Lead scoring

Lead scoring is a method that gives every prospect a score based on who they are and what they do, so you can spot the ones most likely to become customers. Sales teams work the highest scores first instead of calling everyone in order of arrival.

Leads, MQLs, SQLs and the rest of the vocabulary

These terms come up in every scoring discussion, and teams define them differently. The most common meanings:

• A lead is a contact who has shown interest: filled in a form, downloaded content or contacted the company.
• A marketing qualified lead (MQL) is a lead that marketing considers ready to hand over to sales.
• A sales qualified lead (SQL) is a lead that sales has accepted to work on.
• A product qualified lead (PQL) is a user of a free plan or trial who has behaved in a way that suggests they would buy. It is common in software companies.

Leads are also often sorted informally as cold, warm and hot. A score turns that gut feeling into a number, and the threshold decides when a lead moves from one stage to the next.

Fit and intent: two scores, not one

A solid model measures two different things and does not add them up blindly:

• Fit: how closely the contact matches your target. Job title, industry, company size, country, tools in use. In B2B these are called firmographic criteria, and they rarely change.
• Intent: how interested the contact is right now. Pages viewed, forms completed, content downloaded, a demo request. These change every week.

A marketing director who matches your ideal profile perfectly but has never visited your site is not a hot lead yet. A student who downloads everything is not a customer. Keeping the two scores apart, or at least reading them together, avoids both kinds of false positive.

An example scoring grid

This grid is an illustration, not a template to copy. Points should be calibrated on your own sales data.

• Decision-maker title at a company of the right size: +20.
• Visit to the contact page or the offer overview page: +15.
• White paper download: +10.
• Demo request: +40.
• Email opened: +2.
• Personal email address, or a role outside your target: minus 10.
• Unsubscribe: minus 20.
• No activity for two months: minus 10.

With a threshold of 50 points, a lead who combines a good profile, a visit to the contact page and a downloaded white paper goes to sales. Negative points matter as much as positive ones: they take out of circulation the contacts that clutter follow-up lists.

Where the signals come from

Scoring is only as good as the tracking that feeds it. Forms fill in the contact’s profile, page and click tracking gives their behavior, and the source of the visit adds another clue: a lead who arrives from a very specific search is not in the same frame of mind as a visitor who came through a social post. This is where UTM parameters earn their keep: they carry the originating campaign all the way into the CRM. Entry pages such as a well-tracked landing page are often the first place this tracking is set up.

Decay and negative scoring

A contact who was very active six months ago and silent since should not stay at the top of the list. Good models make points expire: a visit counts for a few weeks, then fades. Negative scoring does the opposite job, pushing down signals that suggest a poor fit, such as a competitor’s domain, a student address or a clear unsubscribe. Without these two mechanisms, scores only ever go up and the threshold loses its meaning.

From lead to account: B2B scoring

In B2B sales, the buying decision is rarely made by one person. Account scoring groups signals from several contacts at the same company: the marketing director reading the site, the product manager downloading content, a third person requesting a demo. Taken alone, none of them reaches the threshold. Together they show real interest from the company. This approach mostly serves teams that target a short list of specific accounts.

Checking that the model works

A scoring model is judged by its results. Group your leads by score band (low, medium, high) and compare, for each band, the share that became customers. If high scores convert no better than low ones, the model is not measuring the right thing. Ask the sales team too: a lead rated excellent by the score and written off by sales points to a criterion that needs revisiting. Plan a review every quarter or so, since products and audiences shift.

Common pitfalls

• A score sales does not use. It is worth nothing, however clever the math.
• A frozen model. Products, targets and channels change, so criteria must too.
• Too much weight on email opens. It is a fragile signal, since some mail clients trigger opens without the person reading the message.
• A threshold set too low. Sales drown in lukewarm leads and end up ignoring the alert.
• Too many criteria. Eight to ten well-chosen signals beat thirty.

Which tools to use

Most CRMs and marketing automation platforms, including HubSpot, Salesforce and Pipedrive, include a scoring feature, and each documents how to set it up in its own interface. Without a dedicated tool, a spreadsheet or an automation built with a tool like Make or Zapier can get you started. Some CRMs also offer predictive scoring, calculated automatically from past deals, which needs a lot of data before it becomes reliable. At Mazette.co, the Performance marketing expertise covers these automation chains, where leads are connected, scored and stored in the CRM.