody searches for a category, it does not need a page for search reasons, though it may still deserve one for navigation.
Do not build fifty collections for a store with thirty products. Thin collections with two items each are worse than a smaller number of substantial ones, both for search and for a shopper.
Cross-cutting collections are legitimate and often underused. A product can belong to several: by material, by price, by occasion, by recipient. Those cross-cuts frequently match the way people search better than your primary category structure does.
Be careful with automatically generated combinations. Store platforms make it easy to create a page for every combination of filters, and most of those pages have no text, no purpose and near-identical content. They dilute rather than add, and they are worth controlling deliberately.
The seasonal and gift collections
Worth a separate note because they are among the most valuable and the most temporary.
A collection built for an occasion can capture a query that is enormously more valuable than a general one during its window. The mistake is to build it two weeks before, which is too late for a new site, and to delete it afterwards, which throws away whatever the page accumulated.
Build it early, keep it, and update it each year rather than creating and deleting. A page that has existed for two years is in a different position from one created last week.
What this means for your work order
If you have limited time, and everyone does, the ordering that most reliably helps is:
Collections first, because each one covers a range of queries rather than one.
Then the product pages that matter, meaning the ones people actually buy, written properly rather than transplanted. Our guide to turning a marketplace listing description into a product page covers what has to change in that transplant.
Then the question content, drawn from your own buyer messages, linked from the collections it belongs to.
And not a redesign. A store with well-built collection pages and a plain theme beats a beautiful store with empty category grids.
The honest note
This work is slow, unglamorous, and it does not produce a visible result for a while. It is also the difference between a store that can only ever be found by people who already know its product names and one that can be found by people describing what they want.
That difference is a large part of what owning the store is for, and whether the trade is worth it for your business is the question our pillar on whether you should move at all works through. If your own numbers say the marketplace's category traffic is doing something you cannot replace, that is a real finding, and we have written the case for staying at full length.