Lead scoring is one of the most frequently set up and least frequently used systems in B2B marketing. Most HubSpot accounts have a scoring property defined, but sales reps work from a shared spreadsheet or gut feel because they do not trust the score. The problem is rarely the technical setup. The problem is the input that went into the scoring model.

Lead scoring that sales actually uses requires three things to be true: the score predicts conversion to opportunity, the sales team agrees on what makes a lead qualified, and the workflow routes scored leads to the right rep at the right time. This guide covers how we build HubSpot lead scoring that earns sales team trust.

Start with the definition, not the model

Before opening HubSpot, run an alignment session with sales and marketing leadership. The questions to answer:

What does a marketing-qualified lead look like? Define on demographics (job title, company size, industry) and behaviour (pages visited, content downloaded, events attended). Be specific. “Director or above at a B2B SaaS company with 100 to 1000 employees who has visited the pricing page or downloaded a comparison guide” is a useful definition. “Engaged lead” is not.

What does a sales-qualified lead look like? This usually adds a budget and timeline signal beyond MQL criteria. A sales-qualified lead has explicitly indicated buying intent within a defined window. The signal might come from a demo request, ROI calculator interaction, pricing page session of 90 seconds plus, or a sales-initiated conversation.

What is the disqualification list? Free email domains (gmail.com, yahoo.com), specific industries you do not serve, geographies outside your serviceable market, competitors and student accounts. These should never score, regardless of behaviour.

Get sales leadership to sign off on these definitions in writing. The scoring model then enforces what was agreed, rather than imposing marketing’s view of qualification onto sales.

Demographic scoring

HubSpot’s lead scoring uses a single numeric property that combines demographic and behavioural points. Start with the demographic baseline:

Job title scoring. Map title patterns to score values. Senior decision makers (VP, Director, Head, C-level) get 25 to 35 points. Manager level gets 15 to 20 points. Individual contributors get 5 to 10 points. Use HubSpot’s “contains any of” logic on job title for matching, with separate rules per language if you sell internationally.

Company size scoring. Match against the company’s employee count enriched from Clearbit, ZoomInfo or HubSpot’s own enrichment. Ideal customer profile size bands get 20 points. Adjacent bands get 10 points. Outside bands get 0 points or negative if very far off.

Industry scoring. Map your top-converting industries to positive scores, never-converting industries to negative scores or disqualification. Use HubSpot’s Industry property (auto-enriched if you have HubSpot Sales Hub Enterprise).

Geography scoring. Add positive scores for your primary serviceable markets, negative scores for excluded regions. Be cautious about negative geo scores when running international campaigns because the data quality varies and false negatives can hurt.

Email domain scoring. Free email domains (gmail, yahoo, outlook) get a minus 20 to minus 50 score. Competitor email domains get minus 100, effectively disqualifying. Your own employee domains get minus 100.

Behavioural scoring

Behavioural scoring captures buying intent through actions on your site and in your funnel:

Page visits. High-intent pages (pricing, demo request, free trial signup, product comparisons) get 10 to 25 points per visit. Mid-intent pages (case studies, ROI calculators, integration pages) get 5 to 10 points. Low-intent pages (blog posts, careers) get 0 or 1 point.

Content downloads. Bottom-of-funnel content (vendor comparison guides, buyer’s guides, security whitepapers) gets 25 to 40 points. Middle-of-funnel content (industry reports, ROI calculators) gets 15 to 25 points. Top-of-funnel content (general thought leadership) gets 5 to 10 points.

Email engagement. Email opens give 1 to 2 points (acknowledging Apple’s Mail Privacy Protection makes this less reliable). Clicks give 5 to 10 points. Multiple clicks within a 7-day window give bonus points to capture engagement spikes.

Event attendance. Webinar registration gets 15 points. Webinar attendance gets 30 points. Webinar attendance plus follow-up question or chat interaction gets 50 points.

Demo requests, contact form submissions, calendar bookings. These should typically trigger SQL automatically through a separate workflow, but also add 100 plus points to the lead score to ensure the contact stays scored as hot.

