MQL vs SQL: Definitions, Handoff Rules and the Numbers That Show Where Leads Die

Only 13% of B2B SaaS marketing qualified leads become sales qualified leads, according to First Page Sage's 30-industry table of client data, and the best industry in that table reaches 26%. The rest are rejected, ignored or left to go cold between two teams that often define the terms differently. This MQL vs SQL guide covers the definitions, the handoff rules that keep leads alive, the conversion rate formula and benchmarks, and one real cohort: Spona's 400 MQLs from a two-month window. One note before the numbers: SQL here means sales qualified lead, not the database language.

Get thousands of leads in 4 minutes.

Stop stitching tools together.
Start with verified sales data, delivered in minutes.

SCALE SALES NOW

No contracts · No setup · No onboarding

Where leads die between MQL and SQL: the funnel in numbers

A B2B funnel has five stops: lead, marketing qualified lead (MQL), sales accepted lead (SAL), sales qualified lead (SQL) and opportunity. Most published numbers measure the MQL to SQL step, and the spread is wide. In First Page Sage's table, legal services and real estate sit at 10%, B2B SaaS at 13%, cybersecurity at 15%, and business insurance and HVAC at 26%. The page gives no overall average; the unweighted mean of its 30 rows is 16.1% (our arithmetic).

The older Implisit analysis of hundreds of company pipelines, summarized by Geckoboard, found that 13% of leads became opportunities and that the journey took 84 days on average. Source mattered most: website leads converted at 31.3% and referrals at 24.7%, while purchased lead lists reached 2.5% and email campaigns 0.9%.

Two more numbers explain why the step fails. Forrester's State of Business Buying, 2024, based on more than 16,000 business buyers, found that 91% of purchases stall at some point. In a Gartner Peer Community survey of 200 sales and marketing professionals, 27% of the marketers named "misaligned on internal definitions" as the top alignment challenge, and only 3% of all respondents said their teams had built common metrics.

Every row has an owner and an exit criterion. Where a company cannot fill in both cells, that is where its leads die.

What is an MQL and what is an SQL?

The terms come from the SiriusDecisions Demand Waterfall, a funnel model first published in 2005, revised in 2012 and replaced by the Demand Unit Waterfall in 2017. What MQL and SQL mean in practice depends on who writes the definition; the MQL meaning below comes from the vendors first, then from Spona's own records.

What is an MQL? MQL meaning and definition

HubSpot's lifecycle documentation defines a marketing qualified lead as "a contact or company that your marketing team has qualified as ready for the sales team". Adobe's version is looser: someone who has expressed interest in your products or solutions. Both leave the test to you, and that is where the definition problem starts: an ebook download and a pricing request are both "interest".

Spona's definition is narrower on purpose. A Spona MQL is an ICP fit company with an identified relevant role and an explicit sample, pricing or demo request, excluding duplicates and invalid signups. That produces fewer MQLs and a higher conversion rate, a trade-off the benchmarks section returns to. For the wider qualification stages and frameworks, see what lead qualification is.

What is a sales qualified lead (SQL)?

HubSpot defines an SQL as "a contact or company that your sales team has qualified as a potential customer", with the detail recorded in the Lead Status property. Salesforce's Canada hub lists four conditions: a business case, a contact who can facilitate the decision, a direct or latent need and a timeline. Cognism adds a fast-track rule: a lead who asks for contact, a proposal or pricing becomes an SQL at once and goes to an experienced closer "ideally within minutes, not hours".

Spona's SQL definition: sales confirms a concrete use case, sufficient data fit, a plausible purchasing path and an agreed evaluation or purchase step. The last clause matters most. Without an agreed next step, a good conversation is still an MQL.

HubSpot MQL vs SQL lifecycle stages

In HubSpot, Marketing Qualified Lead and Sales Qualified Lead are values of the Lifecycle Stage property, Opportunity means a contact or company associated with a deal, and the default automation only moves the stage forward. A rejected MQL therefore needs a manual or workflow reset, or it stays an MQL in every report. In Salesforce, teams usually track the stage in the Lead Status field, and converting a lead creates an account and a contact, plus an opportunity by default.

MQL vs SQL: the difference in one table

The key difference between MQL and SQL is who decides and on what evidence. An MQL is a prediction a system makes from fit and behavior. An SQL is a judgment a person makes after speaking to the buyer and hearing a next step. A score can hand a lead over; only a conversation can qualify it.

MQL to SQL handoff rules that keep leads alive

The handoff is where the 27% definition problem turns into lost pipeline. Five rules, each with a source, cover most of the damage.

