StorePilot Store operations setup Store Operations Setup · USD 129
Two hundred and ten products, and 88 of them cannot be found
Product taxonomy, operating calendar and promotion workflow, built against the catalogue as it actually is.
The catalogue has grown by 140 products in two years with no rules about how a product gets listed. This sets those rules and applies them, and the first pass found 88 products that are effectively invisible to anyone browsing or searching your own store.
Eighty eight products are missing a category, a material attribute or a searchable name, which means they appear only if a customer already knows the exact product title. Twenty nine of those 88 have never sold a single unit.
Catalogue
What is actually wrong with the listings
| Group | Products | Revenue share | Action |
|---|---|---|---|
| Complete listings | 51 | 68% | Leave alone |
| One field missing | 71 | 24% | Fix in batch |
| Two or more missing | 59 | 8% | Fix the 30 that sell |
| Never sold, incomplete | 29 | 0% | Delist or complete |
Fifty one complete listings produce 68 percent of revenue. That is the argument for the taxonomy in one line: the products you listed properly are the products that sell, and it is not obvious which way the causation runs, but it is not worth 29 products sitting at zero to find out.
Taxonomy
The rules a product now has to meet
Eight required fields and one naming rule. A product that does not meet them does not go live, which is the only part of this that actually changes anything.
| Required | Why | |
|---|---|---|
| Category, one of 14 | Always | Without it the product exists only via search |
| Material | Always | Your second most used filter |
| Dimensions | Always | The most common support question |
| Title contains the product noun | Always | Astrid Serving Bowl, not Astrid |
| Three images minimum | Always | One in use, one scale reference |
| Care instructions | Where relevant | Reduces returns on textiles and wood |
| Collection tag | Where relevant | Enables cross-selling |
| Cost price | Always | Currently absent on 63 products, so margin is unknowable |
Cost price is the one nobody thinks of as a listing field. Sixty three products have none recorded, which means any margin analysis of your catalogue is currently guesswork on a third of it.
Promotions
Four running at once, and two contradict each other
| Promotion | Started | Ends | Overlaps | Issue |
|---|---|---|---|---|
| Welcome 10 percent | Permanent | No end date | Everything | Stacks with the others |
| Autumn 20 percent, kitchen | 28 September | 31 October | Welcome | Stacks to 30 percent |
| Free shipping over USD 60 | Permanent | No end date | Everything | Threshold below average basket |
| Bundle: board plus knife | 12 September | Never ended | Autumn | Still live, was for one month |
A customer buying a kitchen item today can stack the welcome code on the autumn sale for 30 percent off, with free shipping, on a bundle that was supposed to end in September. Nobody designed that. It is four decisions taken at four different times with nothing holding them together.
Plan
The first four weeks
- This week
Fix the stacking, today
Decide whether welcome stacks with seasonal. Either answer is fine. Thirty percent plus free shipping on kitchen is almost certainly below margin on some of those products, and on 63 of them you cannot tell.
- This week
End the September bundle
It was a one month promotion and it is in week five.
- Week 2
Fix the 30 incomplete products that do sell
Not all 88. The 30 with sales history, where a missing field is costing you something measurable today.
- Week 3
Decide on the 29 zero-sale products
Complete them properly or delist them. Nine have unsearchable names, which is the cheapest thing to test: rename, wait four weeks, then decide.
- Week 4
Backfill cost prices
Sixty three products. Tedious, and it is what makes every future margin question answerable.
- Ongoing
The standard applies to new products from today
This is the part that matters. Fixing 88 products once is housekeeping; the standard is what stops there being 88 again in two years.
- Stacking decision made and implemented
- September bundle ended
- Thirty selling products completed
- Nine unsearchable names rewritten
- Sixty three cost prices backfilled
- Listing standard applied to all new products