As of 13 August 2026, AI can only partly set realistic sales targets for your team.

What to actually do

1. Use a tool built for this

   The route this page recommends

2. Hand it to a person

   Second choice

   A person who owns the outcome does this end to end, worth it when the failure is dear.

3. Do it yourself

   The distant third

   A chat interface, power-user skill, and roughly 1 hour until you can act on the result.

   How to actually do it

   1. Open your CRM and export the last comparable planning periods, including owner or role, opportunities, amounts in GBP, stages, outcomes, close dates, source and sales cycle where available.
   2. Open the current pipeline report and record its date, stage definitions, opportunity values, expected close dates, ownership and any known data-quality problems.
   3. Gather the business plan, revenue requirement, team structure, working days, planned leave, territories, role expectations, product mix and any confirmed headcount changes.
   4. Remove unnecessary personal data from the exports, label every period and currency, and paste the cleaned data and business context into the prompt.
   5. Ask the model for the three scenarios, calculations, assumptions, missing data and risks specified in the prompt.
   6. Create a spreadsheet containing the model's calculations, then compare each proposed target with historical attainment, conversion rates, sales cycle, available selling time and the actual capacity of each role.
   7. Ask your sales leader and finance or revenue operations colleague to challenge the assumptions, approve the scenario and record which targets are finally adopted.
   8. Send the approved targets and their definitions to the team, including the measurement period, exclusions, review point and how changes in territory or capacity will be handled.

   Prompt

   You are helping me propose realistic sales targets for a UK sales team. Do not present your recommendation as a fact or guarantee. Use only the information I provide, show your calculations, state every assumption, and flag missing or unreliable data.
   
   Business context:
   - Planning period: [month, quarter or year]
   - Revenue target or business objective: [details]
   - Sales motion: [new business, account management, renewals or mixed]
   - Team structure and roles: [details]
   - Working days, holidays and planned leave: [details]
   - Expected changes in headcount or territory: [details]
   - Average deal value: [amount in GBP, if known]
   - Sales cycle: [length and stages, if known]
   - Required margin or product mix: [details, if relevant]
   
   Historical data, with the period and definitions clearly labelled:
   [paste CRM export or table here]
   
   Pipeline data as at [date]:
   [paste pipeline totals, stages, values, close dates and owner data here]
   
   Analyse the data and produce:
   1. A recommended target for each role or salesperson, in GBP and in deal or activity terms where the data supports it.
   2. A base case, conservative case and stretch case, with the assumptions and calculations for each.
   3. A comparison with historical attainment, conversion rates, sales cycle and available capacity.
   4. A list of data-quality problems, missing inputs and reasons not to trust any calculation.
   5. The risks of setting the targets too high or too low.
   6. Questions I must answer with my sales leadership team before approving the targets.
   7. A short explanation I can share with the team that does not imply the targets came from an infallible model.
   
   Do not invent benchmarks, conversion rates, market conditions or individual performance claims. Do not recommend targets for a salesperson where the supplied evidence is inadequate. Keep personal information to the minimum needed and identify people by role or anonymous ID where possible.

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