1. Write both definitions down and have both teams sign them. The MQL definition states the fit criteria and the triggering behavior or request; the SQL definition states what the rep must confirm. Review them in the alignment meeting the lead qualification checklist describes, and change them when the data says so: HubSpot's rule of thumb is that a rate below 10% means the MQL criteria are too loose.

The sales accepted lead (SAL) step between MQL and SQL

2. Add an accept-or-reject step, and make the reject reason mandatory. A sales accepted lead is an MQL a rep has reviewed and agreed to work. The step costs one click and produces the most useful dataset marketing can get: why sales said no. 4Thought Marketing recommends a structured reason field with values such as wrong fit, bad timing, already a customer and competitor, and counts a lead as an SQL only once sales takes "a deliberate action in your CRM". Smaller teams often collapse MQL and SAL into one step; keep the reason field even then.

3. Set a response window, and make it short. 4Thought's common starting point is 24 hours for high-intent signals and 48 to 72 hours for lower-intent ones. The research says faster. The Lead Response Management study (Professor Oldroyd of MIT with InsideSales.com data: three years, six companies, more than 15,000 leads and 100,000 call attempts) found a 21-fold drop in the odds of qualifying a prospect when response time stretched from 5 to 30 minutes, and a fourfold drop between 5 and 10 minutes. A 24-hour SLA is the slowest response the whole team may give, not the target for a pricing request.

4. Recycle rejected MQLs with their reason attached. "Bad timing" goes back to nurture with a date, "wrong fit" tightens the MQL rule, "already a customer" fixes the CRM. Score thresholds and decay belong in the lead scoring best practices post; the handoff rule is only that nothing leaves the SAL queue without a reason.

5. Put both teams on one dashboard. Only 3% of the Gartner Peer Community respondents said their teams had common metrics. Understory Agency's three handoff metrics are a workable minimum: first-touch SLA compliance, the share of MQLs worked to a disposition, and conversion rate by the event that triggered the MQL.

MQL to SQL conversion rate: formula and benchmarks

The formula is simple: MQL to SQL conversion rate = SQLs ÷ MQLs × 100. The trap is the time window. Geckoboard's KPI guide warns that counting only MQLs that become SQLs within the same month understates the rate when the sales cycle runs longer. Measure a cohort instead: take the MQLs created in one period and count how many became SQLs within a fixed window, 30 days for most teams and longer for enterprise sales. HubSpot's worked example is 26 SQLs from 200 MQLs, or 13%.

Two other published figures bracket the table. HubSpot puts the cross-industry range at 10% to 20% and calls 13% healthy. Salesforce, citing Gartner research, says 21% of MQLs convert to SQLs; the underlying Gartner page could not be opened, so treat that figure as secondhand. Handle the rest of the search results with care: one agency page in the top ten gives 13%, 13 to 15%, 10 to 20% and 20 to 30% for the same metric, with no source for any of them. A benchmark without a sample size, a period and a definition is an opinion. The next section states all three.

Spona's MQL to SQL conversion rate: 400 MQLs, 120 SQLs, 30% in 30 days

Spona recorded every MQL created in a two-month window and followed each one until a month after the window closed, so every record has a complete 30-day follow-up. The cohort holds 400 MQLs. Sales qualified 120 of them as SQLs within 30 days, a 30% MQL to SQL conversion rate. Spona has confirmed the conversion and the sample against its own records.

The definitions are the two quoted in the definitions section, applied before counting: duplicates and invalid signups never entered the 400, and no lead counted as an SQL without an agreed evaluation or purchase step.

Why 30% sits above the 13% B2B SaaS figure is mostly definitional. First Page Sage counts any contact with purchase intent and the means to buy, without saying how intent is measured; Spona's MQL definition requires an explicit request, which does the filtering a SAL step would otherwise do. The 30-day window pulls the other way: an MQL accepted after day 30 counts as a miss.

What the number does not prove: it is one company, one two-month cohort and one product with a low entry price and a free sample. It is not a market benchmark, it is not an average for Spona's customers, and the next cohort may differ. What it shows is the method: a stated definition, a stated period, a stated window and a count a reader can check.

How to convert an MQL to SQL

Three moves raise the rate without loosening the definitions.

Respond in minutes to explicit requests. The 21-fold finding above is about the first five to thirty minutes. Route pricing, demo and sample requests straight to a rep with a task in the CRM, and keep the 24-hour SLA for content-only MQLs. HubSpot's client example scores a pricing sheet download at 25 points and a newsletter open at 1 to 2, against a 75-point threshold: an explicit request should carry a lead most of the way on its own.

