Skip to content
ToolDeck

Sales & Lead Capture

Lead Scoring by Fit

Every inbound lead is scored against your ideal profile — so you call the right ones first.

Saves about ~2 hrs per week

What it does

  • Scores each inbound lead against your ideal customer profile
  • Weights signals: role, company size, budget words, urgency
  • Outputs a 0-100 fit score with the reasons
  • Ranks the day's leads so you follow up in order

How it works

  1. New lead lands in the sheet/CRM
  2. AI reads the lead data against your ICP rubric
  3. Fit score + reasons written back to the row
  4. Daily ranked list posted or emailed

What's included

  • ICP rubric
  • Scoring prompt
  • Daily ranked list

Set it up yourself

Follow these 4 steps in order. You will need accounts for: Claude, Sheets.

Step 1: Write your ICP rubric

  1. Write your ideal profile as bullet statements: target roles, industries, company sizes, budget language, urgency signals.
  2. Rank them by weight (must-have vs nice-to-have).
  3. Create the rubric as a reusable table or text block in the workflow.
  4. Include an anti-rubric: disqualifiers that tank a score.
  5. Test the rubric against your last 10 won and 10 lost deals — it should separate them.

Watch out: Your successful deals are the ground truth — if the rubric says 'small agency' and you actually close 'enterprise', fix the rubric.

Step 2: Connect the lead source

  1. Feed the workflow from the same source your team uses: form export, CRM, or a leads sheet.
  2. Map fields so the AI sees email, role, company, size, and any free text.
  3. Add a column 'Fit Score' and 'Fit Reason' to the destination.
  4. Test the wire with one lead.

Watch out: Free text (the 'what do you need' answer) carries the most scoring signal — make sure it reaches the AI intact.

Step 3: Add the scoring prompt

  1. Build: 'Score this lead 0-100 against this ICP rubric. Return JSON: {score, strongSignals[], weakSignals[], recommendedAction(reply-now | nurture | drop)}.'
  2. Instruct: score 0 if any disqualifier matches.
  3. Ask for the top 3 reasons — scores without reasons are useless for your judgment.
  4. Paste with your API key.
  5. Run against 5 past leads and compare with what your team actually did.

Watch out: When the AI says 'nurture', define that as a real action (add to follow-up queue) or the score just gets ignored.

Step 4: Schedule the ranking

  1. Run the scoring daily each morning before follow-ups start.
  2. Post the ranked list (score desc) with the top 5 highlighted.
  3. Ship the list to the same channel your team already watches.
  4. Review 'score vs outcome' monthly to keep the rubric honest.
  5. Iterate the rubric once a quarter or after 50 leads.

Watch out: Calibration loop: if your team ignores a low score and wins the deal, the rubric — not the lead — was wrong. Log those.

Frequently asked questions

What does the Lead Scoring by Fit automation do?

Every inbound lead is scored against your ideal profile — so you call the right ones first. Scores each inbound lead against your ideal customer profile

How much time does Lead Scoring by Fit save?

It is estimated to save ~2 hrs per week. Actual savings depend on your volume.

Which tools does Lead Scoring by Fit use?

It works with Claude and Sheets.

What does Lead Scoring by Fit include?

ICP rubric, Scoring prompt, Daily ranked list.

Can I set up Lead Scoring by Fit myself?

Yes. The do-it-yourself guide on this page walks through 4 steps, from getting access to testing the finished automation.