Home · Business · Operations & Logistics · Suppliers & ordering
As of 13 August 2026, AI can compare product catalogues from different suppliers.
This still needs a person who signs their name to it.
Can you do it?
15 minutesto a draft.
1 hourto something you’d act on.
Cost, all in£0
Skill neededpower-user
Who has to check ita colleague
What the alternative costsParseur is an AI document extraction tool for pulling data from emails and PDFs; no alternative price is stated here.
If this goes wrong, you can order an unsuitable product or miss a price, pack-size or delivery difference and create avoidable cost or disruption.
What to actually do
Hand it to a person
The route this page recommends
A person who owns the outcome does this end to end, worth it when the failure is dear.
Use a tool built for this
Second choiceDo it yourself
The distant thirdA chat interface, power-user skill, and roughly 1 hour until you can act on the result.
How to actually do it
- Open the current catalogue or price list for each supplier and save the complete PDFs or spreadsheets, including specification pages, price pages and delivery terms.
- Gather the product categories, required specifications, delivery location, VAT treatment and purchasing constraints that matter to your business.
- Paste the prompt into a chatbot and attach the catalogues, replacing the bracketed requirements with your information.
- Ask the model to produce the comparison table, unmatched-product sections and list of ambiguous matches using only the attached catalogues.
- Open the original catalogues beside the result and check every product code, unit, pack size, price, VAT entry, lead time and minimum order against its source.
- Send the flagged ambiguous matches and any missing questions to the relevant suppliers, then add their written answers to the comparison.
- Ask a colleague who understands the products or purchasing rules to approve the final matches before using the table for an order.
Prompt
Compare the supplier catalogues I provide below. Do not assume that products are equivalent because their names look similar. Extract and standardise, where available: supplier, product name, supplier product code, brand, model, description, dimensions, materials, unit of sale, pack size, price, VAT status, delivery charge, minimum order, lead time, availability, warranty and relevant certifications. Create a table with one row per possible product match and separate columns for each supplier. Include unmatched products in separate sections. Show the original catalogue wording beside any normalised value. Flag missing, conflicting or ambiguous information rather than filling gaps. Distinguish differences in product specification from differences in commercial terms. Calculate no totals or savings unless the necessary figures and units are explicit, and show the calculation if you do. End with a list of questions I should ask each supplier before ordering. Treat the catalogues as the only source of facts. Supplier A catalogue: [PASTE OR ATTACH CATALOGUE] Supplier B catalogue: [PASTE OR ATTACH CATALOGUE] Optional comparison requirements: [ADD PRODUCT CATEGORIES, REQUIRED SPECIFICATIONS, DELIVERY LOCATION AND ANY TERMS THAT MATTER] Return the comparison in a format I can paste into a spreadsheet, followed by a short list of matches that need a human decision. Do not recommend a supplier and do not claim that two products are interchangeable without evidence in the catalogues.
Open it prefilled in ChatGPT or Claude, or copy it into Gemini, which takes no prefill link.
What it gets wrong
- AI cannot reliably decide that two differently described products are technically interchangeable when the catalogues omit a specification.
- It can merge similar product names while missing differences in grade, dimensions, pack size or intended use.
- It cannot know whether a supplier's price, stock position or delivery promise is still current unless you obtain confirmation.
- It does not carry responsibility for an incorrect order, so your purchasing process still needs a human approval point.
Even on a YES, the friction has a name: judgement under ambiguity, verification cost and stakes of error.
How we scored this
Five axes, each scored nought to two by hand: ten means AI carries the task cleanly, and the thresholds that turn a total into YES, PARTLY or NO are published in the methodology. Each axis name links to its definition.
| Axis | Score (0–2) |
|---|---|
| Output | 2 |
| Inputs | 2 |
| Verification | 1 |
| Liability | 1 |
| Effort delta | 2 |
| Total | 8 / 10 |
The methodology and its thresholds are published in full.
FAQ
- Can AI compare two supplier catalogues?
- Yes. It can extract fields from PDFs or spreadsheets, align likely matches and flag differences in specifications and commercial terms. You still need to check the source rows and resolve products that may only look equivalent.
- Can AI compare supplier prices in Excel?
- Yes, if the files clearly state the prices, units and pack sizes. Ask it to preserve the original values and flag missing VAT, delivery charges, minimum orders and different units before comparing costs.
- Can AI tell me which supplier has the better product?
- It can organise the evidence against criteria you provide, but it cannot safely judge an unstated quality or compatibility requirement. A product or purchasing colleague should decide whether the apparent matches meet your operational needs.
- Can AI compare PDF product catalogues?
- Yes, current models can extract catalogue text and tables from supplied PDFs, although scanned or poorly formatted pages may need an extraction tool or manual checking. Compare every important value with the original PDF before placing an order.
Nearby answers
- Can AI email a supplier to place an order?PARTLY
- Can AI find local suppliers for my UK business?YES
- Can AI reorder stock from my supplier?PARTLY
- Can AI write a complaint to a supplier about a bad order?YES
- Can AI check whether a supplier has a product in stock?PARTLY
- Can AI compare supplier minimum order quantities?YES
Assessed by gpt-5.6-luna (gpt-5.6-luna) on 2026-08-13, second-checked by an independent model. Wrong somewhere? Email [email protected] and it gets re-checked.
The newsletter
AI news, new answers and product picks, straight to your inbox.