Confirm three things in the first conversation. Spona's reps work from ten qualification questions; for a small purchase they prioritize three: need (which missing contacts, poor fit or outdated records are costing the team time), timing (which campaign or market launch creates the deadline) and decision criteria (what a 100 account sample must prove to justify buying). Budget and authority get a brief check, not a gate; full MEDDIC is reserved for larger annual or complex accounts.

Do not let a budget screen kill a fundable deal. One recorded Spona example: a founder said there was no data budget, so a rigid screen would have rejected the account. A small sample showed 10 hours of monthly research savings, which created a fundable business case. The order of operations is the lesson: prove the value, then ask about budget. For the fit side of the MQL rule, start from an ideal customer profile both teams agreed on.

When MQL vs SQL stops working: buying groups and self-serve buyers

The people who coined the terms moved on. Introducing the Demand Unit Waterfall in 2017, SiriusDecisions' Kerry Cunningham said "we knew that it wasn't leads anymore" and that accounts must be sourced at the top of the waterfall; Terry Flaherty added that demand needs "a buyer, a business pain and a solution". The data since then supports them. Forrester counts 13 people in the average buying decision. 6sense's 2025 Buyer Experience Report (just under 4,000 respondents) puts first contact with a seller at 61% of the way through the journey, and finds that 94% of buying groups rank their shortlist before engaging. Gartner reports that 75% of B2B buyers prefer a rep-free experience.

So a single-person MQL often arrives late, and the person who filled in the form may be the researcher rather than the decision maker. The fix is not to drop MQL and SQL but to change the unit: qualify the account, count engaged people per account, add product usage and buying signals alongside form fills, and let the SQL conversation map the group rather than one contact. A product qualified lead (PQL) is one such variant; the FAQ covers it.

Where Spona fits in MQL vs SQL, and where it does not

Spona works before the MQL stage. You describe your ideal customer in a chat; the AI mines 40+ data sources, validates every lead and returns qualified leads in about 4 minutes. One Head of Sales at a 20 to 50 person company put the requirement this way: "We wanted a small list that matched our niche, with a reason for every match." Ask every lead source for that reason: it makes a lead easier to accept at the SAL step, because the rep sees why it fits before the first call.

Pricing is per lead: qualified leads cost $0.32 to $4, bulk leads $0.01 to $0.09, each new ideal customer profile takes 1,000 setup credits ($8 to $10), and free sample leads are available. There are no contracts, no setup and no onboarding, so a first test can be 100 leads measured against your current MQL source on the same SAL and SQL criteria.

Where it does not fit: Spona does not run your lifecycle stages, SLA timers or CRM routing; HubSpot and Salesforce do that. It does not replace the conversation that makes an SQL. And a Spona lead is not an inbound request: it arrives fit-qualified, and your outreach has to earn the explicit interest signal that an inbound MQL brings with it. Teams that need account-level intent data will pair it with an intent provider.

Get thousands of leads in 4 minutes.

Stop stitching tools together.
Start with verified sales data, delivered in minutes.

SCALE SALES NOW

No contracts · No setup · No onboarding

MQL vs SQL FAQ

What comes first, MQL or SQL?

MQL. A lead becomes a marketing qualified lead when it meets marketing's fit and engagement rule, a sales accepted lead when a rep agrees to work it, and a sales qualified lead when the rep confirms need, fit, a buying path and a next step. HubSpot's default automation moves lifecycle stages forward only.

What is a good MQL to SQL conversion rate?

First Page Sage's industry table runs from 10% (legal services, real estate) to 26% (business insurance, HVAC), with B2B SaaS at 13%; HubSpot calls 13% healthy and reads anything below 10% as loose MQL criteria. Judge your own rate against your own definitions and a fixed cohort window, not against another company's.

How long does it take to move from MQL to SQL?

HubSpot's guidance is 30 to 90 days for most B2B sales and six months or more for complex enterprise deals. The Implisit analysis found 84 days from lead to opportunity on average. Spona's cohort used a 30-day acceptance window, which fits a low-priced product with a free sample.

What is the key difference between an SQL and an MQL?

Who decides, and on what evidence. An MQL is set by a score or rule in a marketing tool from fit and behavior. An SQL is set by a salesperson after a conversation that confirmed need, fit, a buying path and an agreed next step.

What is a PQL vs an MQL?

A product qualified lead is qualified by what someone did inside your product, such as reaching a usage limit on a free plan, rather than by content engagement. The handoff rules are the same: a written definition, a response window and a reject reason.

How do HubSpot and Salesforce label MQL and SQL?

HubSpot uses the Lifecycle Stage property (Marketing Qualified Lead, Sales Qualified Lead, Opportunity) with sub-stages in Lead Status. Salesforce teams usually track it in the Lead Status field; converting a lead creates account and contact records, plus an opportunity by default.