RevOps
Combine fit and behavior into a score and a routing decision.
- Category
- RevOps
- Signup
- None
- Your data
- Stays in your browser
- Also in
- Claude, via MCP
How do you build a lead scoring model?
Score two things separately: fit (industry, company size, role) and behaviour (pricing page visits, engagement, high-intent actions such as a demo request). Weight each criterion, add the points, and route only leads that clear both an overall threshold and a minimum fit score. This tool uses 40 fit and 60 behaviour points, routing at 65+ with fit of 40+.
Promising but not sales-ready, keep it in nurture until a high-intent action fires. Don’t hand reps a lead that’s all fit and no signal.
Good scoring needs both axes: fit without behavior is a guess; behavior without fit is a tyre-kicker. Threshold on the combination.
About this calculator
A lead scoring model that only measures fit produces confident-looking scores on people who'll never buy, and a model that only measures behavior sends reps chasing tyre-kickers with no budget or authority. This calculator builds a score the way real routing rules should, blending three fit signals and three behavioral signals into one weighted 0-100 score and an explicit routing recommendation.
How to use it
- Rate the three fit criteria, industry/ICP match, company size fit, and job title/seniority, as Strong, Moderate, Weak, or None/unknown.
- Rate the three behavior criteria, pricing/demo page visits, content/email engagement, and high-intent actions like a trial or contact request, the same way.
- Read the overall lead score out of 100, the separate fit and behavior sub-scores, and the recommended routing action.
Methodology
Each rating maps to a point value, Strong = 1, Moderate = 0.66, Weak = 0.33, None = 0. Every criterion also carries a fixed weight: fit criteria total 40 points (industry 15, size 15, title 10), behavior criteria total 60 points (pricing 20, content 15, high-intent action 25). The overall score sums weight × point value across all six criteria.
Fit sub-score and behavior sub-score are each normalized to their own 0-100 scale, weighted points earned within that group divided by that group's total possible weight, so you can see fit and intent separately even though they combine into one overall number.
The routing recommendation isn't a single threshold, it requires both conditions: overall score of 65+ and a fit sub-score of at least 40 routes to sales. A score of 40-64 (or a high score riding entirely on behavior with weak fit) goes to nurture. Below 40 is disqualify or low priority.
That dual condition is deliberate: a lead with strong behavioral signals but poor fit can't clear the sales threshold on behavior alone, since a high-intent action from someone outside your ICP isn't a real sales-ready lead, it's a data point that needs fit qualification first.
The weighted scoring model in this tool
Points per criterion = Weight × Rating value (Strong 1, Moderate 0.66, Weak 0.33, None 0) Lead score = sum of points across all six criteria (maximum 100) Fit sub-score = Fit points ÷ 40 × 100 Behaviour sub-score = Behaviour points ÷ 60 × 100 Route to sales if Lead score ≥ 65 and Fit sub-score ≥ 40; Nurture if Lead score ≥ 40; otherwise Disqualify / low priority
- Fit weights (40)
- Industry / ICP match 15, company size fit 15, job title / seniority 10.
- Behaviour weights (60)
- Visited pricing / demo page 20, content / email engagement 15, high-intent action such as a trial or contact request 25.
- Rating value
- How strongly the lead meets each criterion. Use None / unknown when you have no data.
- Fit sub-score
- Fit points as a share of the 40 available, on a 0 to 100 scale.
Weights are fixed in the tool, so use it to test routing logic, then set your own weights in the CRM from closed-won data. Both major platforms separate the two axes. HubSpot's lead scoring tool builds fit, engagement and combined scores, labels combined scores A1 to C3 (letter for fit, number for engagement) and can decay engagement points over 1, 3, 6 or 12 months. Salesforce Account Engagement scores behaviour with points and grades fit from F to A+, with every prospect starting at D (Trailhead). This tool has no decay, so rate behaviour on recent activity only.
Worked example: three leads through the same rules
A B2B SaaS company selling to mid-size logistics firms tests its routing rules on three leads. Lead A is a strong-fit operations head who visited pricing but has not asked for anything. Lead B is the same person a week later after requesting a trial. Lead C is a student downloading every guide and requesting a trial.
- Lead A fit: industry Strong 15 + size Moderate 15 × 0.66 = 9.9 + title Strong 10 = 34.9 points; fit sub-score = 34.9 ÷ 40 = 87
- Lead A behaviour: pricing Moderate 20 × 0.66 = 13.2 + content Weak 15 × 0.33 = 4.95 + intent None 0 = 18.15; behaviour sub-score = 30
- Lead A score = 34.9 + 18.15 = 53, so Nurture
- Lead B: same fit, intent becomes Strong (+25); score = 53.05 + 25 = 78, fit 87, so Route to sales now
- Lead C fit: industry Weak 4.95 + size None 0 + title Weak 3.3 = 8.25; fit sub-score = 21
- Lead C behaviour: all three Strong = 60; score = 68, which clears 65 but fit is below 40, so Nurture
The rules do what they should: Lead B is sent to a rep the moment intent appears, while Lead C, with a higher score than Lead A, never reaches sales because fit fails. The test of the model is downstream: compare the MQL-to-SQL rate of leads routed by these rules with your current rate. First Page Sage reports 13% for B2B SaaS in its client data; if sales accepts far fewer of your routed leads, raise the fit minimum or the threshold. The MQL calculator shows what a stricter rule does to volume and cost per MQL.
