As of 13 August 2026, AI can clean your 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 30 minutes until you can act on the result.

   How to actually do it

   1. Export the prospect list from your CRM or spreadsheet, retaining a stable row identifier and the fields you use for outreach.
   2. Remove columns that are not needed for cleaning, and confirm that your organisation permits the chosen AI service to process the remaining prospect data.
   3. Write down your rules for duplicates, company-name variants, job-title formatting, country and phone formats, and which fields must never be changed automatically.
   4. Paste the list and those rules into an approved chatbot using the prompt, then ask it to return the cleaned CSV, change log, duplicate report and NEEDS REVIEW list.
   5. Compare the returned row count and excluded rows with the original export, then inspect every proposed merge and every deletion against the original records.
   6. Sample changed names, companies, job titles and contact details against your CRM or an approved current source, and reject any change that cannot be supported.
   7. Save the original export and change log, import only the approved cleaned file into a test view or duplicate CRM list, and check that filters and required fields still work before using it for outreach.

   Prompt

   Clean the prospect list below without inventing or silently deleting information.
   
   List:
   [PASTE THE CSV OR TABLE HERE]
   
   Cleaning rules:
   - Preserve the original row identifier for every row.
   - Standardise spelling, capitalisation, spacing and date or phone formats only when the intended value is clear.
   - Identify exact duplicates and likely duplicates separately.
   - Do not merge records solely because names or company names look similar.
   - Do not guess missing email addresses, job titles, companies, websites or contact details.
   - Flag invalid-looking or incomplete values rather than correcting them by guesswork.
   - Keep the original value beside every changed value.
   - Do not claim that an email address, person, company or job title is current unless the list includes a source and date for that claim.
   
   Return:
   1. A cleaned CSV in the same general structure.
   2. A change log showing the original row identifier, field, original value, new value and reason.
   3. A duplicate report with exact duplicates, likely duplicates and records that must not be merged.
   4. A list of missing, invalid-looking or ambiguous fields.
   5. A short summary of row counts before and after, including any rows excluded from the cleaned file.
   
   Use plain, consistent values and mark anything requiring human approval as NEEDS REVIEW. Do not include personal data from outside the supplied list.

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