You download your supplier’s feed, open it and see a name, a price and a photo. And where the product description should be, there is an empty field or a couple of sentences. Supplier product descriptions often need filling in before they are fit for a product page. In clothing and footwear, the size chart is missing too. This article shows which of these gaps can be filled automatically, which remain the shop’s job, and in what order to tackle them before your products go on sale.
You have the supplier feed, and it has gaps
Missing descriptions and missing size charts are two of the three gaps we see most often in supplier product feeds. The third is missing stock levels, but that is a separate problem, because no tool on the shop’s side can make up for it.
This article is for you if you have already chosen a supplier and are working with their feed. If you are still comparing suppliers, start with our guide to dropshipping supplier data quality – it covers what to check in the feed before you start working with a supplier.
When the feed has gaps, it is tempting to list the products now and add the descriptions later. With a dozen or so items, that works. With several hundred, "later" rarely comes, because there is always something more urgent. And a product page with nothing but a name and a price tells the customer very little and looks exactly like the pages in every other shop selling the same item.
Sort the gaps first – each one needs a different fix
"The feed has no descriptions" actually covers several different situations. Before you fill anything in, go through the feed and split the products into groups:
- no description at all – there is a name, a price, a photo, sometimes a few product attributes,
- a description that exists but is thin – one or two sentences, or text that goes out unchanged to every shop selling that product from the same supplier,
- a description in a foreign language – common with suppliers based abroad,
- no size chart – mostly clothing and footwear.
The first three groups can be filled automatically. The fourth is different: the chart is missing on the supplier’s side, but if the supplier does not provide it, filling the gap falls to the shop. Even so, it takes far less effort than you might think.
No description: AI writes one from what is in the feed
Writing descriptions from scratch by hand does not scale. To put that in perspective: simply adding 1,000 products to a shop by hand takes roughly 160 hours. Writing descriptions from scratch comes on top of that.
That is why descriptions for larger catalogues are generated automatically. What matters is what the description is built from, because that decides whether it is specific or generic. In our service, a description is built from four sources:
- data from the supplier feed – the product name, features and attributes, category, and the source description if there is one,
- the shop’s knowledge of the category – what matters to customers in a given category,
- SEO guidelines – so the description answers what customers are searching for,
- a description template – a shared structure filled in with the details of the specific product.
Product photos can also serve as an extra source – this matters most when the feed has no text at all. The description is then based on the attributes, category, name and what can be seen in the photo.
The same process also produces the page title and description shown in Google results (meta title and meta description). Across thousands of products, filling those fields in by hand is just as unworkable as writing the descriptions.
Thin descriptions, or the same as everyone else’s
Supplier product descriptions go out unchanged to every shop selling those products. Even if the text is correct, your product page then does not stand out – not to customers and not to search engines.
A thin or duplicated description goes through the same process as a missing one: the source text becomes one of the inputs rather than the finished copy. You can supply the keywords yourself if you have your own SEO strategy, or use the suggested phrases.
For more on how AI descriptions relate to SEO, see our article Are AI-generated product descriptions good for SEO?, and for choosing phrases across a whole catalogue, our article on keywords in product descriptions.
Descriptions in a foreign language
A description in German or Polish is no reason to drop a supplier. Translation can happen as part of synchronisation, so the product arrives in your shop in English straight away. Why a basic translation tool falls short on a product catalogue, and what works better, is covered in our article on translating product descriptions.
What AI will not fill in – and why that is a good thing
Generating descriptions has one limitation worth knowing about from the start. AI describes a product based on data. If the feed has no dimensions, composition or capacity, there is nowhere to get them from – and they should not be guessed.
That is a good thing, because a description with an invented specification is worse than no description. The customer buys on the strength of it, and if the specification is wrong, the product comes back.
So if technical data is missing from the feed, fill it in at the source. Ask your supplier for the missing specifications or check them on the manufacturer’s website. The fuller the input, the more specific the description.
Second, a person decides what goes into the shop. You can edit a generated description, approve it, or send it back to be regenerated with your own comment, such as "say more about how it is used" or "shorter".
Size charts: no automation will do this, but you only need to do it once
A size chart is a different kind of gap. It contains a specific manufacturer’s actual measurements, so it cannot be generated from a product name or a photo. Without it, the customer has to guess their size, and every wrong guess is a potential return.
The good news: you do not need a separate chart for every product.
Start with your supplier. Sometimes the supplier has a size chart but simply does not include it in the product feed. Ask whether they have size charts for the brands you sell. If not, check the manufacturer’s website – many clothing and footwear brands publish their own size charts.
Prepare one chart per brand or category. Sizing is often shared across a whole brand or product line, though not always – check this for each brand. Instead of hundreds of charts, you usually end up with a handful or a dozen or so, depending on how many brands you carry.
Assign it to many products at once. Popular e-commerce platforms have apps, plugins or modules that let you assign one size chart to many products, or a whole category, in one go. Check what is available for your platform.
To be clear: automatically attaching size charts is not a standard feature of the integration. Whether and how it can be done depends on the data the supplier provides and on what the shop needs.
What to do before you list your products
A short checklist:
- Sort the gaps in the feed – no description, thin description, foreign-language description, no size chart.
- Ask your supplier for the missing data – technical specifications and size charts.
- Prepare size charts – one per brand or category.
- Have descriptions generated from the data – then review them and approve them or send them back for changes.
- Only then list the products.
The order matters. If you fill in the specifications with your supplier first, the descriptions are built from fuller data and need fewer changes.
Summary
Missing descriptions in a supplier feed are no reason to drop the supplier or to spend weeks writing by hand. Missing and thin descriptions can be generated from the data that is in the feed, foreign-language descriptions translated as part of synchronisation, while a person decides what goes into the shop. Size charts stay with you, but you only need to prepare them once per brand or category.
If your catalogue is missing descriptions, find out more about our AI product description service.