Dear MQL haters.

Open LinkedIn on any given day and you'll find someone declaring that MQLs are a relic. A gumball machine. A vanity metric that floods sales with junk and gives marketing something to celebrate that doesn't actually matter. The criticism has become so mainstream that bashing MQLs is practically a content strategy.

And some of that criticism has a legitimate kernel. If you're chasing leads that aren't a good fit — people at the wrong companies, with no buying authority, who downloaded a white paper on a whim — then yes, your MQL process is generating noise, not signal. Nobody wants that.

But here's what gets lost in the bashing: the people who built lead scoring cared. They had meetings. They debated which behaviors should carry more weight. They thought carefully about target personas, account tiers, and threshold criteria. Building a good scoring model is detailed, thoughtful work — and the teams that did it well created something genuinely useful. Reducing all of that to “the gumball machine” is dismissive of real effort and real expertise.

It's also worth asking who benefits from declaring MQLs dead. In many cases, the loudest voices are people selling an alternative — an account-based platform, a new methodology, a different software category. They need MQLs to be the villain so their product can be the hero. That's not objective analysis. That's competitive positioning dressed up as thought leadership.

So if the concept of scoring and prioritizing inbound leads isn't inherently broken — and it isn't — then what actually went wrong?

The marketing qualified lead was designed to do something simple and useful: put a post-it note on an inbound lead that says “this one is worth your attention.” That's it. It's a prioritization signal. A way for marketing to say to sales: out of all the leads that came in this week, these are the ones you should look at first.

When done right, that prioritization is built on three dimensions:

Here's what nobody talks about when they bash MQLs. The reason the scoring model got gutted isn't a marketing problem. It's an organizational power dynamic.

Sales teams are — understandably — terrified of missing opportunities. They don't want any system sitting between them and a potential deal. Even if they helped design the scoring model. Even if they agreed to the thresholds. The anxiety is always the same: what if the scoring is wrong and I miss a deal because marketing held it back?

That fear is real and it's legitimate. Nobody wants to find out after the fact that a hot lead got buried because the scoring model underweighted it. And in most organizations, sales has more political power than marketing. So when sales says “don't hold anything back” — the scoring model bends.

But here's the part that doesn't get said out loud: sales is now doing the scoring manually. Every lead that comes over, the AE or SDR opens it up and makes a judgment call — is this the right person? Is this a good company? Is this worth my time? They're doing exactly what the scoring model was designed to do, except now they're doing it one lead at a time, in their heads, with no consistency and no scale.

The scoring didn't disappear. It just moved from a system to a person. And it got worse in the process.

This is the part that's hard to say diplomatically, but it's true: MQLs got a bad reputation largely because the people with the most power to influence the scoring model are the same people who demanded it be gutted. Sales pushed the bar to the floor, and then complained that what came over wasn't good enough. The concept gets blamed for a failure that was imposed on it.

Meanwhile, the people writing LinkedIn posts about “the MQL gumball machine” are rarely acknowledging this dynamic. It's easier — and gets more engagement — to declare a concept dead than to explain the organizational politics that broke it. The nuance doesn't fit in a headline.

And the marketers caught in the middle? They know the scoring should be better. They know three-dimensional scoring would produce higher-quality leads. But they also know that the last time someone tried to hold leads back for better scoring, sales escalated it, and the scoring model got stripped down again. So they send everything over, take the criticism, and move on.

Scoring and prioritizing inbound leads is not a broken idea. It's a necessary one. The alternative — sending every lead to sales unscored and letting individual reps make quality judgments one at a time — is more expensive, less consistent, and harder to optimize. The path forward isn't to abandon MQLs. It's to reclaim the three-dimensional model and give it the organizational backing it needs to work.

But the MQL debate doesn't stop at scoring. There's a broader industry conversation happening — and it's worth understanding what's actually useful in it and what's just noise.

The industry has framed this as an either/or. You're either an MQL-driven organization (focused on individual leads) or an ICP-driven organization (focused on target accounts). Pick a side.

The ICP approach — ideal customer profile — says: define your target accounts first. Use data to identify the 200, 500, 1,000 companies that are the best fit for your product. Tier them into enterprise, commercial, and SMB. Filter by company size, geography, tech stack, industry. Then focus your marketing and sales effort on those accounts, not on random inbound leads from companies you'd never sell to.

