Lead Qualification & Routing

What Is a Marketing Qualified Lead (MQL)? Definition, Criteria and How to Measure It

Alec Phillips
November 20, 2026
Summarize using
SHARE

Short answer: A marketing qualified lead (MQL) is a contact your marketing team judges more likely to buy than other leads, based on who they are (fit) and what they've done (engagement), like downloading a guide or revisiting your pricing page. An MQL is a prediction, not confirmed intent, so judge it by the pipeline it produces.

‍

Your MQL number went up last quarter. Pipeline didn't move. Sales says the leads are cold, marketing says sales doesn't follow up, and both teams are reading the same dashboard.

That argument usually starts with a definition nobody wrote down. "Marketing qualified" means whatever your scoring model says it means, and most scoring models were built from guesses about which clicks predict a purchase.

This guide covers what a marketing qualified lead is, the criteria teams use to define one, how it differs from an SQL and an AQL, and how to measure marketing qualified leads so the number tells you something. If you're already past the definition and fighting the handoff, our breakdown of the MQL-to-SQL conversion rate goes deeper.

What are Marketing Qualified Leads (MQLs)?

Marketing qualified leads are the contacts marketing has decided deserve sales attention, or focused nurture, because they match your target buyer and have engaged enough to suggest interest. In most funnels they sit between a raw B2B sales lead and a sales qualified lead.

The stage name comes from the Demand Waterfall, the funnel model SiriusDecisions built for B2B teams and that is now part of Forrester. It gave marketing and sales a shared set of stages: inquiry, MQL, sales accepted lead (SAL), sales qualified lead (SQL) and closed deal. Forrester's guidance on the Demand Waterfall still starts with the same advice: agree on what counts as an MQL before you measure anything.

Most CRMs build the stage in. HubSpot's default lifecycle stages define a Marketing Qualified Lead as a contact or company your marketing team has qualified as ready for sales. The label is standard. What earns it isn't.

Here's a typical MQL. Say a demand gen manager at a 300-person software company downloads your benchmark report, opens two nurture emails and visits your pricing page twice in five days. Her title matches your persona and her company fits your ideal customer profile. Your scoring model adds up the points, she crosses the threshold, and she becomes an MQL.

What makes a lead marketing qualified?

A lead becomes marketing qualified when it crosses a threshold your team sets across two kinds of criteria: fit (does this person and company look like your buyer?) and engagement (have they acted like someone interested?). Most teams combine both into a single lead score.

Criteria typeCommon signalsWhat it suggestsWhat it can't tell you
Fit: the personJob title, seniority, departmentThey could influence a purchaseWhether they're looking now
Fit: the companyIndustry, employee count, revenue, tech stackThe account could buyWhether there's a project or budget
Engagement: contentReport or ebook download, webinar registration, email clicksInterest in the topicInterest in buying
Engagement: high intentPricing page visits, return visits, demo or contact requestPossible evaluationWhat problem they're solving or who else is involved
Negative signalsStudent or competitor email, careers page visits, unsubscribesProbably not a buyerLittle; these are usually right

A few rules keep the criteria honest:

  • Write the definition down. If sales and marketing can't both recite it, each team will apply its own.
  • Weight fit before engagement. A quiet account that matches your profile deserves more attention than an eager student.
  • Decay old activity. A download from eight months ago isn't interest today.
  • Pull hand-raisers out. Someone who asks for a demo shouldn't wait in the same nurture queue as someone who read a blog post.

Where the point values come from is its own problem. Our piece on how an MQL scoring model works shows how thresholds get set, and our roundup of the best lead scoring tools for B2B teams compares the software that runs them.

What's the difference between an MQL, an AQL and an SQL?

An MQL is marketing's prediction that a lead might buy. An SQL (sales qualified lead) is a lead sales has vetted and confirmed as an opportunity worth pursuing. An AQL (Agent Qualified Lead) is a buyer who has articulated their needs, demonstrated fit, and agreed to a next step in a conversation with an AI agent. Docket coined the AQL term.

The difference is who did the qualifying and what evidence they had. MQLs are inferred from clicks. SQLs are confirmed by a rep, usually after outreach and a discovery call. AQLs are confirmed by the buyer, in their own words, while they're still on your site.

