Lead scoring implementation

Turn an agreed definition of a useful lead into rules that can be checked on individual records. Decide which records to score, which attributes and actions contribute, how older activity is treated, and what decision a score supports.

A sales hand-off also needs eligibility and ownership rules; the score alone does not supply them.

Define the decision

Name the decision first, such as which enquiries need sales review. Agree who owns the model and what evidence a reviewer needs alongside the score.

Keep fit and engagement distinguishable: a suitable prospect may show no recent interest, while an active contact may fall outside the target market.

Weights and thresholds are choices to assess against outcomes, not proof that the model predicts conversion.

Data used to build the score falls into three groups. Demographic data covers the lead's own attributes, such as name, job title, company size, industry and location. Firmographic data describes the company: size, revenue, industry, location and tech stack. Behavioural data records engagement, including website visits, whitepaper downloads and webinar attendance.

Building an ideal customer profile and incorporating company-based information can help prioritise leads from high-value companies.

Examples of score criteria

HubSpot examples include a company fit score using Region, Company size and Industry; a Northeast company with 50-200 employees in Hospitality scores highly. A contact engagement score can score highly after more than five campaign interactions.

A deal score can mark deals likely to close when the company is in the correct industry and contacts are decision makers who have recently attended sales meetings.

Scale to teams and regions

You can create multiple scores per object to meet the goals of different teams or regions. A support team, for example, may need a different score to track the engagement of existing customers.

If your business operates in multiple regions, the qualities of a good-fit customer may differ across each location, requiring different fit scores. Set clear labels for each score property to keep teams organised.

Configure the approved rules

Check that the platform can express each rule without changing its meaning. HubSpot offers contact fit, engagement and combined scores with Marketing Hub Professional or Enterprise; contact scores require Marketing Hub.

Company fit, engagement and combined scores are available with Marketing Hub or Sales Hub, and deal scores are available with Sales Hub only. These products are available with Professional or Enterprise subscriptions, subject to the stated object-specific requirements.

Customer Insights – Journeys scores leads using demographic attributes and interactions. Check the actual subscription, permissions, feature settings and event sources before using either product example.

In HubSpot, creating a score requires Edit permissions for Lead Scoring; View permissions allow access to the tool and record score values. Navigate to Marketing > Lead Scoring and click Create score. Choose the object: contacts (Marketing Hub only), companies, or deals (Sales Hub only).

Select the score type next. Deal scores are combined scores by default; contact and company scores can be fit, engagement, or combined.

Name the score, set criteria and points, configure inclusion lists, set associated properties and thresholds, then review and turn on the score.

In Customer Insights - Journeys, go to Lead Management > Scoring Models and select New. Provide a name and specify conditions to target a specific audience and avoid generating extra scores.

Conditions are based on attributes of the audience entity, in this case lead attributes; for example, you can target leads from a specific country or region. The builder provides Attributes (demographic information) and Interactions as data sources.

Lead Scoring Models: HubSpot vs Microsoft Customer Insights – Journeys

Implementing a Lead Scoring Model: Key Steps

Score groups and inclusion lists

HubSpot calculates scores from criteria set for event and property rules in score groups. Each score must have at least one score group, but you can include multiple groups that add up to the total score.

Inclusion lists determine which records will be scored, so you can exclude records that should not receive a score.

Permissions and data privacy

In Customer Insights - Journeys, an administrator must enable the Lead Management feature switch under Settings > Overview > Feature switches. Fields with field-level security (FLS) enabled cannot be used in lead scoring model conditions.

Microsoft notes that automatic scoring models can introduce data privacy problems; organisations are responsible for ensuring they operate in full compliance with relevant data privacy regulations.

Control the hand-off

Decide which repeat actions add new evidence and limit weak repeated signals where suitable controls exist. Specify how old engagement loses influence without assuming that fit points or an earlier sales status will disappear. Missing or conflicting data needs a review outcome.

Set the score threshold alongside any further eligibility checks. The resulting action might create a review task; the lead-routing process should determine the destination and confirm ownership.

Record the score and model version used for the decision.

Qualify leads with the score

In Customer Insights – Journeys, qualification criteria are based on scoring models, and only live models can be used. You can add multiple models — for example, an engagement model and a demographic model — and qualify leads that meet both models or at least one.

Actions when the criteria are met update the lead's Status Reason to Marketing Qualified or Sales-Ready to Yes. Select Publish to make the qualification live.

Review the result

Before activation, compare expected scores for contrasting fit and engagement, repeated and old events, missing values and changed properties with the configured calculation. After activation, inspect selected records, their score histories and the resulting review decisions. Recheck existing records and connected workflows after a model change.

Lead Scoring Implementation Metrics (Australia)