Lead Scoring Model

Draft a documented lead scoring model grounded in your CRM's actual conversion history: weighted attributes, clear MQL and SQL thresholds, and a real lead scored as a worked example.

For Demand Gen Managers Marketing Ops Sales Ops & RevOps Startup Marketing Leads
Sample prompt to Joy's Assistant

Draft a lead scoring model for [product] using our CRM's closed-won data: weighted attributes on a 100-point scale, MQL and SQL thresholds, and one recent lead scored as a worked example.

What it can look like

An illustrative example. Your output will reflect your own sources, data, and prompts.

Draft a lead scoring model for our logistics operations platform using our CRM's closed-won data: weighted attributes on a 100-point scale, MQL and SQL thresholds, and one recent lead scored as a worked example.

I based the weights on the last 12 months of closed-won deals in your connected CRM: 214 conversions. Seniority and pricing-page visits predicted conversion far more strongly than industry or email opens, and the weights reflect that.

Lead scoring model: draft

The model scores every lead out of 100 points across six attributes. Seniority fit carries the most weight at 25 points, with pricing-page visits and demo or webinar activity at 20 each: those three signals appeared in over 80% of your closed-won deals.

Attribute weights: 100-point scale

Seniority fit25
Pricing-page visit20
Demo or webinar20
Company size fit15
Industry fit10
Email engagement10
  • MQL: 55 points or more. Routed to nurture with a sales notification. In your history, conversion jumps from 4% below this line to 19% above it. That breakpoint is why the threshold sits at 55.
  • SQL: 75 points or more. Routed to an AE for outreach within one business day; leads above 75 closed at 31%.
  • Below 55: stays in nurture. Re-scored whenever new activity lands in the CRM.

Worked example: a lead created Tuesday, scored line by line.

AttributePriya Nair, Ops Director at Meridian LogisticsPoints
Seniority fitDirector-level, operations: exact ICP match25 / 25
Pricing-page visitVisited pricing twice this week20 / 20
Demo or webinarRegistered for the webinar but didn't attend0 / 20
Company size fit240 employees, inside the 100–500 target band15 / 15
Industry fitAdjacent (3PL) rather than core freight5 / 10
Email engagementOpened 4 of the last 6 sends, one click6 / 10

Priya scores 71, a clear MQL, four points short of SQL. One demo booking or a webinar attendance pushes her past 75, which is exactly what her next nurture email should offer.

Want me to score this week's leads against the model, or test how many would qualify with a 60-point MQL threshold instead?

From CRM history to scoring model

Lead Scoring Model looks at your CRM's closed-won history to see which attributes actually preceded conversion, then drafts a full scoring model: weighted attributes on a 100-point scale, MQL and SQL thresholds set at real conversion breakpoints, and a worked example so everyone can see the math.

  1. Connect your CRM data

    Point Joy at your connected CRM or add a lead-history file to the Knowledge Center. What matters is which leads converted and what you knew about them beforehand.

  2. Ask for the model

    Name the product or segment and ask for a 100-point model with MQL and SQL thresholds. Add house rules up front ("competitors always score zero") if you have them.

  3. Review weights and thresholds

    Get the attribute weights as a chart, the thresholds with the conversion evidence behind them, and a recent lead scored line by line so the model is concrete, not abstract.

  4. Adopt it where you work

    Copy the documented model into your marketing automation platform's scoring rules and your team wiki, and use the worked example in the sales-marketing handoff conversation.

  5. Make it one click for your team

    Save this ask as a custom command on the assistant your team already uses, then customize it with your own sources and wording, so anyone on the team can run it in one step.

Make it yours

Evidence-Backed Weights

Each attribute's weight reflects how strongly it preceded conversion in your own data, not a template someone published in 2019.

Threshold Logic

MQL and SQL cutoffs land where conversion rates actually jump in your history, so the lines mean something.

Worked Examples

Any lead scored line by line on request, the fastest way to settle a 'why did this one qualify?' debate.

What-If Rescoring

Change a weight or threshold in the chat and see how last month's lead volume would have qualified under the new rules.

Account Scoring

Score accounts instead of individual leads for ABM motions, with intent and engagement rolled up per account.

Negative Scoring

Add point deductions for disqualifiers (students, competitors, free-mail domains) to clean the MQL queue.

Decay Rules

Define how behavioral points expire, so a pricing-page visit from March doesn't inflate a lead in September.

Model Audit

Run your existing model against recent closed-won data and find the weights that no longer earn their points.

Frequently Asked Questions

What is a lead scoring model?

A lead scoring model assigns points to a lead's attributes and behaviors (job title, company size, pricing-page visits) and uses thresholds to decide who is marketing-qualified (MQL) or sales-qualified (SQL). JoySuite drafts the model from your CRM's conversion history rather than generic templates.

How are the attribute weights decided?

Joy compares your closed-won deals against leads that never converted and weights each attribute by how strongly it separated the two groups. Every weight in the output comes with its evidence, so the model is something your team can inspect and challenge.

Where should the MQL and SQL thresholds go?

At the score levels where conversion rates actually jump in your history, not at round numbers. In the drafted model each threshold cites its breakpoint (for example, conversion rising from 4% to 19% at the MQL line), so the cutoffs are defensible.

Does JoySuite score leads inside my marketing automation platform?

No. Joy drafts and documents the model, scores individual leads on request in the chat, and re-scores examples when you change a weight. You copy the final rules into your automation platform, which does the ongoing scoring at scale.

How often should a lead scoring model be rebuilt?

Review it every two to three quarters, or after anything that changes your funnel: new pricing, a new segment, a new product. Because Joy rebuilds from current CRM data on demand, a refresh is one ask instead of a multi-week project.

Related Recipes

Marketing

Budget Allocator

Recommend budget splits based on historical cost-per-acquisition.
Marketing

Email Campaign Copy

Draft email sequences for campaigns and nurtures.
Marketing

Marketing Dashboard

Visualize key marketing metrics in one view, ask and see.

Ready to score leads your sales team trusts?

Join the waitlist and be first to try this workflow when JoySuite launches.

Join the Waitlist Browse All Recipes