All posts
Lead generation

The MQL Is the Wrong Unit of Progress

Gartner found 67% of B2B buyers now prefer a rep-free purchase. Forrester calls MQL-driven go-to-market structurally obsolete. Most small firms still count a form fill as the win.

HA
Hamza AliFounder, Fixora
4 min read
The MQL is the wrong unit of progress, Fixora Journal banner
Fixora · Lead generation

A unit invented for a market that no longer exists

The marketing qualified lead made sense in a world where a buyer had to talk to you to learn anything. Marketing found the person, sales educated them, the handoff was the moment of progress, and counting it was reasonable.

That world is gone. Gartner's March 2026 survey of 646 B2B buyers found 67% prefer a rep-free purchasing experience, and 45% had used AI during a recent purchase. Forrester has gone further, describing MQL-driven go-to-market as structurally obsolete.

Buyers now do most of the evaluation before they identify themselves. By the time somebody fills in your form, the shortlist usually exists.

What the MQL gets wrong, specifically

It counts an individual when the decision is collective. A typical B2B purchase involves a group, and you will never meet most of it. The person who downloaded your guide may be researching on behalf of a committee that has already formed opinions, which is the buying committee problem in miniature.

It counts an action, not an intent. A guide download and a pricing page visit followed by a returning session are wildly different signals. Scoring both as leads flattens the only distinction that matters.

It rewards volume over fit. Anything measured becomes a target. Teams measured on MQLs produce MQLs: gated fluff, lead-gen ads aimed at people who will never buy, and a sales team that stops returning calls because the last thirty were worthless.

It arrives too late to be interesting. If most evaluation happens before contact, the form fill is closer to the end of the buying process than the start. Optimising for it means optimising for the last ten percent.

What to count instead

You do not need a revenue-attribution platform for this. Four things, all available to a five person company.

  1. Accounts showing activity, not individuals. Group your enquiries and your site behaviour by company domain. Three people from the same firm reading your pricing page in a fortnight is the actual signal, and it is invisible if you count contacts.
  2. Qualified conversations, not qualified leads. The countable unit that correlates with revenue is a real conversation with someone who has a budget, a problem, and a timeline. Everything before it is input.
  3. Pipeline created and pipeline converted. Two numbers, monthly. This is the honest version of marketing performance and it survives contact with a finance director, unlike an MQL count.
  4. Self-reported source. One optional free-text field on the form. It is the only instrument that sees the untracked half of your demand, for the reasons set out in the attribution gap.

The handoff matters more than the label

Whatever you call the unit, the transition is where deals die quietly. Two rules cover most of the damage.

Speed. Response time dominates almost every other factor in whether a first conversation happens at all, which is why the five minute rule is the highest-return operational fix in lead generation. A perfect scoring model attached to a next-day response is worth less than a crude one attached to a five minute response.

Context. Send what the person actually read, not just their job title. "They read the pricing page twice and the migration article" changes the first sentence of the call.

What this means if you are small

The good news is that the enterprise version of this problem needs software and yours does not. You have few enough accounts to read the list yourself.

Once a week, look at who enquired, which companies appeared more than once, what they read before they wrote, and which conversations turned into scope. That is buying-group analysis. It takes twenty minutes and it beats a scoring model built on invented point values.

And keep publishing for the people who are not in market yet, because most of them are not, and they will arrive already convinced or not at all.

Frequently asked

So we should delete lead scoring? If your scoring model has never been validated against closed revenue, it is decoration. Either validate it or drop it.

Is gated content dead? Gating works when the asset is genuinely worth an email address. Gating a blog post to manufacture leads mostly buys you bad data and unsubscribes.

Can Fixora rebuild our funnel reporting? Yes: account-level grouping, qualified conversations, pipeline created, and one source question on the form. Tell us your stack and get the plan within 48 hours.

Have a project that needs this?

Tell us what you are building. We reply within 24 hours.

Start a project

The service behind this

AI Marketing Services

A full marketing function without the headcount: strategy, content, campaigns, and outbound, run as one system and reported on the numbers that decide whether it paid.

What this includes

Keep reading