MQL vs SQL: why the handoff number lies and what to track instead

The marketing qualified lead was invented for a reasonable purpose. Sales said marketing sent junk. Marketing said sales did not follow up. Somebody proposed a threshold, both sides agreed to it, and the argument stopped. That is genuinely useful. It is also the entire origin story, and it explains everything wrong with how the number is used now.

An MQL is a definition of when marketing is allowed to stop caring. It is a handoff marker. Somewhere along the way it became a target, then a board metric, and now it is the number that determines budget for a function whose job is to create demand rather than count form fills.

MQL vs SQL is the wrong question

Most teams that ask about MQL vs SQL are really asking why the handoff feels broken. The honest answer is that the MQL is not one population, it is several, stitched together by a scoring model that adds points for behaviours with completely different meanings. A person who downloaded a pricing sheet and a person who downloaded an ebook about industry trends both get points. Only one of them is close to buying.

The arithmetic nobody puts on the slide

Take a plausible mid market SaaS quarter: 900 MQLs, of which 620 are content downloads, 190 are webinar registrations, and 90 are demo requests. Sales works the list and books 74 meetings. Of those, 68 came from the 90 demo requests.

If you set a target of 1,100 MQLs next quarter, you already know which lever gets pulled. Nobody triples inbound demo requests in ninety days. They gate another report and buy more traffic, and the number goes up while pipeline stays flat.

Every metric that averages two different populations will be gamed by whichever population is cheapest to grow.

What the MQL hides

• Intent. A person who asked to speak to you and a person who wanted a PDF are not the same person, and no scoring model reliably tells them apart after the fact.
• Timing. Most of the market is not buying this quarter. Counting them as leads makes an audience problem look like a conversion problem.
• Everything unmeasured. The buyer who read three of your pages, heard you mentioned in a community, and typed your name into a browser six weeks later shows up as direct traffic and gets no credit.

MQL vs SQL vs the numbers that actually predict revenue

The two numbers I would report instead

1. 1

   Hand raisers, counted honestly

   Demo requests, pricing enquiries, trials that reached activation, inbound replies that asked for a call. One number, no scoring model, no blending with content downloads. This is the closest thing marketing has to a real time demand signal.

2. 2

   Qualified pipeline created, dated to creation

   The currency finance already speaks. Date it to when the opportunity was created rather than when it closed, so you are measuring this quarter's marketing rather than last year's.

Everything else is diagnostics. Traffic, downloads, engaged accounts and impressions all belong in the working dashboard where the team makes decisions. None of them belongs in front of a CFO, because none of them survives the follow up question.

The objection, and the answer

The volume metric feels safer because it moves smoothly and always has an explanation. Smooth numbers are comfortable precisely because they are disconnected from a market that does not move smoothly at all.

What to do with the scoring model you already built

You do not have to delete it. A behavioural score is still useful for prioritising who sales calls first inside a list of hand raisers, or for deciding which accounts get a sequence versus a cold outbound touch. The mistake is not having a score. The mistake is putting the score's total on the same slide as revenue, as if the two were the same kind of number.

Keep the score as a sorting tool inside the working team. Keep hand raisers and pipeline created as the two numbers that leave the room. That separation is the entire fix, and it costs nothing to implement beyond one uncomfortable meeting where you tell finance the MQL number they have been tracking for two years is retiring.

Why the MQL survives even when everyone knows it is broken

The honest reason MQLs stick around is not analytical, it is organisational. A marketing leader who reports hand raisers is reporting a small, honest, sometimes flat number. A marketing leader who reports MQLs can almost always find a lever, another gated asset, another webinar, another paid social push, that makes the number go up by next month. Volume metrics survive because they are easy to defend in a meeting where nobody wants to say the quarter was quiet.

Sales carries its own version of the same incentive. A large MQL number gives a sales leader cover when the quarter underperforms: the pipeline was thin because marketing sent poor leads, not because reps missed activity targets. Both sides keep the number alive because both sides can use it as an excuse when the real conversation, about whether the product and the market fit, gets uncomfortable.

