What Should Replace the MQL?

Should you still be reporting MQLs?

Report them internally if your process depends on them, but stop treating the number as a result. An MQL counts one person raising a hand. The thing you actually sell to is a group of people at one company, and a count of hands tells you almost nothing about whether that group is moving.

This is not a new complaint, and most marketing leaders already half agree with it. What is usually missing is a replacement that fits the tools a team already owns, so the MQL survives by default because nothing else is defined.

So here is a framework built from properties that already exist in a standard CRM, with the three measures we would put in its place.

What is an MQL supposed to mean?

A handoff, not a quality judgement. HubSpot's lifecycle stage documentation defines a Marketing Qualified Lead as a contact or company that your marketing team has qualified as ready for the sales team, and a Sales Qualified Lead as a contact or company that your sales team has qualified as a potential customer.

Read those two definitions together and the honest reading is that MQL means marketing thinks this is worth a conversation. It is a routing decision. It was never designed to be a measure of pipeline quality, and it behaves badly when used as one.

Notice also that HubSpot's definitions say contact or company. The stage can apply at the account level, which most teams never use. That one detail is the beginning of the fix.

Why does the MQL break down?

Three reasons, and they compound. First, it counts individuals when the decision is collective, so five people from one target account look like five times the progress of one person from a perfect-fit account.

Second, it is usually triggered by engagement alone. Someone downloading a report is engaged with your content, which is a different thing from being in a buying process. A competitor's intern and a VP with budget generate identical records.

Third, the stage is sticky. HubSpot notes that default automatic updates to the lifecycle stage property will only move the stage forward, and that you must clear the value manually or via a workflow before setting an earlier value. So a contact who went quiet two years ago is still, technically, qualified.

What should replace it?

Three measures instead of one, because the MQL was doing three jobs badly. We would report account fit, buying group coverage, and sales acceptance, as separate numbers that cannot be averaged into a comfortable total.

Account fit answers whether this is a company you should want. Buying group coverage answers whether you have reached enough of the right people inside it. Sales acceptance answers whether the people who have to act on it agree it is real.

The reason to keep them separate is that each one fails differently. High fit with no coverage means a targeting success and an outreach failure. High coverage with low acceptance means marketing and sales disagree about what good looks like, which is a conversation, not a metric.

How do fit and engagement come apart?

Deliberately, and most CRMs now model it for you. HubSpot's lead scoring tool distinguishes an engagement score, which qualifies records based on their actions and interactions such as visiting your website, subscribing to your newsletter, clicking a CTA, or opening a marketing email, from a fit score, which qualifies records based on demographic information through property values such as their age, job title, company size, or annual revenue.

The combined score is where it gets useful. HubSpot describes a combined score as using both engagement and fit criteria, and documents a nine-tier value where letters indicate fit, with A high and C low, and numbers indicate engagement, with 1 high and 3 low. Its own example is that a low-fit but highly engaged contact would have a value of C1.

C1 is the record every MQL count quietly inflates. Very active, wrong company. Separating the two axes means you can see that pattern instead of averaging it away, and it costs nothing because the tooling already does it.

What does buying group coverage look like in practice?

A count of filled roles at an account, not a count of contacts. HubSpot's account-based marketing setup documents a Buying role contact property that identifies the role that a contact plays during the sales process, with options including Decision Maker, Budget Holder, and Blocker, and notes that multiple contacts can share the same role.

That property turns a vague instinct into something you can report. For each target account, how many of the roles you care about are occupied by a known contact who has engaged. Three of four is a real number with a clear next action. Eleven MQLs is not.

The Blocker option is the one worth dwelling on, because no MQL model has a slot for a person whose engagement is bad news. Knowing that a security lead at a target account has been reading your docs is useful precisely because it is not a positive signal. Our notes on defining an ICP cover how to decide which roles matter for your deal shape.

How do you pick target accounts without guessing?

With a tiering property you commit to and review. HubSpot documents an Ideal customer profile tier company property that shows how close a company matches your ideal customer profile in three tiers, with Tier 1 an excellent fit and Tier 3 acceptable but lower priority, and notes it can be customised.

It also documents a Target account property that identifies the companies you are marketing and selling to as part of your account-based strategy, as a single checkbox set to true for qualifying companies. A checkbox is deliberately crude, and that is the point: it forces a yes or no rather than a score nobody acts on.

The discipline is in maintenance, not in the field. A target account list that nobody prunes becomes a list of everyone, and a tier that nobody reviews becomes decoration. Put a date on it and revisit quarterly.

What role do intent signals play?

They tell you when to pay attention, not who is qualified. HubSpot describes intent signals as capturing high-value actions and news updates of companies you are tracking, helping you prioritise accounts, identify engagement opportunities, and personalise outreach.

The data sources are worth understanding before you trust them. HubSpot says signals include news about funding or when a company starts to research a topic you care about, uses reverse-IP technology to update company visit data including visitor counts and page views from monitored organisations, and derives CRM signals from conversations in HubSpot such as emails, transcripts, and notes.

Reverse IP is the part to hold loosely. It gives you a company-level hint from unidentified traffic, which is genuinely useful for prioritisation and a poor basis for claiming an account is in market. Treat it as a reason to look, not as a stage change. Our notes on AI lead qualification cover where automation helps with this and where it oversteps.

What should you actually report to the board?

Target accounts engaged, coverage within them, and pipeline created, with the fit and engagement split visible. Three numbers and one breakdown. No composite score, because a composite is where the bad news hides.

Keep the MQL count in the operational dashboard if your routing depends on it. It is a fine trigger for a workflow and a bad headline. The distinction between a number that fires an automation and a number that describes the business is one most marketing reporting never makes.

And report the disagreement. If sales accepted forty percent of what marketing qualified, that gap is the single most useful thing on the page, and averaging it into a conversion rate removes the only information worth discussing. Our notes on structuring a marketing board report go further into that.

What we would do if we were starting over

Define the buying roles before building any scoring. The roles are a statement about how your deals actually work, and every scoring decision downstream depends on them. Teams usually build the score first and then try to reverse engineer the roles from it, which never quite fits.

Then wire fit and engagement as separate scores from day one, even if you only report the combination. Once they are merged you cannot pull them apart retrospectively, and the C1 problem stays invisible.

Where we are honest about our own position: we build the websites, content engines, and automations that feed this system rather than running the sales side of it. So treat this as a framework from someone who sees a lot of these setups from the inside, not as a quota carrier's playbook. The parts we are confident about are the modelling and the reporting, because those are where the website and the CRM meet. Our notes on demand generation versus lead generation sit next to this one.

Will the MQL survive?

As plumbing, yes. As a reported outcome, it is already losing credibility with the people who read the reports, and the tooling has quietly moved on: account level lifecycle stages, separate fit and engagement scores, buying role properties, and company level intent signals are all standard now rather than enterprise features.

Our bet is that the teams who switch earliest will not be the ones with the best tooling, since most teams already own what they need. They will be the ones willing to report three awkward numbers instead of one comfortable one.

If you are rebuilding how your site and CRM hand off to each other, and you want a second opinion on what to measure, we are happy to think it through with you. Find us at phoenix.studio.