As of 13 August 2026, AI can only partly build a B2B prospect list.

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 1 hour until you can act on the result.

   How to actually do it

   1. Open a spreadsheet and create columns for company name, website, location, industry, fit evidence, buying signal, source, target job title, contact route, confidence and checking status.
   2. Write down your offer, ideal customer profile, exclusions, target number and buying signals, then gather relevant company websites, directories, trade sources and existing CRM exports.
   3. Paste the prompt and your sales information into a chatbot, attach or paste the permitted source material, and ask it to produce the ranked table without inventing missing details.
   4. Copy the returned rows into the spreadsheet and remove duplicate companies, existing customers, competitors and any organisation that matches your exclusions.
   5. Open each company website and source link, then compare the stated industry, location, size and buying signal against the evidence in the row and mark each fact as confirmed, changed or unknown.
   6. Check every proposed contact route and job title against the company website or an authorised business database, removing personal details that are not clearly published for business contact.
   7. Ask the chatbot to rewrite the surviving rows into a short outreach brief using only the confirmed facts, then have a colleague check the final list and your planned lawful basis and opt-out process before sending.

   Prompt

   Build a UK B2B prospect list from the information below.
   
   Our offer: [describe the product or service]
   Ideal customer profile: [industry, company size, location, business model and likely use case]
   Buying signals: [events, hiring, technology changes, regulation, funding, expansion or other observable signals]
   Exclusions: [industries, company sizes, locations or situations to exclude]
   Target number of prospects: [number]
   Available sources or URLs: [paste sources, directories, company pages or exported data]
   
   Return a table with these columns:
   1. Company name
   2. Company website
   3. UK location
   4. Industry
   5. Evidence that it matches the ideal customer profile
   6. Evidence of a buying signal, with the source and publication date where available
   7. Likely department or job title to contact
   8. Public business contact route, if available
   9. Confidence level: high, medium or low
   10. A short reason this company belongs on the list
   
   Use only information in the supplied sources or information you can clearly identify as current public information. Do not invent companies, people, job titles, contact details, buying signals or dates. Do not include personal email addresses or phone numbers unless they are clearly published as business contact details. Mark missing information as unknown. Deduplicate companies, separate fact from inference, and rank the list by fit and strength of evidence. After the table, list every item that needs manual checking before outreach.

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