What is account scoring?

Account scoring evaluates and prioritizes companies using explicit criteria such as customer fit, recent engagement, and relevant buying signals. The score belongs to the company record. Evidence can come from that company's attributes and from activity across associated contacts.

The output is a priority signal, not automatic proof that a company will buy. A rules-based score of 70 does not mean a 70% purchase probability unless a separately validated model gives it that interpretation.

Why account scoring matters

B2B buying activity is often distributed across several people. One person evaluates technical requirements, another checks commercial terms, and another approves spending. Looking at each contact in isolation can hide the account's overall context.

Lead scoring is the broader prioritization discipline and can include accounts. Account scoring focuses on the company-level details: which people belong together, how their evidence is combined, and which action the company deserves.

That distinction supports account-based marketing. The target is an organization with a buying situation, not simply the individual who clicked most recently.

How account scoring operates

Start by defining the scored entity. Decide whether the account represents a parent company, subsidiary, business unit, or another commercial boundary. Match contacts to that entity before aggregating their behavior.

Then separate fit from timing. Fit asks whether the company resembles the ICP: relevant industry, size, region, use case, or technical environment. Engagement and timing ask whether recent activity deserves attention now.

Set rules for associated-contact evidence. Repeated visits from one person should not necessarily outweigh activity from several relevant members of a buying committee. Cap repetitive contributions, remove duplicates, and preserve who generated each signal.

Apply recency deliberately. An old evaluation should not continue raising current priority indefinitely. Use a defined observation period or reduce the contribution of aging events.

Finally, connect score ranges to actions. A high-fit, quiet account may need research. A high-fit account with recent relevant activity may warrant outreach. A low-fit company should not become sales-ready merely by accumulating clicks.

A SaaS example

Consider an illustrative security software company. Its scoring rules award an account 40 fit points because its size, industry, and technical environment meet the team's criteria.

Recent evaluation activity from relevant contacts contributes another 30 points after the model applies its event caps. The company receives a total score of 70.

The account owner can inspect both components and the supporting evidence. If the engagement evidence becomes stale, the timing contribution falls while the fit component remains.

A second company might also total 70 through a different combination. The two accounts should not automatically receive the same action. Component visibility lets the owner distinguish strong fit with limited timing from weaker fit with heavy activity.

These points illustrate a local rule set. They are not recommended weights or a conversion benchmark.

Common mistakes

Summing every contact score rewards large databases and duplicate records. Aggregate the evidence that matters, not every available number.

Treating intent data as confirmed purchase intent creates another error. Research activity can indicate interest without establishing budget, authority, or a current project.

Combining prospects and existing customers under one unexplained score also causes confusion. Acquisition priority and customer health answer different questions and need different success criteria.

Keep fit and timing visible beneath the total. A useful account score should let an owner explain which company deserves attention, what changed, and what evidence would alter the next action.

For more practical account prioritization guidance, join the newsletter below.