MQL vs SQL: The Definitions, and Why Most Companies Get Them Wrong

An MQL has shown interest. An SQL has demonstrated a real requirement and the ability to act on it.

That is the whole distinction. Everything else is implementation detail — but the implementation detail is where most B2B companies quietly lose money.

The definitions

MQL — Marketing Qualified Lead. Someone whose behaviour suggests interest: downloaded a guide, attended a webinar, visited pricing repeatedly, filled in a form. Marketing decides they are worth a closer look.

SQL — Sales Qualified Lead. Someone sales has examined and accepted as a genuine opportunity: there is a real need, budget is plausible, timing is workable, and you can reach whoever decides.

The transition between the two is the handoff, and it is where the arguments start.

Side by side

Where it breaks

The threshold is arbitrary. Most scoring models were invented in a meeting. Someone decided a whitepaper download is 10 points and a demo request is 50, and the number 100 became the definition of “qualified.” If nobody ever validated those weights against actual closed revenue, the model is measuring activity, not intent.

Marketing is paid for volume. If the target is MQL count, MQL count is what you get. The threshold drifts downward because that is the easiest way to hit the number, and lead quality degrades quietly.

Sales does not trust the leads. So they ignore them, work their own network, and the entire measurement apparatus becomes theatre. This is extremely common and rarely admitted.

The definitions were never agreed. Marketing counts an MQL as a win. Sales counts it as noise. Both report against their own definition, and the numbers cannot be reconciled.

The fix everyone knows and few do

Define both terms jointly, in writing, with sales holding veto power. Then work backwards from closed revenue rather than forwards from a scoring model.

Take your last twenty closed customers. What did they have in common before they became opportunities? Company type, size, trigger event, role of first contact, timing. That is your qualification model, and it is derived from evidence rather than invented in a workshop.

Then agree a rejection loop: when sales rejects an MQL, they record why, and marketing sees it. Without that loop, nothing improves.

The better question

Both terms describe a person’s relationship to your marketing. Neither says much about whether the company genuinely needs what you sell.

A more useful question: what evidence do we have that this company has an actual requirement?

That is a different kind of qualification. Not “did they download something” but “is there observable evidence of a commercial need” — a company already buying what you sell, at volume, from someone else. Expansion, hiring, procurement activity, regulatory triggers, supply disruption.

For businesses selling across borders, this is often directly observable in trade records. For domestic and services businesses the signals differ, but the principle holds: evidence of buying behaviour beats inferred interest, every time.

An MQL tells you someone visited your website. Evidence tells you a company has a problem worth money. The second is a far better basis for spending a salesperson’s week.

Practical definitions to start from

If you need something workable today:

MQL: Matches your ideal customer profile and has taken an action indicating a live problem — not a passive content download.

SQL: A human conversation has confirmed a real requirement, a rough sense of budget, a workable timeline, and access to whoever decides.

Then measure the only ratio that matters: MQL to closed revenue. Not MQL to SQL. If MQLs convert to revenue at a rate you would not accept from any other channel, the definition is wrong regardless of how many you generate.

Related: Outbound vs inbound · How we qualify before handover