MQLSQLAQL
Who qualifiesMarketing automation and scoring rulesSales or SDR teamAn AI agent in a live conversation
EvidenceFit data plus engagement (downloads, page views, email clicks)A discovery call against criteria like BANT or MEDDICThe buyer's stated need, fit and agreement to a next step
When it happensWhen the score crosses a thresholdAfter sales outreach and a callDuring the website visit
What sales receivesA contact record and an activity historyA qualified opportunityA booked meeting with the full conversation
Main riskActivity mistaken for intentSlow follow-up lets good leads go coldOnly as good as the traffic and the approved knowledge behind the agent

One buyer can look very different in each record. Our comparison of AQL vs MQL walks through the same visitor recorded both ways, and our guide to what an agent qualified lead is covers the criteria in full.

How do you measure marketing qualified leads?

Measure marketing qualified leads by what happens to them after they're created: how many sales accepts, how many become opportunities, and how much pipeline and revenue they produce. MQL volume on its own tells you how busy your marketing was, not whether it worked.

Most teams know this and report volume anyway. In The Trust Gap, a survey of 192 marketing, demand gen and RevOps practitioners that Heinz Marketing fielded for Docket in September 2026, nearly 90% said they report MQL volume to leadership, but only 7% of them believe it predicts pipeline. Our write-up of that survey on sales and marketing alignment has the full results.

These seven metrics tell you whether your MQLs are working:

MetricFormulaWhat it tells you
Lead-to-MQL rateMQLs ÷ new leadsHow strict your threshold is
MQL acceptance rateMQLs sales accepts ÷ MQLs passed to salesWhether sales trusts the definition
MQL-to-SQL conversion rateSQLs ÷ MQLsWhether the score predicts a real sales conversation
MQL-to-opportunity rateOpportunities from MQLs ÷ MQLsWhether MQLs turn into pipeline
MQL-sourced pipelineTotal opportunity value from MQLsMarketing's contribution in dollars
Time to first touchHours from MQL to first sales contactWhether the handoff is fast enough to matter
Cost per MQLProgram spend ÷ MQLsEfficiency, read only beside the conversion rates

How do you set an MQL benchmark?

Build your benchmark from your own last two to four quarters, split by source. Numbers borrowed from other companies rarely transfer, because every company sets its threshold differently. A loose threshold shows a high lead-to-MQL rate and a low MQL-to-SQL rate. A strict one shows the reverse.

  • Pull every MQL from the last six months. Keep the source and the score at handoff.
  • Match each one to its outcome. Accepted, SQL, opportunity, closed-won, or nothing.
  • Compare by source. Webinar MQLs and pricing-page MQLs rarely convert at the same rate.
  • Find the score band where conversion jumps. That's where your threshold belongs.

For the downstream numbers that predict pipeline better than MQL counts, see our list of B2B revenue metrics beyond MQL and the pipeline metrics for CMOs that boards ask about.

Why do MQLs often fail to predict pipeline?

MQLs often fail to predict pipeline because they score one person's activity, while B2B purchases are made by groups and the most telling signals never reach the score. Three gaps do most of the damage.

  • Buying is a group decision. In Forrester's Buyers' Journey Survey, 2023, 93% of B2B buyers were part of a buying group of two or more people, and 71% were in a group of four or more. As Forrester's analysts put it, "The individuals identified as MQLs are not buying alone."
  • Activity isn't intent. A student writing a thesis and a VP comparing vendors can download the same three assets and earn the same score. The score can't tell them apart. A short conversation can.
  • The questions that matter never reach the form. Buyers arrive with specific questions about pricing, integrations or fit. A form captures a name and an email, so the most useful information leaves when they do.

Forrester's own answer is the B2B Revenue Waterfall, which moves teams from individual leads to opportunities and buying groups in the CRM. Our post on why MQLs don't convert to pipeline covers the five most common causes inside a single funnel.

What do MQLs still do well?

MQLs still do a useful job as an internal marketing signal. They show which campaigns and channels attract people who fit your buyer, and they give marketing a countable stage in the CRM. The trouble starts when an MQL is treated as proof that a buyer is ready for sales.

  • They make campaigns comparable. You can see which programs bring in people who fit, quarter over quarter.
  • They give nurture a trigger. Crossing the threshold can start a sequence or an SDR task.
  • They're already built in. Every major CRM and marketing automation platform supports the stage, so reporting exists on day one.
  • Finance understands them. Budget models built on MQL volume are easy to explain.

