Product Qualified Lead: When Usage Is the Evidence

A product qualified lead is a person, or more usefully an organisation, that has used your product and done enough with it to suggest a genuine intention to buy. The evidence is what they did rather than what they read. Where a marketing score infers interest from page visits and downloads, product qualification observes somebody importing real data, inviting a colleague or completing a workflow end to end. Those actions cost effort, and effort is a far better predictor of purchase than attention. The term became widespread alongside product-led growth, where the product demonstrates its own value before any salesperson gets involved.

Choosing the signals

The right signals represent investment rather than curiosity. Importing data, connecting another system, inviting colleagues, completing a core workflow, returning on several separate days, and approaching a plan limit all indicate somebody is trying to make the product part of how they work. Logins, sessions and time in app are poor signals, because a confused user and an engaged one produce similar numbers.

Setting the threshold

Choose the threshold empirically. Compare what converting accounts did in their first two weeks against what non-converting accounts did, and keep the actions where the gap is large. Then set the bar where the resulting volume matches what your team can actually follow up, because a qualification rule that produces more accounts than anyone can contact will be applied informally by whoever looks first.

Account level as well as user level

For most business products the account is the more meaningful unit. One person exploring is weak evidence. Three or four colleagues from the same organisation active in the same week implies an internal conversation has already taken place, which is one of the strongest signals available and is invisible in a user-only view.

A worked threshold example (illustrative)

An account qualifies when three conditions are met within fourteen days: at least fifty records imported, at least two active users, and one core workflow completed from start to finish. Consider three trials. The first has imported two hundred records with four users but has not completed a workflow: they have set up and not yet reached value, so the right action is a prompt about the missing step, not a sales call. The second has one user who completed a workflow with sample data: they are evaluating, and belong in nurture. The third meets all three conditions and should reach a salesperson the same day, in a conversation that references exactly what they built. The numbers are illustrative; the discipline of defining them in advance and testing them against conversion is not.

The decision it drives is when a human should intervene in a self-serve motion. Contacting every trial is uneconomical and unwelcome; contacting none leaves revenue on the table, because accounts that reach real usage frequently need one conversation about pricing, security or rollout to proceed. A threshold turns that into a rule rather than a guess.

It also changes the content of the conversation. A salesperson approaching a qualified account already knows what the customer built, which problems they were solving and where they stopped. That removes most of the discovery a cold conversation would need and makes a much shorter, more specific interaction possible, which is usually what a self-serve buyer wants.

Treating signup as qualification

The most common error, and it undoes the entire benefit. A signup is a registration; a large proportion of registrations never return. Calling all of them produces the same low-quality experience as calling everyone who downloaded a guide, while consuming more goodwill because the person is in the middle of evaluating something.

Picking signals that flatter the product

It is tempting to build thresholds around features the company is proud of rather than actions that predict purchase. The test is always the same: did converting accounts do this more than non-converting ones? Where the answer is no, the signal is measuring enthusiasm inside the company rather than intent outside it.

Acting too slowly

The value of a usage signal decays in days rather than weeks. A message referencing something a person set up an hour ago lands very differently from the same message a fortnight later, when the trial has expired and the project has moved on. If qualification is reviewed weekly rather than triggered on the event, most of the advantage over a marketing score has already been given away.

Look at trajectory rather than totals. An account whose usage is rising across a fortnight is in a different position from one that had a burst on day one and has been quiet since, even where the cumulative numbers match. Breadth matters too: usage spreading across several people or several parts of the product is a stronger signal than depth in one place by one person.

Keep fit in the picture. Heavy usage by an organisation that cannot buy, a student, a competitor, a business in a segment you do not serve, is a cost rather than an opportunity, and a purely usage-based threshold will route it straight to a salesperson. Requiring both a usage threshold and a fit threshold avoids that without adding much complexity.

These four qualification labels rest on quite different evidence and justify different responses.

The model is not universal. It requires a product that can deliver value without help and enough self-serve volume to be worth modelling. Where implementation or training is needed before anything works, usage signals remain useful for prioritisation and expansion, and the qualification decision keeps resting on a conversation.

Usage signals are only actionable when they reach the people doing the follow-up. Start with what a CRM is, then look at AI CRM for behavioural scoring, sales automation for in-context sequences, lead management software, WhatsApp CRM for reaching trial users where they reply, and product features.