It's a good idea. It's also not new. This is database marketing — a practice that's been around for decades. The only thing that's changed is the branding. “ICP” sounds more modern than “target account list,” but the underlying concept is the same: know who you're going after and focus your resources there.

The critics position ICP as the antidote to MQLs. Stop chasing individual leads, they say. Focus on accounts. And they're not wrong about the direction — account-level thinking is important. But framing it as a replacement for lead scoring is where the argument falls apart. You still need to know which individuals at those accounts are engaging, what they're doing, and how to prioritize them. That's scoring. That's what MQLs were designed to do.

When you shift to an account-based approach, the terminology changes. Instead of a marketing qualified lead, you start talking about a marketing qualified account — an MQA. It sounds like a different concept. But when you look at what's actually underneath, it's the same building blocks.

Each MQL has three dimensions: behavior fit, person fit, and account fit. When you have multiple MQLs at the same company, the account fit circle is shared — they all work at the same place. What varies is the behavior (what each person did) and the person fit (what role each person holds). Stack those individual MQLs together and you get an MQA — a view of collective engagement at a single account.

This is why the MQL-vs-ICP debate is so frustrating. If you do MQLs right — with all three dimensions — you're already capturing the data you need to build an account-level view. The MQA isn't a competing framework. It's what happens when you aggregate well-scored MQLs at the same company. The individual scoring feeds the account-level picture.

And when you see an account where multiple people have strong behavior fit, strong person fit, and the account itself is a tier-A target — that's your qualified buying group forming. You didn't need to abandon MQLs to get there. You needed to do them well and then look at them from a higher altitude.

Here's where the conversation gets genuinely useful. The best idea to come out of this whole debate isn't ICP itself — that's table stakes. It's the concept of the qualified buying group.

The insight is simple but powerful: B2B decisions aren't made by individuals. They're made by groups of 3 to 5 people — sometimes more — who collectively evaluate, champion, and approve a purchase. If you're only tracking individual leads, you're seeing fragments. If you're only tracking accounts, you're seeing the container but not what's inside it. The buying group is the unit that actually matters.

The qualified buying group takes the best of both worlds. It uses ICP to define which accounts matter. It uses behavior and person-fit scoring to track what's happening at the individual level. And then it rolls it up to the buying group — asking not “did one person do something?” but “is a group of decision-makers at a target account collectively showing intent?”

That's a fundamentally better question. And it's the one the industry should be organizing around — instead of arguing about whether MQLs are dead.

The MQL isn't dead. The ICP isn't a revolution. And the LinkedIn hot takes aren't helping. What's actually useful is much simpler than the debate makes it seem:

First, score your leads properly — with all three dimensions (behavior, person, account), not just behavior. That's the MQL done right. Second, define your target accounts and tier them. That's ICP — good practice, not a paradigm shift. Third, track and prioritize buying groups, not just individuals. That's the qualified buying group — the real evolution that bridges the gap between lead-level signals and account-level strategy.

None of these are competing ideas. They're layers of the same system. And the teams that treat them as complementary — instead of picking sides in an industry debate — are the ones that end up with a demand engine that actually works.

Whether you call it MQL, MQA, qualified buying group — the label doesn't matter. The underlying principles have always been sound: score behavior, evaluate the person, consider the account. What's changed is that you're no longer limited to what you can manually build inside a marketing automation tool.

And let's be fair — the MAPs were actually pretty good at giving you the scaffolding for scoring. That was one of their genuine strengths. But if you've set up scoring a few times, you know the ceiling. The rules are static. The data sources are limited to what the MAP can see. The output is a number — 25, 50, 75 — that means different things to different people. It works, but it's basic.

What's happening now is that the same core principles are being executed at a level that wasn't possible before, because three things have converged:

This isn't a new framework replacing an old one. It's the same framework — behavior, person, account — with the execution constraints removed. You're not throwing out MQL scoring. You're doing it in a way that's no longer limited by what one person can manually configure in a MAP. More inputs, smarter analysis, and output that a sales rep can actually act on without needing to interpret a number.

The predictive scoring companies were the first wave of this — applying statistical models to lead prioritization. Good idea, early execution. What's happening now takes that further, with AI that can process a much wider set of signals in real time and explain its reasoning in plain language.