Keep the MQL as a marketing metric. Retire it as the sales handoff. If you're weighing that change, our take on whether MQL is the right metric for B2B marketing teams lays out the trade-offs.

How do you improve your MQL definition?

Improve your MQL definition by rebuilding it from closed-won data with sales in the room, then adding at least one step where the buyer confirms intent directly. Five steps get you there.

  1. Start from deals you won. List the fit and engagement signals that appeared before those opportunities opened, and drop the ones that didn't.
  2. Agree on it with sales. Put the definition, the threshold and the follow-up time on one page, signed off by both leaders.
  3. Route hand-raisers straight to sales. Demo and contact requests skip the score.
  4. Add negative scoring and decay. Subtract points for non-buyer signals and let old activity expire.
  5. Add a confirmation step. Ask your qualifying questions when the buyer is most engaged, in a short conversation instead of a longer form.

Our roundup of how to improve your MQL definition collects ten practitioners' approaches, and our playbook on how to qualify website visitors without a form shows what step five looks like in practice.

How does Docket qualify buyers to create MQLs?

Docket, The Demand Platform, generates MQLs from website visitors engaging with conversational AI instead of traditional form fills and content downloads. Every conversation is written back to Salesforce, HubSpot, Dynamics 365 or Marketo, so leads arrive with what the buyer actually said attached, not just a score.

What the buyer asks the Docket Demand Agent (pricing, integrations, rollout timing) lands on the lead record, where your scoring and routing rules can use it. Buyers who go further become AQLs (Agent Qualified Leads) and book a meeting on the spot. The Buyer Context Graph builds qualification across visits.

Instead of a contact record and a score, sales gets the conversation: what the buyer asked, the problem they described, how they fit your criteria and the meeting they booked, written back to Salesforce, HubSpot, Dynamics 365 or Marketo. The website agent also recognizes returning visitors and their company, so buyers don't start over. That memory lives in the Buyer Context Graph, Docket's continuously updated profile of each visitor and their company.

Factors.AI booked 23 meetings in two weeks with Docket's website agent, over 5x their baseline, and 77% of those meetings were booked outside business hours.

This is the inbound side of what buyers now search for as agentic marketing: AI agents that act when the buyer acts, inside rules you set. You don't have to switch off your MQL model to start. Run AQLs alongside your MQLs for a quarter and compare which ones sales accepts and which ones become pipeline. Our guide to AI lead qualification covers how to set up that test.

FAQ

What does MQL stand for?

MQL stands for marketing qualified lead. It's a lifecycle stage for contacts that marketing has judged likely enough to buy that they deserve sales attention or focused nurture.

What is an example of a marketing qualified lead?

A director of marketing at a mid-size software company who downloads your benchmark report, attends a webinar and visits your pricing page would be an MQL in most scoring models. She fits your ideal customer profile and has engaged several times, but she hasn't asked to talk to sales or confirmed a project.

What is the difference between a lead and an MQL?

A lead is anyone who has shared contact information, for example by subscribing to a newsletter. An MQL is a lead that also meets your fit and engagement criteria, so marketing believes they're more likely to buy than the average lead.

What is a good MQL-to-SQL conversion rate?

It depends on how strict your MQL threshold is, so a number from another company won't tell you much. Track your own rate by source over several quarters, then move the threshold to the score band where conversion to opportunity jumps.

Are MQLs still relevant in 2026?

Yes, as an internal marketing metric for comparing campaigns and triggering nurture. As a sales handoff they're weaker, because most B2B purchases involve a buying group and a score can't confirm intent. In Docket and Heinz Marketing's Trust Gap survey, respondents named qualified pipeline, revenue and SQLs as primary top-of-funnel indicators more often than MQLs.

Who owns the MQL definition?

Marketing usually owns the scoring model, but the definition only works if sales agrees to it. Write it down together, including how quickly sales commits to follow up, and review it every quarter against closed-won deals.

Turn more interest into pipeline

Your buyers already showed up and showed interest. Whether that interest becomes pipeline depends on how quickly someone finds out what they need.

Want to see how that works on your site? Book a demo, or ask Aura, Docket's own agent, a question on docket.io.

The Trust Gap

New research fielded by Heinz Marketing on behalf of Docket, based on 192 marketing, demand gen and RevOps practitioners. Three handoffs, one number almost nobody believes, and standard any new lead source will have to meet.
Check it out