Marketing Qualified Lead: The Threshold Between Interest and Follow-Up

A marketing qualified lead is a contact that marketing considers ready for a salesperson to follow up. It is a judgement rather than a fact, normally expressed as a score crossing a threshold, and it combines two quite different kinds of evidence: whether this is the sort of organisation and person you sell to, and whether they have behaved like somebody with a live problem. The label exists to solve a coordination problem between two teams, and it works only to the extent that both teams have agreed what it means. Where they have not, it becomes a number marketing reports and sales ignores.

Two dimensions, not one

Fit covers attributes that do not change with attention: industry, size, location, role, business model. It answers whether this contact could ever be a good customer. Behaviour covers what they have done: pages viewed, materials requested, emails replied to, messages sent, meetings attended. It answers whether something is happening now. Collapsing the two into a single score is the most common design error, because it allows a very engaged bad fit to outrank a quiet good one.

Thresholds and decay

A threshold should be set on both dimensions rather than on a total, and the behavioural half should decay. Interest is evidence about a moment; without decay, a contact who read three pages in March sits permanently above the line while somebody active this week queues behind them.

A worked scoring example (illustrative)

Score fit out of 10: correct industry 4, employee band 3, decision-making role 3. Score behaviour out of 10: pricing page visited 4, guide downloaded 2, email replied to 3, three or more sessions 2, webinar attended 3, capped at 10. Require at least 6 on each. A contact in the right industry and size band, but in a junior role, scores 7 on fit; if they visited pricing and replied to an email they score 7 on behaviour, and they qualify. A contact who downloaded four guides, attended a webinar and scored 10 on behaviour, but works in a different industry entirely, scores 3 on fit and does not qualify. The second case is where most single-score models fail, and the two-dimension version handles it without any special rule.

It exists to allocate a scarce resource: salespeople's time. Most businesses generate more contacts than they can call, so somebody has to decide who gets a conversation this week. Doing that with an explicit, reviewable rule is better than doing it by whoever shouts loudest or replies fastest, and it makes the decision improvable, because a rule can be tested against outcomes.

It also creates a boundary that both teams can be measured against. Marketing is accountable for producing contacts that meet the standard; sales is accountable for working them promptly and giving feedback when they do not meet it. That accountability only exists if the definition is written down and the acceptance rate is measured, which is precisely the part that most organisations skip.

An undefined or unilateral definition

Where marketing sets the definition alone, it drifts towards what marketing can produce rather than what sales can use, and the acceptance rate falls quietly. Agreeing the fit criteria and the behavioural threshold jointly, and reviewing them against outcomes every quarter, is the whole mechanism. Without it the term is simply a synonym for a form fill.

Making the count a target

Any target expressed as a number of qualified leads can be met by lowering the threshold, and under pressure it will be. The result is a rising count, a falling acceptance rate, and a sales team that stops trusting the source. Targeting accepted leads or opportunities created removes the incentive entirely, since neither improves when a definition is loosened.

Ignoring the feedback loop

A scoring model built once and never revised will be wrong within a year, because the market, the product and the content all move. The correction mechanism is recording why sales rejected a lead, and comparing score bands against eventual outcomes. Two quarters of that data will usually reveal that one or two attributes carry most of the predictive value and several carry none.

The single most informative number is the acceptance rate: the share of qualified leads that sales agrees to work. A very low rate points at the model or the definition. A very high rate, close to everything being accepted, often means the threshold is too conservative and good contacts are sitting in nurture.

After that, look at conversion by source and by score band. If leads from one channel qualify at the same rate but convert at half the rate of another, the scoring model is not capturing whatever distinguishes them. And check volume against capacity: a definition that generates far more leads than the team can call will be re-applied informally by individual salespeople, at which point the whole exercise has been replaced by preference.

These four labels mark different points and are frequently used as though they were synonyms.

A fifth term, the product qualified lead, sits outside this sequence. It is based on what somebody has actually done inside a product during a trial or freemium tier, which is a considerably stronger signal than any marketing engagement, and it usually justifies its own route rather than being folded into the same threshold.

Qualification is a scoring problem sitting on top of a data problem. Start with what a CRM is, then look at lead management software for capture and routing, AI CRM for behavioural scoring, sales automation for nurture below the threshold, CRM for small business, and sales use cases.