MQL to SQL conversion rate by industry (for checking a scoring threshold)
| Industry | MQL to SQL rate |
|---|---|
| B2B SaaS | 13% |
| Software development | 14% |
| Cybersecurity | 15% |
| IT & managed services | 13% |
| Financial services | 13% |
| Fintech | 11% |
| Manufacturing | 16% |
| Transportation & logistics | 19% |
| Staffing & recruiting | 12% |
What makes a good lead scoring model?
A good model is judged by what happens after routing, not by how sophisticated it looks:
- It separates fit from intent. Both HubSpot and Salesforce report the two separately for this reason; a high total from behaviour alone should not reach sales.
- It predicts conversion. Leads above the threshold should convert to SQL and to customer at clearly higher rates than leads below it. Check this each quarter against closed-won data.
- It is agreed with sales. The threshold and the follow-up time belong in a written sales and marketing SLA.
- It decays. Old engagement should not keep a lead hot. HubSpot's decay setting handles this; in other systems, use date-limited criteria.
A decision rule: if routed leads convert to SQL at a rate close to unrouted leads, the weights are not predicting anything; rebuild them from closed-won patterns, as described in the lead scoring model guide. For fit alone, use the lead grading calculator.
Lead scoring in India vs the US
Indian B2B enquiries often arrive by WhatsApp, phone or marketplaces such as IndiaMART, where there is no page-view history to score. Rate behaviour from what the conversation shows (asked for pricing, asked for a demo) and log it in the CRM, or those leads will always score as low intent. Company size data for Indian SMBs is patchy in enrichment tools, so None / unknown is common for fit; a lower fit minimum for verified enquiries may be justified, but test it. Zoho CRM consulting in India covers how to capture these signals from marketplace and WhatsApp enquiries.
US B2B teams usually have richer data: enrichment tools fill company size and role, and marketing automation tracks page views and email engagement. The risk shifts to over-scoring engagement such as email opens, which privacy features in mail clients can inflate. Weight high-intent actions (demo request, pricing page, trial) far above passive engagement, as this tool does with 25 and 20 points against 15. For larger teams, the HubSpot lead scoring consultant guide covers how to configure fit and engagement groups.
Common lead scoring mistakes
- Adding fit and behaviour into one number with no minimum on fit, so heavy content readers outside the ICP reach sales.
- Guessing Moderate when data is missing instead of marking None / unknown, which inflates scores.
- Never decaying engagement, so a lead who was active a year ago still looks sales-ready.
- Setting weights by opinion and never comparing routed leads' conversion with closed-won data.
- Changing the threshold without telling sales, so MQL counts jump and trust in the model falls.
- Scoring email opens heavily even though automated privacy features can register opens nobody made.
Sources
- Overview of the lead scoring tool, HubSpot Knowledge Base. Accessed 2026-10-18.
- Get Started with Lead Grading, Salesforce Trailhead. Accessed 2026-10-18.
- MQL to SQL Conversion Rate By Industry: 2026 Report, First Page Sage. Accessed 2026-10-18.
Go deeper
FAQ
Because a high-intent action, requesting a trial, contacting sales directly, from someone with a poor ICP match is usually a sign of the wrong prospect showing real interest, not a qualified buyer. The model requires fit above 40 specifically to prevent behavior alone from triggering a sales handoff.
Grading (see the Lead Grading Calculator) measures fit alone, an A-F letter grade purely on firmographics like industry, size, budget, authority and geography. Scoring, this tool, blends that fit dimension with behavioral and intent signals into one number and a routing decision, the two are meant to work together in a grade × score qualification matrix.
Rate what you can verify honestly and mark the rest as None/unknown, since guessing Moderate on an unverified signal will inflate the score past what your actual data supports. It's better to route conservatively on solid data than confidently on assumptions.
There is no standard number; it depends on your weights. Set the threshold so that sales can follow up every routed lead within your agreed response time, then check that routed leads convert to SQL at a clearly higher rate than the rest. This tool uses 65 out of 100 with a fit minimum of 40.
Explicit scoring uses information the lead gives or that you can verify about them, such as role, company size and industry; it measures fit. Implicit scoring uses behaviour, such as page visits and content downloads; it measures interest. Salesforce Account Engagement calls the first a grade and the second a score.
Yes, for signals that clearly reduce the chance of buying, such as competitors, students, job seekers or unsubscribes. This tool does not subtract points; marking a criterion as None gives zero. In a CRM, add negative criteria or a disqualification rule so these leads never reach sales.
Review it each quarter against conversion data and whenever your ICP, pricing or main lead sources change. Look at the share of routed leads sales accepts and the win rate of routed versus non-routed leads. Change weights in small steps and record each change so results stay comparable.
Predictive scoring finds patterns in historical data automatically, but it needs enough closed-won and closed-lost records to learn from and can be hard to explain to sales. Rule-based scoring, like this tool, is transparent and works with little data. Many teams start with rules and move to predictive once data volume allows.
Numbers looking off?
If the model says one thing and the pipeline says another, that gap is usually where the revenue is. A 30-minute call finds it.
It starts with a 30-minute call. Pick a time below.