AI and automation
From Spreadsheet to B2B Product Catalogue
Prepare B2B product data for a searchable catalogue with clear variants, consistent specifications and better quote requests.

Your sales team knows which pump fits a customer’s requirements. A buyer looking at the spreadsheet sees three nearly identical product names and a column called “Size.” Before choosing catalogue software, decide what the buyer needs to compare and which product details your team can reliably maintain.
Start with a product family buyers already ask about
Choose one family with a manageable number of products and real differences between them. Ask the people who answer quote requests which details customers ask for repeatedly. For a supplier of industrial equipment, those might be dimensions, operating conditions, compatible accessories and whether installation is available.
Bring together the spreadsheet, supplier data sheets and current sales material. Give conflicting values an owner. A specification copied into three documents should have one reviewed source before it appears on a product page.
Distinguish the product from the orderable variant
A product family explains what the item does. A variant records the configuration somebody can request, such as a particular size, material or connection. Use a stable identifier for each variant. Preserve supplier identifiers separately where the purchasing team uses them.
Put units and their meanings in the data model. If the spreadsheet contains a value of 50, decide whether it represents millimetres, litres or a nominal size before anyone uses it to filter results. Keep unknown values distinct from zero and from attributes that do not apply.
Build filters around actual selection criteria
Start with the attributes that change the buying decision. A filter for mounting type may be useful even if it only has two values. A long list of supplier codes may be useful for exact search and distracting as a filter. Normalise synonyms where they mean the same thing and retain meaningful differences.
Put the most important specifications in the page itself. Keep the technical PDF available for details such as drawings and installation instructions. Label the document with a version or date if buyers need to know which specification they are using.
Carry the chosen product into the quote request
When a visitor requests a quote, preserve the selected variant, the product page and the quantity if supplied. Give them a way to add their application or ask for help choosing. The receiving team should see the same product reference the buyer saw.
Agree which values can be published. When availability changes frequently and your website has no verified stock feed, explain how availability is confirmed. If pricing depends on volume or configuration, state what the sales team needs to prepare a quote.
Test a small import before expanding
- Import one family with several variants and a deliberately incomplete record.
- Ask a colleague to find a product using the terms a customer would use.
- Compare the displayed specifications with the approved source.
- Send a marked quote request and check the receiving record.
- Update one specification and confirm where the change appears.
Record who owns data corrections and how often the catalogue is reviewed. A pilot exposes the amount of product data work required before the project commits to a full import.
CONSAI designs digital platforms and enquiry journeys. If your product information sits across spreadsheets and PDFs, share one product family and a sample quote request.



