MQL to SQL Conversion Rate: Why It's Low and How to Fix It

Short answer: your MQL to SQL conversion rate is the share of marketing qualified leads that sales accepts as worth pursuing. Published B2B benchmarks put it at roughly 13 to 21%. When yours is well below that, the leads usually aren't the main problem. It's more often the handoff: an MQL definition sales never agreed to, a lead score that doesn't predict anything, slow follow-up, MQLs that never reach an owner, or lifecycle data you can't trust. You can check each of those in HubSpot.

How to work out your MQL to SQL rate

The common way is to take this month's SQLs and divide by this month's MQLs. That compares two different groups of people. A lead that became an MQL on the 28th probably won't be an SQL until next month, so the rate moves around with timing and looks worse after a busy month of lead gen.

Use a cohort instead:

1. Take contacts that became MQLs in a set window, using the Became a Marketing Qualified Lead date property.
2. Count how many of them have since reached SQL or a later stage. A contact now at Opportunity or Customer still counts.
3. Divide the second number by the first.

Leave a gap between the end of the window and today, so leads have had time to convert. The free audit uses contacts that became MQLs between 30 and 180 days ago. If it usually takes sales longer than 30 days to qualify a lead, push the window back further.

Before you trust the number, check how many recent SQLs have no MQL date at all. If a demo form sets contacts straight to SQL, your hottest leads skip MQL and drop out of the rate completely. Why are my HubSpot lifecycle stages wrong? covers that one and six other causes.

What's a good MQL to SQL conversion rate?

Gradient Works' 2025 B2B sales performance benchmarks, updated in March 2026, give 13 to 21%. They don't attribute that range to a named study, so treat it as a reference point rather than a target. Published benchmarks vary a lot, and some quoted online are much higher.

Your own trend is more useful. If your rate has dropped over the last two quarters while lead volume held up, something in the handoff changed.

Why is my MQL to SQL conversion rate so low?

Five causes, in roughly the order I'd check them.

1. Sales never agreed the MQL definition

If marketing wrote the definition on its own, sales has no reason to trust the leads it sends them. Leads get a quick look, a judgement on a hunch, and nothing after that.

Get your sales lead in a room for an hour. Agree what has to be true before a lead goes to sales, write it down where both teams can see it, then build the MQL workflow from that. Make a rejection reason required when sales turns a lead down. After a month you'll know whether the problem is lead quality or follow-up, instead of arguing about it.

2. The lead score doesn't predict anything

If you have more than one score (fit, engagement, combined, or one per team or region), test the one your MQL workflow uses. Take contacts created in the last year that have that score. Split them into a top half and a bottom half by score, then compare how many in each reached SQL. I use 1.5x as a rough bar. If the top half isn't converting at least that much better, the score isn't telling sales anything.

Also check what your MQL workflow runs off. HubSpot stopped updating the legacy HubSpot Score property at the end of August 2025, and new contacts don't get one. A workflow waiting for that score to pass a threshold won't fire for anyone created since. Move it to a score from HubSpot's lead scoring tool. Each score there writes to its own score property, so check the workflow reads the right one.

3. Follow-up is slow

A demo request answered two days later has usually gone cold or booked a call with someone else. In the report it then looks like a bad lead.

HubSpot already tracks this. Two default contact properties, Date of first engagement and Lead response time, show when sales first touched a lead and how long that took. Report on contacts created in the last 90 days that have an owner, and look at Lead response time. If you're being told the leads are bad, check this number first.

4. MQLs never reach anyone

Filter contacts where lifecycle stage is MQL and contact owner is unknown. That's the list of leads marketing qualified that nobody in sales was told about. Then look at MQLs that do have an owner but have had no lead status change and no logged activity for a couple of weeks. Those reached someone and still went nowhere.

The usual causes are a routing workflow that only enrols some forms, a round robin still assigning to people who've left, or a second tool overwriting the owner. The lead routing post covers the fixes.

5. The stage data is wrong

If forms, imports, the Salesforce sync and several workflows all set lifecycle stage, nobody controls the numbers the rate is built on. Contacts skip MQL, or sit at stages their activity doesn't support. Fix the stages before you rely on the rate, because every other check here depends on them.

Better conversion or more leads?

Say you get 200 MQLs a quarter and 8% become SQLs. That's 16. Get conversion to 15% and the same 200 MQLs give you 30. To get 30 SQLs at 8%, you'd need 375 MQLs, which is 175 more leads to buy for the same result.

If a big share of your current MQLs never got a first touch, more leads mostly means more leads that never get a first touch.

What it looked like at one company

At a 65-person B2B tech company I worked at, MQL to SQL was under 4%. It was the usual argument: sales didn't rate the leads, and marketing didn't think they were being followed up. We rewrote the MQL definition with the sales team and rebuilt the lead scoring, and a few months later it was about 33%. Follow-up went from days to under an hour, and the number of leads it took to create one opportunity went from roughly 180 to about 48 within three to six months.

What to do next

Work out your rate as a cohort first. If it's well under the published range, or you can't get a number you'd defend in front of sales, run the free Lead Leak Audit. It does the cohort maths for you and checks causes 2 to 5 above in your portal.

If you'd rather talk it through, the Free Lead Leakage Check is 20 minutes on one lead process, with a one-page summary within 1 working day. The full fix (MQL definition, lead score, routing and a handoff dashboard) is the Lead Handoff Sprint.