Lead Scoring Criteria: Build a Model Sales Trusts

Lead scoring promises that sales will spend its time on the prospects most likely to buy. In practice, many scoring models are ignored within months. Salespeople notice that high-scoring leads include students downloading every guide and competitors browsing the pricing page, while a perfect-fit buyer who called once sits at a low score. The problem is rarely the scoring tool. It is the choice of lead scoring criteria: which attributes and behaviours earn points, how many, and what takes points away. This guide explains how to choose criteria from evidence, separate fit from interest, use negative scoring and decay, set thresholds with sales, and check that the model keeps working.

Fit and interest are different questions

Every useful scoring model answers two questions:

• Fit: does this lead look like the customers we win and keep? This comes from attributes such as industry, company size, role and location.
• Interest: is this lead engaging with us now? This comes from behaviours such as visiting key pages, replying to emails, attending a webinar or requesting a demo.

Combining them into one number hides important differences. A perfect-fit lead with no recent interest needs nurturing; a highly engaged lead with poor fit may need a polite, low-effort response. Many teams keep two scores, or a grid of fit and interest, and route leads based on both.

Choose lead scoring criteria from evidence

The best criteria come from your own history, not from a template. Start by pulling closed deals from the last year or two, both won and lost, and comparing:

• Which industries, company sizes and roles appear most among won deals?
• Which first actions did eventual customers take, such as a demo request, a pricing page visit or a referral?
• Which attributes or behaviours were common among leads that never progressed?

Then talk to the salespeople who handle leads every day. They know which signals are misleading, such as a free-email address that turns out to be a small but valuable business, or a job title that sounds senior but rarely buys.

If you have enough history and want to go further, our guide to AI lead scoring explains how to learn weights from CRM data rather than setting them by hand.

Weight criteria by strength of signal

Not all criteria are equal. A demo request says far more than an email open. Group criteria by strength:

• Strong signals: demo or quote requests, replies to sales emails, a referral from an existing customer, a target role at a target company size
• Moderate signals: pricing or case study page visits, webinar attendance, a relevant industry
• Weak signals: newsletter sign-ups, single blog visits, email opens

Give strong signals points large enough that one of them matters, and keep weak signals small so they cannot add up to a sales-ready score on their own. Email opens, in particular, are unreliable because some mail clients load images automatically.

Use negative scoring

Points should go down as well as up. Common negative criteria:

• Competitor or own-company email domains
• Roles that never buy, such as students or job seekers
• Locations you do not serve
• Unsubscribes or bounced emails
• Visits to the careers page only

Negative scoring stops obviously unsuitable leads from reaching sales, however many guides they download.

Let interest decay

A pricing page visit last week is a signal; one from last year is not. Interest points should fade over time, either by expiring after a set period or by reducing gradually. Fit points usually do not decay, because a company's industry rarely changes, but they should update when the underlying data does, such as a change of role.

Worked example: a starting set of lead scoring criteria

The table below shows a hypothetical starting model for a B2B software company selling to mid-sized firms. Points and thresholds are illustrative and should be set from your own deal history and sales feedback.

CriterionTypePointsNotesCompany size in target rangeFit+20From enrichment or form fieldTarget industryFit+15Industries common among won dealsDecision-making roleFit+15Based on title mapping agreed with salesLocation not servedFit-40Effectively excludesDemo or quote requestInterest+40Strongest behaviour signalPricing page visitInterest+10Expires after 30 daysCase study downloadInterest+5Expires after 30 daysWebinar attendedInterest+10Expires after 60 daysCompetitor email domainFit-50Effectively excludesUnsubscribedInterest-20Stops nurture scoring

In this example, a lead is passed to sales when fit reaches a set level and interest reaches another, or immediately on a demo request from a lead with acceptable fit.

Set thresholds and routing with sales

Thresholds decide what happens at each score, so agree them with the sales team, not just marketing. Define what a sales-ready lead means, how quickly sales will follow up, and what happens to leads that sales rejects. Feed rejections back with a reason, such as wrong role or no budget, so criteria can be adjusted. Write the agreed definitions down in one short document that both teams can find, and revisit it whenever products, target markets or sales capacity change. Our guide to CRM lead routing covers assigning qualified leads to the right person quickly once they reach the threshold.

Check that the model works

Review the model every quarter:

• Do higher-scoring leads convert to opportunities and customers at higher rates than lower-scoring ones?
• What share of sales-ready leads does sales accept?
• Are there won deals that never reached the threshold? What did they have in common?
• Are some criteria awarded to almost everyone, making them meaningless?

Adjust one or two criteria at a time and watch the effect, rather than rebuilding the model each quarter.

Common mistakes with lead scoring criteria

• Copying a template instead of using your own won and lost deals
• One combined score that hides the difference between fit and interest
• Weak signals, such as email opens, adding up to sales-ready scores
• No negative scoring or decay
• Thresholds set by marketing without sales agreement
• Never checking whether scores predict conversions

Frequently Asked Questions

What are lead scoring criteria?

They are the attributes and behaviours that add or remove points in a lead scoring model, such as company size, role, demo requests, page visits and unsuitable email domains.

How do you choose lead scoring criteria?

Compare won and lost deals to find attributes and behaviours common among customers, then refine with feedback from the salespeople who handle leads.

What is negative lead scoring?

It removes points for signals that a lead is unlikely to buy, such as competitor domains, locations you do not serve or unsubscribes.

How often should a scoring model be reviewed?

Quarterly is a sensible rhythm: check whether higher scores convert better, how many sales-ready leads sales accepts and whether won deals were missed.

Build a scoring model your sales team trusts

We design lead scoring criteria from your deal history and set them up in your CRM with routing and review. See our AI automation solutions, or talk to us about lead scoring.

About the Author

Manav Sharma is the founder of Insyrge and a CRM solutions architect. He works across Zoho, Salesforce, HubSpot, Odoo and more, designing the CRM setups, integrations and automations that keep sales and operations teams working from the same data.