---
title: "Lead Scoring: How to Build a B2B Lead Scoring Model"
description: "How to build a B2B lead scoring model from closed-won data: fit vs engagement scores, a sample scorecard, thresholds, decay, HubSpot setup, and AI fit scoring."
url: https://getaob.com/lead-scoring
---

# Lead scoring: how to build a B2B lead scoring model that works

**Lead scoring** ranks leads by how likely they are to buy, so reps call the right people first. Most B2B scoring models fail quietly. Someone picks point values in a meeting, nobody checks them against real deals, and a year later reps ignore the score because everyone sits at 45.

This guide builds one from your own closed-won data, with a sample lead scorecard, thresholds, decay, the HubSpot setup, and where an AI prompt helps. Scoring is one step in a bigger process; for capture, routing and speed to lead, see [B2B lead management](https://getaob.com/b2b-lead-management).

By [Garrett Wolfe](https://garrettawolfe.com), founder of OneGTM, employee #9 at Unify, and author of the [State of GTME report](https://stateofgtme.com). Updated November 8, 2026.

Key takeaways

- Score fit (who they are) and engagement (what they did lately) as two separate numbers. Adding them together hides a bad-fit lead behind a burst of clicks.
- Build the scorecard from closed-won and closed-lost deals. Every attribute should earn its points by showing up more often in deals you won.
- Use negative points and hard disqualifiers. Students, competitors and companies too small to buy should never reach a rep.
- Let engagement decay. A pricing page visit from last spring is not a reason to call today.
- Check the model by score band. If the top band books no more meetings than the middle, the weights are wrong.

## What lead scoring is: fit score vs engagement score

A lead score answers two questions that move at different speeds. Is this the kind of company and person who buys from us? And are they showing interest right now? The first is a **fit score**. The second is an **engagement score**, and when the activity happens off your own site (a funding round, a new VP, a job post) it is usually called intent or a [buying signal](https://getaob.com/buying-signals).

|  | Fit score | Engagement or intent score |
| --- | --- | --- |
| Measures | Match with your ideal customer profile | Recent interest and timing |
| Inputs | Industry, employee count, revenue, region, tech stack, job title, seniority | Demo requests, pricing page visits, email clicks, webinar attendance, hiring or funding news |
| Changes | Over quarters, when the company or person changes | Over days and weeks, and should fade |
| Main job | Decide who is worth contacting at all | Decide who to contact first |

Keep them as two fields. A combined view is fine for reporting, and HubSpot shows one as a grid (more on that below), but the two numbers should never be summed into one. A student who downloads five ebooks would otherwise outrank a VP at a perfect-fit account who visited the pricing page once.

## How to build a B2B lead scoring model from closed-won data

1. **Pull the deals.** Export the last 12 to 24 months of closed-won and closed-lost opportunities, with the company and the main contact on each. Fifty won deals is a workable start. Fewer than 20 and you are mostly scoring opinion, so lean on fit attributes your team agrees on and revisit in a quarter.
2. **Fill in the fields.** Enrich every record with the attributes you plan to score: industry, headcount band, region, tech, title and seniority. A field that is blank on half your records cannot be scored. [HubSpot data enrichment](https://getaob.com/hubspot-data-enrichment) covers how to keep them filled.
3. **Compare won against lost and against the whole database.** For each value, work out the win rate, or simply how much more often it appears in won deals than in your total lead pool. A value that shows up in 40% of wins and 10% of leads earns real points. One that appears equally in both earns none, however much the team likes it.
4. **Turn the gaps into points.** Give the biggest gaps the most points, add negative points for values common in lost deals, and set hard disqualifiers that zero the score.
5. **Backtest before launch.** Score last year's leads with the new model and check that won deals land in the top band. If they don't, adjust before any rep sees it.

Want this built for your team?

Tell me what you're trying to fix. The call is with me, and I'm the one who'd build it.

Book a call with Garrett

## A sample lead scorecard

Here is what a fit scorecard looks like for a hypothetical B2B software company selling to finance teams at mid-sized companies. Your attributes and points will differ; the shape is what to copy. Each row has a reason, and the reason should come from your deal data.

| Attribute | Points | Why |
| --- | --- | --- |
| Industry is software, fintech or professional services | +20 | Most closed-won deals came from these three |
| 200 to 2,000 employees | +20 | Big enough to have the problem, small enough to buy without a procurement cycle |
| 50 to 199 employees | +8 | Wins here, but smaller deals |
| Uses NetSuite or Sage Intacct | +15 | The integration is the main reason buyers pick the product |
| Title is CFO, VP Finance or Controller | +20 | Signed or championed most won deals |
| Title is accountant or analyst | +5 | Useful users, rarely the buyer |
| US or Canada | +10 | Where the sales team and the integrations are |
| Recently raised a round | +10 | Finance teams rebuild their stack after funding |
| Personal email domain | −15 | Rarely converts, often a student or job seeker |
| Fewer than 20 employees | Disqualify | No deal under this size has closed |
| Competitor or agency domain | Disqualify | Research, not buying |

Cap the fit score at 100 and keep the engagement scorecard separate: a demo request might be worth 40, a pricing page visit 15, an email click 3. Low-value actions like email opens are best left out entirely, since privacy features and security scanners trigger opens nobody made.