Score decay

Without decay, a lead who downloaded a piece of content 2 years ago looks identical to a lead who downloaded it yesterday. Set up score decay through HubSpot workflows:

Behavioural scores should decay over time. We typically reduce behavioural scoring weights by 50 percent after 90 days and remove them after 180 days. Demographic scoring stays constant (a Director is still a Director).

The decay logic runs through a recurring workflow that re-evaluates contacts every 30 days, recalculating their score based on activity within the last 90 days only. This keeps scores aligned with current intent, not historical.

Thresholds and routing

Score thresholds turn into lifecycle stages and routing rules:

0 to 49 points: Subscriber. Lead is in the database, gets marketing nurture content, not actively in pipeline.

50 to 99 points: Lead. Lead is engaging meaningfully, gets nurture sequence appropriate to interest area, no sales touch yet.

100 to 199 points: Marketing Qualified Lead (MQL). Lead matches ICP and shows meaningful intent. Automatic email to BDR with lead context. BDR has 24-hour SLA to make first outreach.

200 plus points: Sales Qualified Lead (SQL). Lead is ready for direct sales conversation. Route to account executive based on territory or named account assignment.

Demo Request, Contact Form, Calendar Booking: Override scoring, route directly to sales as SQL regardless of score.

HubSpot workflows handle the routing. Use round-robin assignment for inbound territories, named-account routing for ABM programmes, and automatic Slack notifications when a high-score lead is created so reps see the alert in real time.

Feedback loop from sales

The most important part of lead scoring, and the part most teams skip. Set up a structured feedback mechanism:

Lost Reason on every MQL that does not convert to opportunity within 30 days. HubSpot has a Disqualified lifecycle stage. Use it. Capture lost reason as a dropdown: Not ICP, Wrong Timing, No Budget, Competitor, No Decision Maker, Other.

Won Reason on every opportunity that closes. Capture which marketing touches contributed (HubSpot’s attribution reporting), what content the lead engaged with, and which scoring signals were most predictive.

Monthly review with sales leadership. Walk through the disqualified MQLs and ask: should this have scored as MQL? What signal misled us? What do we add or remove from the model?

Update the scoring model quarterly based on this feedback. We typically refine the model 3 to 5 times in the first year, then settle into stable yearly tuning after that.

Common scoring mistakes

Scoring all email engagement equally. An email open from a generic newsletter is not the same as an open of a personalised one-to-one email. Segment scoring by email type, with sales-triggered emails scoring 5x to 10x marketing nurture emails.

Scoring blog post visits at high values. Blog traffic includes journalists, students, competitors, and casual researchers. Cap blog post scoring at 1 to 2 points per visit, and disqualify if blog visits are the only signal after 90 days.

Not weighting recent behaviour higher. A lead who downloaded 4 assets in the last week is hotter than a lead who downloaded 4 assets over 18 months. Implement velocity scoring: bonus points for multiple high-value actions within a 14-day window.

Letting demographic scoring dominate. If 60 percent of the score comes from job title and company size, you end up with high-scoring leads who have shown zero interest. Cap demographic scoring at around 40 percent of typical MQL threshold (around 40 points if MQL is 100).

No disqualification logic. Without explicit disqualification, you waste BDR time on competitors and students. Set up clear minus 100 score rules for known disqualifying patterns.

Tooling beyond HubSpot

For larger B2B programmes, HubSpot’s native scoring has limitations. Workarounds and add-ons we use:

MadKudu for predictive scoring. Uses machine learning trained on your historical conversion data to weight signals automatically. Plug into HubSpot through the integration, override or supplement the native scoring property.

Clearbit Reveal for company identification. Captures anonymous traffic and matches to companies, adding visitor-firmographic signal to lead scoring even before the contact form submission.

6sense or Demandbase for intent data. Buyer intent signals from third-party research behaviour. Add these as scoring signals at the company level, increasing scores for accounts showing intent for your category.

For accounts under 10,000 dollars per month in HubSpot spend, the native scoring is usually enough. Above that, the third-party tools start to earn their cost.