Lead scoring rubric prompt: free ChatGPT template
Updated · By Robert Breen
Use this when your team argues about which leads deserve a call today and you want a written scoring system instead of gut feel. The prompt compares deals you won with deals you lost, picks out signals that show up on a form (team size, timeline, how they work today) and turns them into a points rubric with a cutoff for a fast reply. Rating each lead against the rubric is a separate job, covered by the lead qualification prompt.
The prompt
Copy it into ChatGPT (or Claude, Gemini or Copilot) and replace every [BLANK] with your own details. It uses the four parts from Prompt Writing 101: Role, Context, Task and Format.
Role: You are a sales operations analyst designing a lead scoring rubric that a busy rep can apply in under a minute. Context: We are [COMPANY] and we sell [WHAT WE SELL]. Facts about deals we won: [WON DEALS]. Facts about deals we lost or that went quiet: [LOST DEALS]. Fields we collect on every new lead: [LEAD FIELDS]. Task: Using only the facts above, find the signals that separate won deals from lost ones. Build a rubric of four to six signals, each scored from our lead fields, with points and one sentence on why, citing the deals behind it. Suggest a score for a same-day reply and a score below which the lead gets a nurture email. Do not invent conversion rates. List signals you would want but can't score from our fields. Format: A table with the columns: signal, how to score it, points, why (with the deals it came from). Below it, the two cutoffs in one line each, then a short list titled "Fields to add". Plain language, no emoji.
Fill in the blanks
[COMPANY]- Your company name.
[WHAT WE SELL]- The product and who buys it, in a sentence.
[WON DEALS]- Five to ten recent wins, one line each: size, timeline, how they found you, what they used before. No names needed.
[LOST DEALS]- The same kind of lines for deals you lost or that went silent. Without these the model can't tell a signal from noise.
[LEAD FIELDS]- The questions on your form or intake sheet, exactly as they appear.
Example, filled in
A made-up example from the free lesson Build an AI Demo Request Form That Qualifies Leads. Northwind Scheduling Software is the made-up company from the demo request form lesson, where the form asks for name, email, what they need help with and team size. The won and lost deal notes below are made up for this example.
Role: You are a sales operations analyst designing a lead scoring rubric that a busy rep can apply in under a minute. Context: We are Northwind Scheduling Software and we sell online scheduling software to small service businesses like cleaning companies, salons, lawn care crews and repair shops. Facts about deals we won: a cleaning company with 12 cleaners on a spreadsheet that kept breaking, wanted to switch this month; a lawn care crew of 8 missing jobs from double-booking; a salon with 6 stylists wanting to start before the holidays; a repair shop with 5 techs after a trial. Facts about deals we lost or that went quiet: a solo groomer just looking; a 3-person cleaning team who asked only about price; a salon of 9 with no timeline; a student researching for a class. Fields we collect on every new lead: Name, Email, What do you need help with?, Team size. Task: Using only the facts above, find the signals that separate won deals from lost ones. Build a rubric of four to six signals, each scored from our lead fields, with points and one sentence on why, citing the deals behind it. Suggest a score for a same-day reply and a score below which the lead gets a nurture email. Do not invent conversion rates. List signals you would want but can't score from our fields. Format: A table with the columns: signal, how to score it, points, why (with the deals it came from). Below it, the two cutoffs in one line each, then a short list titled "Fields to add". Plain language, no emoji.
What a good answer looks like
- Team size of 5 or more and a stated timeline score highest, and the "why" column points to the cleaning company and the salon.
- A signal for a named pain (spreadsheet breaking, double-booking) that the rep scores from the "What do you need help with?" answer.
- "Asked only about price" or "just looking" earns zero or negative points, tied to the lost deals.
- "Fields to add" suggests things like timeline or current tool, because the form doesn't ask for them yet.
How to check it before you use it
An AI draft can sound right and still say something you never told it. Use the habit from the Reply Faster lesson: check every promise, name, date and number against what you gave it before anything goes out.
- Eight deals is a small sample. Treat the rubric as a first guess and look at it again after the next month of leads.
- Make sure every signal can be scored from the form alone. If a rep would have to call first, it belongs in discovery.
- Score five recent leads by hand. If a lead you know is weak lands above the cutoff, adjust the points.
Practice it in the free lesson
Build an AI Demo Request Form That Qualifies Leads: Build a "Book a demo" form in Lovable for Northwind Scheduling Software and connect it to n8n — the AI agent rates each lead hot, warm or nurture, saves it to Google Sheets and emails the right next step. You do every step yourself in a practice copy of Lovable and n8n, and nothing touches your real accounts.
Also useful: Build an n8n AI Agent That Writes Deal Follow-Ups · Prompt Writing 101: Role, Context, Task, Format
Terms used here
Related prompts
- Lead qualification prompt: Sort leads into Hot, Warm and Nurture with your own written rules. This ChatGPT prompt quotes the words behind each rating and…
- Ideal customer profile prompt: Describe the companies you should chase, built only from facts about your best current customers, with the traits they share and…
- Weekly leads summary prompt: Paste a week of form leads from your Google Sheet and get counts by type, who is still waiting, patterns with row numbers and…
More sales prompts
- LinkedIn connection request prompt: Write a short LinkedIn connection note with one real reason to connect and no pitch, plus a useful first message for after they…
- Maintenance agreement renewal prompt: A renewal email for a maintenance plan member: the visits they really had, the renewal terms stated plainly, and canceling made…
- Objection handling prompt: Prepare honest answers to sales objections with a ChatGPT prompt: ask a question first, use only your facts and limits, and know…