Guide
Site search analytics for distributors: the five reports that matter
Site search analytics is the record of what visitors type into a storefront's search box, what they were shown, and what they did next. On a distributor storefront it is the most direct demand signal the business owns, because buyers type exactly what they intend to order. This guide covers the five reports worth reading and what to do with each.
On this page
- Why search data is a demand signal, not a web metric
- Report 1: Zero-result searches
- Report 2: Popular and rising queries
- Report 3: Conversion by query
- Report 4: Code versus word searches
- Report 5: Query to catalog gap
- How to run a weekly search review
- Privacy and data handling
- Where the data comes from matters here too
- Frequently asked questions
Why search data is a demand signal, not a web metric
Search data on a B2B storefront is a demand signal because every query is a purchase intent stated in the buyer's own words or codes. Retail analytics tells you what people browsed. Distributor search analytics tells you what customers tried to buy, including the things you do not stock.
Three patterns are worth watching every week:
- A zero-result search for a code you do not carry is a request for a new product, from a customer who already wants it.
- A spike in searches for a brand is a demand forecast, usually a week or two ahead of the order spike.
- A high-search, low-order query is a pricing, availability or relevance problem on a product buyers clearly want.
Report 1: Zero-result searches
The zero-result report lists every query that returned nothing, grouped by normalised text, ranked by frequency, with the accounts that ran it. It is the first report to open and the one with the most direct actions. Every row is one of three things.
| Cause | Example | Action | Owner |
|---|---|---|---|
| Product not stocked | geek bar pulse x | Add to the purchasing review list | Purchasing |
| Code or alias not mapped | EB-4420-BR | Map the alias in the ERP item master | Catalog |
| Synonym missing | RAW cones 1 1/4 | Add the synonym to the search configuration | Ecommerce |
Set a zero-result target for your catalog and review the trend. A benchmark across Shopsense storefronts will be added when it has been measured. Treat the report as a weekly queue rather than a dashboard. Every row should leave with an owner.
Search report
Sample dataTop queries
vs word
Zero-result queue
- geek bar pulse xPurchasingStock it
- EB-4420-BRCatalogMap alias
- RAW cones 1 1/4EcommerceAdd synonym
Report 2: Popular and rising queries
The popular queries report ranks queries by volume over a period. The rising queries report ranks them by week-over-week growth. The second is more useful than the first. Popular queries confirm what you already know. Rising queries show what is about to be popular, which is the signal purchasing needs.
Filter both by customer group or region. A flavour trending in one state and not another is actionable in a way a national total is not.
Report 3: Conversion by query
Conversion by query shows, for each query, how many searches led to a click, an add-to-cart and an order, and the revenue attributed. Sort by lowest conversion among high-volume queries. Those are the result pages that need work.
Low conversion on a high-volume query usually has one of four causes: the right product is ranked too low, the product is out of stock, the price is uncompetitive, or the buyer wanted a variant that is not carried. The report narrows it to a short list. The stock and price data on the result rows usually settles it.
Report 4: Code versus word searches
The code versus word report shows what share of searches are SKUs, part numbers or barcodes rather than product words, and how each converts. Distributors are often surprised by how high the code share is. The report tells you where to invest. A high code share with low code conversion means alias and barcode fields are incomplete in the ERP. A high word share with low conversion means synonyms and attributes need work. See the guide to SKU and part number search.
Report 5: Query to catalog gap
The catalog gap report combines zero-result queries with rising queries to produce one ranked list of products buyers want that you do not stock, weighted by how many accounts asked. It is the report to hand to purchasing once a month. It turns the search log into a buying list with evidence attached.
How to run a weekly search review
A weekly search review takes 30 minutes and three people: ecommerce, catalog and purchasing. The agenda does not change.
- Open the zero-result report. Assign every row above the volume you care about to an owner.
- Open rising queries. Flag anything that grew sharply week over week to purchasing.
- Open conversion by query. Pick the three worst high-volume queries and diagnose each from the result rows.
- Record the actions. Next week, check whether the fixed queries left the reports.
Privacy and data handling
Search analytics should tie queries to the business account, not to the individual person typing, unless there is a clear need and the privacy policy says so. Retention should be stated on the plan. Shopsense attributes queries to the account. User-level reporting is opt-in. Retention periods are not published on this site until they are fixed, and there is no privacy page to align them with yet.
Where the data comes from matters here too
Analytics is only as good as the stock and price data on the result rows. If the search index holds a synced copy, the conversion report cannot tell whether a buyer left because the item was out of stock or because the storefront wrongly said it was. When the index reads the ERP directly, the analytics inherit the truth. Shopsense is a B2B ecommerce search engine for distributors and wholesalers that connects directly to ERP systems. See the guide to real-time inventory and pricing.
Frequently asked questions
Keep reading
Next step
See Shopsense on your own catalog
Send us a product export or connect a sandbox ERP and we will show you live search on your data.