How to migrate off MQL reporting without a fight

1. 1

   Run both numbers in parallel for one quarter

   Do not cut the MQL cold. Report it alongside hand raisers and pipeline created for one full quarter so everyone can see the gap between the volume story and the revenue story with their own eyes, rather than being told about it.

2. 2

   Bring the conversion table, not the argument

   The table showing 810 content leads producing 6 meetings against 90 demo requests producing 68 does more persuading in five seconds than any slide of reasoning. Numbers that contradict each other in the same room end the debate faster than opinions do.

3. 3

   Let sales pick the qualification bar for hand raisers

   If sales helps define what counts as a hand raiser, the number becomes a shared target instead of a marketing invention. This is the same move that made the original MQL work when it was invented, applied to a better definition.

4. 4

   Retire the MQL from the board deck first, the CRM last

   You do not need to rip the scoring field out of the CRM to stop reporting it upward. Change what leaves the building before you change what the sales team sees day to day.

What this means for how marketing spends its time

Once hand raisers and pipeline created are the numbers that matter, the calendar changes on its own. Gating another report to hit an MQL target stops making sense, because gating adds volume leads, not hand raisers. Time moves toward the pages and channels that produce demo requests directly: comparison pages, pricing pages, case studies with real numbers, and the paid campaigns that target people already searching for a solution rather than people mid-research on a broad topic.

It also changes what a good quarter looks like to report internally. A quarter with 40 hand raisers and 24 qualified opportunities is a better quarter than one with 1,200 MQLs and the same 24 opportunities, even though the second one looks more impressive on a chart with more bars in it. The chart with fewer bars is the one that is telling the truth.

The self reported field that does more work than the scoring model

One field on a demo request form beats most of what a scoring model tries to infer indirectly. Ask how the person heard about you, in a free text box, not a dropdown of channels somebody in marketing ops picked eighteen months ago. Read the answers weekly. Buyers name podcasts, colleagues, a specific comparison page, a competitor's sales rep who irritated them, in language no attribution model would ever produce, and that language tells you what to fund next far faster than a multi touch report does.

This is cheap to add and expensive to ignore. Most teams that add the field stop reading it after the first month, because it does not roll up into a dashboard automatically. Reading forty rows of free text a week is not glamorous work. It is also more accurate than the attribution software most of those same teams pay for.

The scoring math, worked through

Most MQL models look reasonable on the whiteboard. Five points for a pricing page visit, three for a webinar attendance, two for opening three emails, ten for a demo request. The problem shows up once you run real behaviour through the formula rather than the example the vendor used in the demo.

How to run the audit yourself

1. 1

   Pull every MQL from the last full quarter into one sheet

   Export the raw list with the behaviour that triggered the score attached to each row, not just the final score.

2. 2

   Tag each row as intent or volume

   Demo requests, pricing enquiries, and trial activations are intent. Downloads, webinar registrations, and email opens are volume. Most CRMs already store this; the work is deciding the split once, in writing.

3. 3

   Calculate the meeting rate for each group separately

   Do not average them. The gap between the two rates is usually large enough that nobody needs a statistician to see it.

4. 4

   Take the two numbers to the next pipeline review

   Not as an argument, as a fact. Let the room draw its own conclusion about which number belongs on the board slide.

The question nobody asks in the pitch meeting

The question I get most often once a team has seen this split is whether volume signals are worthless. They are not. A person who reads three blog posts a week for two months and then requests a demo has been shaped by that volume, and I would not want a scoring model to ignore it. The mistake is not tracking volume. The mistake is reporting it upward as if it were the same currency as an opportunity, when the two numbers behave nothing alike once you separate them. Track both if you want, but never let them share a row on the same slide again, because the moment they share a row the volume number will always win the argument on size alone.