## Thresholds, score decay, and the MQL handoff

A score is only useful once it triggers something. Set two or three bands and decide what each one does:

| Band | Rule | What happens |
| --- | --- | --- |
| Sales now | Fit 70+ and a high-intent action (demo request, pricing visit, reply) | Becomes an MQL, gets an owner and a call task within minutes |
| Outbound | Fit 70+ with no recent engagement | Goes into a targeted outbound sequence, timed to a signal where one exists |
| Nurture | Fit 40 to 69, or engagement without fit | Marketing email and retargeting, rescored as data changes |
| Ignore | Below 40 or disqualified | No sales time at all |

Pick the threshold from your backtest: the score above which leads converted to meetings at a rate worth a rep's time. Then agree it with sales in writing. An MQL definition marketing set alone is the fastest way to get scores ignored.

**Score decay** keeps engagement honest. Halve or remove engagement points after 30 to 90 days so that a lead who went quiet drops back to nurture on its own. Fit points should not decay, though they should be recalculated when someone changes job or a company crosses a size band.

The handoff itself is a routing problem: who owns the account, round robin or territory, and how fast the first touch goes out. That part lives in your CRM workflows and a router; see the comparison of [lead routing tools](https://getaob.com/lead-routing-tools) and how we wire it in [HubSpot automation](https://getaob.com/hubspot-automation).

## HubSpot lead scoring: how the tool works

HubSpot replaced its old score properties with a dedicated lead scoring tool. Per HubSpot's knowledge base, you could no longer create the old properties from May 1, 2025, and existing ones stopped updating on August 31, 2025. If a workflow or list still reads an old score, it is reading a frozen number. The details below are from [HubSpot's lead scoring documentation](https://knowledge.hubspot.com/scoring/understand-the-lead-scoring-tool?utm_source=onegtm&utm_medium=referral&utm_campaign=lead-scoring) (updated September 2, 2026) and [its note on the old score properties](https://knowledge.hubspot.com/crm-setup/set-up-score-properties-to-qualify-leads?utm_source=onegtm&utm_medium=referral&utm_campaign=lead-scoring), checked November 8, 2026.

| Question | Answer (checked November 8, 2026) |
| --- | --- |
| Which plans | Marketing Hub or Sales Hub, Professional or Enterprise |
| Score types | Fit and engagement for contacts and companies; combined for contacts, companies and deals |
| What each hub scores | Contacts need Marketing Hub, companies work with either hub, deals need Sales Hub and only take a combined score |
| Labels | Fit and engagement scores show High, Medium or Low. A combined score shows a grid from A1 to C3, where the letter is fit (A is high) and the number is engagement (1 is high) |
| Limits | You set an overall score limit and optional limits per criteria group, so one group cannot drown out the rest |
| Decay | For engagement and combined event groups only, every 1, 3, 6 or 12 months |
| AI scores | Marketing Hub Enterprise, with at least 25 converted and 25 non-converted contacts to learn from |

The A1 to C3 grid is the useful part, because it keeps fit and engagement apart while still giving reps one label to sort by. A practical HubSpot setup: build the fit score from enriched company and contact properties, build an engagement score with decay on, use the combined label in rep views, and trigger a workflow when a lead reaches A1 or A2. If your best fit signal comes from outside HubSpot (an AI classification, a product usage count, a funding date), write it to a custom property first and score that property.

## AI fit scoring with an LLM prompt

Rules break on messy data. Job titles are the obvious case: "Head of Money", "Finance & Ops Lead" and "Controller (Interim)" all need to land in the same bucket, and a list of keywords will miss half of them. Industry is the other. A company's LinkedIn industry often says "Software Development" when what matters to you is whether it sells to hospitals.

A short prompt handles both well. Give the model the title, or the company's website text, and ask it to pick one label from a fixed list, with a one-line reason. Then score the label in your scorecard like any other field. We built a title-fit AI prompt for Patlytics, and SuperDial's TAM was scored in Clay with AI industry classification, alongside HubSpot rep views. The [best LLM for GTM](https://getaob.com/best-llm-for-gtm) page covers which models fit this job and what they cost.

### Where AI scoring goes wrong

- **Free-form scores.** Asking a model to "score this lead from 1 to 100" gives numbers that drift between runs and cluster around 70. Ask for a label from a fixed list and let your scorecard assign the points.
- **No "unknown" option.** Without one, the model guesses. A thin website should come back as unknown and get no points, not a confident wrong industry.
- **Untested prompts.** Hand-label 50 to 100 records, run the prompt on them, and read every disagreement before running it on the full database.
- **Scoring what you could look up.** Headcount, location and funding come from data providers more cheaply and more reliably. Save the model for judgement calls.
- **Never re-running.** Companies pivot and people change jobs. Re-run classification when the inputs change, at least once a quarter.

