Foundations
Part of Shopping feeds basics: the fields, channels and marketplace fees behind a UK catalogue
Before you read shopping feed demand signals, check the basics
Shopping feed demand signals are easy to misread. This guide sets out five checks, the England and UK scope, and a worked example with labelled pound figures.
What to take away
- The most common mistake is treating a single metric as a demand signal. Feed demand signals only mean something when two or more sources point the same way at the same time.
- Start with scope. Most published retail and ecommerce data from the Office for National Statistics covers the UK as a whole, so do not label it England-only.
- Set inclusion criteria before you rank anything. Decide which signals count, which periods you compare, and which product categories you exclude.
- Compliance is part of demand. Google Merchant Centre terms and the ICO's business-to-business marketing rules both shape what you can promote and to whom.
- Build a written signal checklist, then test it against one category for a quarter before you widen it.
Where demand signals come from
A demand signal is evidence that shoppers want a product now or soon. Feed-level signals include impressions, clicks and conversion rate by item. Search-level signals include query volume for the product type. Stock and returns data add a third layer.
Three demand signal layers
Feed-level
- What it shows
- Impressions, clicks, conversion
- Typical misread
- Bid change spike
- Read with
- Search and stock data
Search-level
- What it shows
- Query volume by product type
- Typical misread
- Intent not yet in feed
- Read with
- Own impression share
Stock and returns
- What it shows
- Availability and return rates
- Typical misread
- Stockout looks like lost interest
- Read with
- Conversion by item
No single layer is reliable alone. A spike in impressions can come from a bid change rather than real interest. A fall in clicks can come from a feed attribute error. Read them together.
For market context, UK business and ecommerce statistics from the ONS are a sensible starting point. Check the release notes, because many series are UK-wide rather than England-only.
Set your inclusion criteria
Before comparing anything, write down what qualifies as a demand signal for your catalogue. Use these tests.
Five inclusion criteria for signals
- Defined period, such as rolling 28 days
- Traceable to a product identifier in feed
- Comparable across same category and season
- Survives check against stock and returns
- Named owner reviews it each month
Geography matters here. If you sell only into England, say so, and treat UK-wide data as context rather than proof. If you sell across Great Britain, note that Scotland and Wales sit in the same data series but may behave differently.
Five signals worth tracking
Feed health signals
Disapprovals, missing GTINs and image failures suppress visibility. A clean feed is the baseline, not a demand signal in itself. Review Google Merchant Centre terms of service for the data quality and account compliance rules that apply.
Search demand signals
Query volume for a product type shows intent before it reaches your feed. Compare it with your own impression share to see whether you are capturing that demand or losing it.
Conversion signals
Conversion rate by item shows whether demand converts once seen. Pair it with average order value to avoid chasing high-volume, low-margin lines.
Stock and returns signals
Repeated stockouts on a rising item point to unmet demand. High return rates on the same item point to a mismatch between the listing and the product.
Trade and B2B signals
If you run a trade catalogue, business-to-business outreach has its own rules. The ICO explains business-to-business marketing requirements that apply to how you contact trade buyers.
Worked example with pound figures
A homeware seller lists 400 lines and compares the last quarter with the same quarter a year earlier. Impressions rise 18 per cent, clicks rise 9 per cent, and conversion rate falls from 2.4 per cent to 1.9 per cent.
Average order value stays at £55. Revenue per 1,000 impressions falls from £1,320 to £1,045. That is a drop of £275 per 1,000 impressions.
The seller checks stock and finds 30 lines out of stock for part of the quarter. Those 30 lines account for most of the conversion fall. The demand signal was real, but the constraint was availability, not interest.
These figures are illustrative. Substitute your own feed exports before drawing conclusions.
Comparing business models
Signal priority changes with how you sell. A marketplace-led seller watches feed health and conversion. A wholesale seller watches trade enquiries and repeat orders. The guide to shopping feed business models in England sets out how these differ before you pick metrics.
Whichever model you use, keep one shared definition of a qualified signal. Otherwise teams argue about numbers rather than acting on them.
Common questions
Do these signals apply across the UK?
Mostly, yes for feed mechanics, since platform rules apply UK-wide. Retail data often covers the UK, so check each release before labelling it England-only.
How many signals should I track at once?
Three to five is workable. More than that and review becomes a reporting exercise rather than a decision tool.
How often should I review them?
Monthly for feed health and quarterly for demand trends. Seasonal categories may need a shorter cycle.
Where does the market guide fit?
The England market guide for 2027 covers scope and definitions in more depth once you have a working checklist.
