MQL adoption as a primary metric fell from 72.0% to 55.3% in a single year (6sense, 2024). High-performing B2B teams responded with one decision: demote the MQL to an early signal, and report and optimize against sourced pipeline and cost per opportunity instead. It counts interest; the business needs intent.

This is not a lead-scoring tweak. Up to 75% of buy-side leaders say their core measurement approaches underperform (IAB State of Data, 2026), and the MQL is the most visible symptom. The rest of this covers why it broke and exactly what to put on the board in its place.

Is the MQL dead?

55.3% of B2B teams still run the MQL as a primary metric, down from 72.0% a year earlier (6sense, 2024). So the MQL is not dead, it is demoted: a 16.7-point fall for a metric that ran demand gen for a decade. What changed is not the definition of a lead. It is what teams trust the number to tell them.

An MQL was always a proxy. It said a person crossed a scoring threshold, downloaded the thing, opened the emails. That was useful when the buying journey ran through forms and gated content. The proxy held. It is holding less well now because most of the buying journey happens before a form is ever touched, which means the MQL is measuring the visible slice of a mostly invisible process.

So the honest read is a demotion, not a funeral. Plenty of teams still generate MQLs and still find them useful as an early signal. The high performers stopped treating that signal as the scoreboard.

Why does the MQL mislead demand gen teams?

13% is the average MQL-to-SQL conversion rate, so hitting 100% of an MQL goal can deliver only about 30% of the pipeline target (The Digital Bloom, 2025). That gap is the core problem. An MQL number can climb while the pipeline the business needs stays flat, because the two are only loosely connected.

The MQL rewards volume at the top of the funnel, where volume is cheap and easy to manufacture. Lower the score threshold, run a content offer, and MQLs climb. None of that guarantees a single additional deal. When marketing hits its MQL goal and sales still misses quota, the argument that follows is predictable: marketing says it delivered leads, sales says the leads were junk, and nobody can settle it because the metric everyone agreed to doesn't map to revenue.

We have watched this play out inside enterprise demand programs more than once. The dashboard is green, the MQL target is met, and the CRO opens the quarterly review by asking how much pipeline marketing sourced. The MQL number cannot answer that question. It was never built to.

What should B2B teams measure instead of MQLs?

50x is the cost-per-opportunity reduction one B2B security company (Armorblox) saw after switching from MQL-based lead scoring to a cost-per-opportunity model, from roughly $40,000 to about $800, while lead-to-conversion rose from around 2% to over 30% (Metadata.io / Armorblox, 2024). Cost per opportunity is the metric that replaces the MQL as the primary target. The metric they optimized against changed, and the economics followed.

The shift is from counting interest to counting intent and money. Instead of "how many people crossed a score threshold," the questions become: how much qualified pipeline did we create, what did each opportunity cost to generate, and how fast are deals moving between stages. Those metrics are harder to game because they only move when something real happens in the CRM.

The Armorblox result is not a promise that every team gets a 50x cut. It is proof that moving the optimization target changes what the media buys. When you optimize toward opportunities rather than form fills, the algorithms and the budget chase the audiences that actually buy, and cost per lead in that case fell from about $1,000 to roughly $50 as a byproduct.

MQL model vs. pipeline and opportunity model

Both models measure something real. The difference is what each one lets you optimize and what the board sees.

Why did B2B ad measurement break in the first place?

Up to 75% of US buy-side leaders say their core measurement approaches, including attribution analysis, incrementality tests, and marketing mix models, underperform (IAB State of Data, 2026). The MQL problem is one symptom of a broader measurement crisis: the tracking got harder and the buying got more complicated at the same time.

The buying side moved first. A typical B2B buying group is now 6 to 10 decision-makers, each gathering 4 to 5 pieces of research independently (Gartner, 2023). No single form fill, and no single scored lead, represents that group. The MQL captures one person raising a hand while the other eight members of the committee research silently and never fill anything out. The metric sees a fraction of the buying reality and reports it as the whole.

Measurement also got structurally harder as signal loss, longer cycles, and fragmented channels piled up. That is why so many leaders now rate their own attribution and mix models as underperforming. The response we are seeing is not a better lead score. It is a move toward outcome metrics that hold up even when the path to them is impossible to fully trace.

How do you make the internal case to move off MQLs?

13% is the number to put in front of the board: that is the average MQL-to-SQL conversion rate, so an MQL goal can be fully met while pipeline lands near 30% of target (The Digital Bloom, 2025). That comparison reframes the debate from "is the MQL bad" to "does our goal predict the revenue we owe the board." It almost never does.

Then propose a parallel-run, not a rip-and-replace. Keep generating and scoring leads, but add cost per opportunity and sourced pipeline as the metrics you report up and optimize the media against. Run both for a quarter and let the CRO see which number tracks the deals that actually closed. The MQL becomes a diagnostic signal; pipeline economics becomes the scoreboard.

One move: rebuild this quarter's marketing dashboard so the top line is sourced pipeline and cost per opportunity, and demote MQLs to a supporting row. If the top line can't answer "how much revenue did marketing influence," it isn't the right top line.

This is the work we do with enterprise demand teams: re-anchoring the measurement model on pipeline and opportunity economics, then rewiring the media and the CRM stages so the numbers the board sees are the numbers the business runs on. Moving Parade builds the optimization target first, then buys against it, because the metric you optimize toward is the one you get more of.

Frequently Asked Questions

What single metric should replace the MQL on the board?

2.4 is the average number of metrics a marketing team reports to its board (6sense, 2024). Spend those slots well: make sourced pipeline the top line and cost per opportunity the efficiency line, and drop the MQL to a supporting row. If the top line can't answer "how much revenue did marketing influence," it is the wrong top line.

Why can a rising MQL count still miss real buyers?

84% of B2B buyers pick a preferred vendor before ever contacting sales (6sense, 2024). A form-fill MQL captures a hand-raise that often comes after the decision is effectively made, and misses the buyers who researched quietly and never filled anything out. Volume at the top does not mean influence where it counts.

What has to be true in your CRM before switching to cost per opportunity?

Two things gate the switch: clean opportunity stages and reliable stage-entry dates. Cost per opportunity and stage velocity are only as trustworthy as the CRM behind them. Fix stage definitions and hygiene first, or the new scoreboard inherits the old data problems and the revenue leader stops trusting it.

How long does it take to move off the MQL?

Run both models in parallel for one quarter. Keep scoring and generating leads, but report sourced pipeline and cost per opportunity alongside the MQL. Let the revenue leader watch which number tracked the deals that actually closed, then make pipeline the scoreboard and the MQL a diagnostic signal.

When is the MQL still worth keeping?

Keep the MQL as a routing and early-warning signal, especially in higher-velocity motions where a fast hand-raise genuinely predicts a deal. The move is demotion, not deletion: an MQL is a useful leading indicator when it feeds a pipeline scoreboard, and a misleading one when it is the scoreboard.