HubSpot Lead Scoring 101: A Guide for Beginners

HubSpot lead scoring assigns a numeric score to a record based on its properties and behaviour, so your sales and marketing teams can prioritise outreach around who is actually ready to buy rather than who simply exists in the database. Generating leads is not enough on its own; knowing which ones are worth a call today is what the score is for.

Which object you can score, and how to make a score reflect an entire buying group rather than one person, are the two questions this guide answers.

What Is HubSpot Lead Scoring?

Lead scoring works by evaluating a record's demographic properties and behavioural signals, such as job title, company size, page visits and email engagement, and converting them into a single number that reflects readiness to convert. HubSpot offers two approaches: manual scoring, where you set the point values yourself, and predictive scoring, which uses machine learning to weight the same kind of signals automatically. A well-configured score exists so sales spends time on the records most likely to close, not on every record that happens to be new.

Which Object You Can Score Depends on Which Hub You Hold

Lead scoring is not one universal setting; it is a tool gated by subscription and further split by object. Score properties need Marketing Hub or Sales Hub Professional or Enterprise, and within that, contact scores are Marketing Hub only, company scores are available on either Marketing Hub or Sales Hub, and deal scores are Sales Hub only, with AI-built contact scores reserved for Marketing Hub Enterprise. A Sales-Hub-only portal cannot score a contact at all, and a Marketing-Hub-only portal cannot score a deal. Check which Hub the score you want actually needs before you plan a rollout around it, because the object you assumed you could score may not be the one your subscription unlocks.

One detail works in your favour and is worth knowing rather than assuming away: scoring is retroactive at creation. When a score is first turned on, HubSpot evaluates records based on their current and historical property values and actions, not only from that point forward, so a portal that just inherited a contact list from a migration does not start every record at zero.

How to Score the Buying Group, Not Just the Person Who Filled the Form

The contact who submits a form is rarely the person who approves the spend. Gartner's 2025 B2B buyer survey found that technology buying groups now range from 5 to 16 people across as many as four functions, and a score built only on the one contact who converted answers a narrower question than how engaged the whole group actually is.

HubSpot has a native answer to this, and it is built from two pieces working together. For company and deal scores, event criteria are based on the associated contacts' own activity, and you can customise exactly which associated contacts count toward it. Separately, a rollup property can sum a number property, such as each contact's individual score, across every contact associated with a company, giving you a company-level total built from the group rather than from one record. Put those two together and a compliance or procurement stage that only ever touched three of your six committee roles shows up in the company score as exactly that: partial engagement, not full readiness. Rollup and calculation properties carry their own subscription tier and a limit on how many you can create, so confirm both before building the property, but the mechanism itself is native and does not need a workaround workflow.

Manual vs Predictive Scoring

Manual scoring means you or your team decide which actions and attributes add or subtract points: visiting a pricing page might add ten, downloading a whitepaper might add five, unsubscribing might subtract points entirely. It takes more upkeep but gives you full control over what a score actually means for your business, which matters more the more specific your buying process is.

Predictive scoring uses HubSpot's machine learning to identify patterns across your historical data automatically, requiring less manual maintenance and adapting as behaviour changes over time. Many teams end up running both: predictive scoring for contact-level volume, manual rules layered on top for the specific signals, like a compliance-stage property, that only your business knows matter.

Keeping a Score Workflow From Going Quiet

If your scoring logic runs through a workflow rather than a native score property, the same enrollment rules that govern every HubSpot workflow apply to it. Re-enrollment is off by default, so a record that already passed through the scoring workflow once will not be re-evaluated on a later qualifying event unless re-enrollment is switched on for a trigger that is itself eligible; activity-based triggers can never be made eligible at all. Turning re-enrollment on afterward is forward-only, so records missed while it was off stay missed until someone enrols them manually. A score that quietly stopped updating three months ago looks identical to a score that is current, which is exactly what makes it dangerous. Our guide to HubSpot workflows covers the enrollment rules in full.

How to Set Up Lead Scoring in HubSpot

Predictive scoring, available on Marketing Hub Enterprise:

1. Go to Settings, then Properties, and find the Predictive Lead Scoring section.
2. Enable it and let HubSpot analyse historical data to identify converting patterns.
3. Review the predictions periodically and adjust your process around what the model surfaces.

Manual scoring, available wherever the score properties tool is unlocked:

1. Decide which behaviours and attributes matter, and assign point values to each.
2. Build the rules in the Lead Scoring section of Properties, mixing positive and negative scores.
3. Review the rules regularly, especially after a change to your sales process or your buying-committee model.

Where lead status and lifecycle stage fit into the score is its own question, and it is worth reading on its own: our guide to lifecycle stage versus lead status covers where a compliance review actually belongs in the two properties, rather than repeating that here.

Scoring a committee accurately, and reporting it back to leadership as one pipeline instead of a set of disconnected contact scores, is what our RevOps work is built to set up. If your portal can score a contact but cannot yet answer how engaged the whole account is, talk to us about the rollup and the properties behind it.