## Predictive lead scoring

Predictive scoring trains a model on past conversions and lets it set the weights. It can find patterns a person would miss, and it saves the manual work of maintaining points. The catch is volume. A model needs a few hundred conversions to learn anything stable, and early-stage B2B teams with 30 deals a year don't have them. For most teams under that line, a hand-built scorecard from closed-won data, checked every quarter, will beat a predictive model and is easier to explain to sales.

## How to check that a lead scoring model works

The most common failure is that everyone scores the same. Half the database sits between 40 and 50, so the score cannot tell a rep who to call. Run two checks before launch and again every quarter.

**Distribution.** Plot how many leads land in each score band. You want a long middle and a small, clear top tier. If most records share a score, either the attributes don't vary across your data (everyone is in the US, everyone gets the same industry points) or blank fields are scoring zero across the board. For one vertical SaaS client with a TAM of about 45,000 companies, a rebuild moved the median score from 0.25 to 0.50 and left 2,001 accounts above 0.75, which gave reps a clear first list. The [GTM engineering guide](https://getaob.com/gtm-engineering) walks through that rebuild.

**Conversion by band.** For each band, compare the rate of meetings booked and of closed-won. The rates should climb with the score. If the top band converts no better than the middle, the weights are rewarding the wrong things, and the fix is usually to go back to the won and lost comparison with fresher deals.

Watch one more thing: whether reps use it. If they sort by something else, ask why. They are often right about an attribute the model is missing.

## How OneGTM builds lead scoring into the CRM

We start from the client's own closed-won and closed-lost deals, enrich the fields the scorecard needs, and write fit and signal scores to properties in HubSpot or Attio so routing, rep views and reporting can use them. Where a field needs judgement, like title fit or industry, we classify it with a tested prompt. Then we check the distribution and conversion by band before reps see it, and again after the first quarter. Scoring is usually one part of a wider build that includes [signals](https://getaob.com/buying-signals), outbound and routing; [the RevOps guide](https://getaob.com/revops) covers where it sits. Not sure where your gaps are? The free [GTM scorecard](https://getaob.com/gtm-scorecard) grades your whole setup in twelve questions.

## Lead scoring FAQ

### What is lead scoring?

Lead scoring is a way to rank leads by how likely they are to buy. Each lead gets points for who they are (fit: industry, company size, job title) and for what they do (engagement: demo requests, pricing page visits, email clicks), and the score decides who sales calls first and who stays in nurture.

### What is a lead scorecard?

A lead scorecard is the table behind a lead score: each attribute you score, the points it adds or subtracts, and the reason it is there. A good one fits on a page, is built from your closed-won and closed-lost deals, and includes negative points and disqualifiers as well as positive ones.

### How does HubSpot lead scoring work?

HubSpot's lead scoring tool, in Marketing Hub or Sales Hub Professional and Enterprise, lets you build fit, engagement and combined scores. You set criteria groups with points and a score limit, HubSpot labels each record High, Medium or Low, and a combined score shows fit and engagement as a grid from A1 to C3. Engagement points can decay every 1, 3, 6 or 12 months.

### What is the difference between a fit score and an engagement score?

A fit score measures whether the company and person match your ideal customer profile, and it changes slowly. An engagement score measures recent activity such as form fills, email clicks and page views, and it should fade over time. Keep them as separate fields so a burst of activity from a poor-fit lead never looks like a hot lead.

### What is predictive lead scoring?

Predictive lead scoring uses a model trained on your past conversions to weight attributes automatically, instead of a person assigning points by hand. It needs enough history to learn from. HubSpot's AI scores, for example, need Marketing Hub Enterprise and at least 25 converted and 25 non-converted contacts.

### How do you know if a lead scoring model works?

Look at two things. The score distribution should have a clear top tier, so if most leads land on the same score the model is not separating anything. Conversion to meeting and to closed-won should rise by score band, so a model where the top band converts no better than the middle needs new attributes or new weights.

Want a lead score your reps actually sort by?

## Related

- [B2B lead management](https://getaob.com/b2b-lead-management): The lead management process step by step: lifecycle stages, fit scoring vs signals, routing, speed to lead, and the stack.
- [Ideal customer profile (ICP) guide](https://getaob.com/ideal-customer-profile): What an ICP is in B2B, ICP vs buyer persona, how to build one from closed-won data, a fill-in template, and a worked example.
- [HubSpot automation and workflows](https://getaob.com/hubspot-automation): HubSpot workflows for lead routing, lifecycle stages, enrichment, and scoring, plus custom code where native tools stop.
- [HubSpot data enrichment](https://getaob.com/hubspot-data-enrichment): Native HubSpot enrichment vs a waterfall, which fields to refresh and when, and the rules that keep the CRM clean.
- [Best lead routing tools](https://getaob.com/lead-routing-tools): RevenueHero, Chili Piper, LeanData, Calendly and more for form to booked meeting, with pricing.
- [Buying signals guide](https://getaob.com/buying-signals): The 15 B2B buying signals worth tracking, how to detect each one, and how to turn a signal into outreach.

## Talk to OneGTM about lead scoring

Tell me what you're building. I reply the same day.

Book a call at [getaob.com/#contact](https://getaob.com/#contact) or email [team@getaob.com](mailto:team@getaob.com).