Lead qualification: four gates, run in the cheapest order
BANT and MEDDIC qualify leads you are already talking to. This is the version for leads who have never heard of you, run cheapest gate first.
Most lead qualification frameworks assume you're already talking to someone. BANT, MEDDIC and the rest are questions you ask a person. With outbound, nobody's talking to you yet, so qualifying a lead has to rely on what you can see from outside the company, and the order you check things in matters more than the checklist itself.
For outbound, lead qualification works best as four checks, run cheapest first. Does the company match your profile? Are they already solving the problem? Is there a dated reason to get in touch now? Can you write one specific sentence about them? The last check needs someone reading carefully, so you want it applied to twenty companies, not four hundred.
The four checks, in order
Each check costs more per company than the one before it. A database can run the first across ten thousand rows almost for free, while the last one is a person reading for ten minutes. Getting the order right is the difference between an afternoon's work and a fortnight's.
1. Profile. Does this company match who actually buys from you, rather than who could? That's an ideal customer profile question, and if your profile is a description rather than a filter, this check won't remove anything.
2. Problem. Are they already solving this another way? A clinic that already has online booking would be switching, not starting, which is a different pitch and usually a harder sale. Checking costs one page load.
3. Timing. Is there a dated event that makes this the right month? It's worth being strict about which events count as evidence.
4. The sentence. Can you write one line about them that names something specific and checkable, with a link? This is the one check you can't hand off entirely, and it's the one that decides whether the message is worth sending at all.
Why people run them backwards
The fourth check is the interesting one. Reading about a company is more fun than filtering a spreadsheet, so people start there, get through eleven companies by lunch, and decide outbound doesn't scale. It doesn't, if you read four hundred companies. It's fine if you read nineteen.
The other reason is that the first three checks feel like throwing away opportunities. What they really throw away is work: a company removed at the second check was never going to reply to an email about a problem it had already solved.
Which qualification checks can you hand to a tool?
The first three, with spot-checks. A tool can draft the fourth, but the decision stays with you.
| Check | Can a tool do it? | What goes wrong when it does |
|---|---|---|
| Profile | Yes | List data is often wrong, so spot-check a sample by hand |
| Problem | Mostly | It reads the homepage and misses the feature two clicks deep |
| Timing | Yes, with sources | It reports months-old events as if they were current |
| The sentence | It can draft; you decide | Asked to find something, it always will |
That last row is the one to watch when you're buying software. A system judged on how many qualified leads it produces will produce plenty. Restraint has to be designed in deliberately, which is why a proper evaluation checks what a tool does when there's nothing to find.
Where each check should get its evidence
For the profile check, read the company's own website rather than the list vendor's fields. Headcount bands, industry codes and practitioner counts are wrong often enough that a check built on them will let through companies that don't exist as described.
For the problem check, look for the feature itself, not the marketing around it. A homepage that mentions online booking and a booking page that actually works are two different findings.
For the timing check, insist on a date. We see how much this matters in our own data: of the 843 conversations Revtive's agents flagged as worth replying to between July and September 2026, only 39% came with a date at all, and among those, about a quarter were more than eight months old by the time they were found. An event without a date isn't a timing signal, it's a fact of unknown age.
The sentence check needs a person, or a tool that shows its work the way a person would. Revtive's prospecting agent runs the first three and drafts the fourth with a source link attached, and that's the split we'd recommend generally: automate the checks that fail quietly, and keep a human on the one that decides whether to send.
What the funnel looks like
The illustration in our explainer on prospecting follows a made-up list of 400 physiotherapy clinics through these four checks: 212 match the profile, 61 don't already have online booking, 19 have a dated reason to get in touch, and 6 produce a sentence worth sending. The exact numbers are invented, but the shape is typical. Most of the list goes early and cheaply, and only a small group ever gets read properly, which is why the whole thing takes a morning rather than a week.
When this approach is wrong
For inbound leads, most of it doesn't apply. Someone who filled in a form has already passed the first three checks by acting, which is also why a hand-raise shouldn't sit in the MQL queue. Use BANT or something like it and just ask them.
If you only have about fifty target accounts in total, you don't need checks at all. You have a list, so read the whole thing. Frameworks are for triage, and at that size there's nothing to triage.
And if every company passes, something's off. When four hundred rows turn into two hundred "qualified" accounts, the checks aren't filtering anything, and the usual culprit is a profile that's too loose. Narrowing it until it feels uncomfortable is the fix, even though it's the last thing anyone wants to do in a quarter with a pipeline target.