As of 13 August 2026, AI can create a supplier performance scorecard.

What to actually do

1. 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.

2. Use a tool built for this

   Second choice

3. Do it yourself

   The distant third

   A chat interface, chat-fluent skill, and roughly 30 minutes until you can act on the result.

   How to actually do it

   1. Open the latest supplier records and collect the measurement period, supplier identifiers, delivery results, quality results, pricing information, response times, lead times and service incidents.
   2. Open the relevant supplier contracts, purchase terms or internal policy documents and copy the agreed targets, service levels, scoring bands and any existing weighting rules into one working document.
   3. Remove unnecessary personal or confidential information from the working data, retain supplier identifiers needed for comparison, and separate each supplier's records clearly.
   4. Paste the working document into a chatbot with the prompt above and ask it to produce the scorecard, formulas, assumptions and validation checklist.
   5. Copy the generated table into your spreadsheet and compare every input, calculation and overall score with the original supplier records and agreed targets.
   6. Ask a colleague who knows the purchasing process to check the proposed weights, missing-data treatment, ranking and interpretation before sending the scorecard to suppliers or using it for a decision.
   7. Record the approved scoring rules and the review date in the scorecard, then send the final version to the people responsible for supplier performance management.

   Prompt

   Create a supplier performance scorecard from the information below.
   
   Business context: [describe what we buy and what good supplier performance means]
   Suppliers: [list supplier names or IDs]
   Measurement period: [start date to end date]
   Source data: [paste delivery, quality, price, responsiveness, lead-time, service or other records]
   Agreed targets: [paste the target for each measure]
   Scoring rules: [paste any existing scoring bands or explain the scale, such as 0 to 5]
   Weights: [paste the percentage weight for each measure; if none exist, propose a neutral draft and label it as a proposal]
   Audience: [name the people who will use the scorecard]
   
   Use only the supplied information. Do not invent measurements, targets, weights, supplier names or explanations. If a required value is missing, mark it as missing and list the exact data needed. Keep supplier IDs where confidentiality is important.
   
   Return:
   1. A clear table with one row per supplier and one column for every measure.
   2. The formula or rule used for every score and the overall score.
   3. A ranking only where the data and rules support a fair comparison, otherwise state why ranking is not valid.
   4. A short explanation of each supplier's strongest result, weakest result and any missing data.
   5. A separate list of assumptions and proposed rules that I must approve before use.
   6. A validation checklist showing how a colleague can compare every score with the source data.
   Use plain British English and do not present a draft rule as an agreed business decision.

   Open it prefilled in ChatGPT or Claude, or copy it into Gemini, which takes no prefill link.