Customer discovery interviews, and how many are enough
B2B customer discovery interviews: the sample size the research converges on, why your segments change that number, and the questions that get honest answers.
Read
Disqualifying is not losing. It is the decision that a deal should never have entered the pipeline, and the documented prerequisite for scoring leads at all is having made it at least forty times.
Every sales team has a deal that has been in the pipeline for seven months. Nobody is working it, nobody has heard from the buyer since August, and nobody has killed it either, because killing it means the number goes down and the number going down is a conversation. So it sits there at forty percent, and the forecast is wrong by exactly that much.
Disqualifying is the decision to stop. It is not the same as losing, and treating the two as one thing is why most teams avoid it. A lost deal is one you competed for and did not get. A disqualified prospect is one you worked out should never have been in the pipeline, which is a cheaper mistake and a more useful record.
Disqualification goes wrong in both directions. Reps hold dead deals because the pipeline is a performance metric, and they also drop live deals after two unanswered emails because silence feels like rejection. Neither is a decision. These are the conditions that genuinely settle it.
What is not on that list: a single unreturned call, a price objection, a competitor already in place, or a quarter ending. Those are conditions to work, not reasons to stop. Which signals belong on your own version of this list comes out of the entry criteria, and that is the subject of the ideal customer profile for outbound and its four hard tests.
This sounds administrative and it is not, because the mechanics decide whether the decision is reversible. Worth reading the documentation rather than guessing at it.
Microsoft's Dynamics 365 Sales documentation on lead qualification sets out the default disqualify reasons as Lost, Can't Contact, No Longer Interested and Canceled. The record is not deleted. It moves to an inactive state and stays available, so a lead disqualified today can be reactivated and qualified later.
Read that as a permission rather than a footnote. The reason reps resist disqualifying is a belief that it throws the relationship away, and in a system configured like this it does not. It files it. The practical consequence is that a disqualification with a reason attached is recoverable, and a deal quietly abandoned in a late stage is not, because nobody will ever look at it again.
The four default reasons also do more work than they look like they do. "Can't Contact" and "No Longer Interested" are different diseases with different cures: the first is a prospecting and routing problem, the second is a qualification problem. A team that records both as "Lost" has given up the ability to tell them apart.
This is the part that changes how we think about it. If you ever intend to score or prioritise leads with anything other than instinct, the disqualified records are half the training data, and the requirement is documented rather than theoretical.
The same Microsoft documentation sets the prerequisite for predictive lead scoring plainly: before you begin, ensure your organisation has at least 40 qualified leads and 40 disqualified leads from the past two years. The model works by analysing the patterns from leads that were qualified versus those that were disqualified. Without the second set there is no contrast to learn from, and the feature simply does not turn on.
So a team that never disqualifies is not merely carrying a bloated pipeline. It has made its own data unusable for the exact purpose that would have told it which deals to work. The documented scoring bands are worth knowing too, because the lowest one is explicitly a disqualification prompt rather than a nudge.
| Grade | Score range | Documented action |
|---|---|---|
| A | 75 to 100 | High priority, contact immediately |
| B | 50 to 74 | Strong potential, schedule follow up soon |
| C | 25 to 49 | Monitor, review periodically or automate nurturing |
| D | 1 to 24 | Consider disqualifying or routing to automated nurturing |
One caveat the documentation states and we would repeat louder: the score is informational, not prescriptive. Sellers retain full judgement. A grade D on a named account you have been building toward for a year is a prompt to think, not an instruction to delete. The model also needs the data before it is any good, and the honest note in the same documentation is that organisations with limited lead history should focus on capturing quality data as they build toward the minimum thresholds.
There is a specific, documented trap for teams that bulk-process leads instead of deciding on them one at a time. Microsoft's troubleshooting guide for lead qualification notes that when you qualify a lead through the Leads grid, the system creates an account or contact even if a duplicate record exists, because the duplicate detection rule is disabled by design in that flow. Qualify from the lead record form instead and the rule runs and warns you.
That is how a database ends up with the same company three times under slightly different names, and it happens fastest on teams that clear a backlog of stale leads by mass-qualifying rather than facing the disqualify button. The same guide notes the other end of it: you cannot convert or qualify a lead that is already closed, so the record state is not advisory.
The broader discipline this belongs to is unglamorous and covered separately in CRM hygiene for small sales teams. Disqualification is the part of it that needs management cover rather than a rule.
Disqualification is the same judgement as qualification, made later and with more information, so the two should share a definition. If your qualification criteria are written down in a form that survives scrutiny, the disqualify decision is mostly reading them back: a qualification framework you can defend to a regulator covers what those fields should hold and, importantly, what to leave out.
If you are choosing between named frameworks to hang this on, the comparison is in BANT, MEDDIC and CHAMP without pretending one of them wins. None of them has a disqualify step worth the name, which is part of the point.
And the deals that should have been disqualified and were not are the ones that show up as stalls. The five places they die are in why B2B deals stall, and at least two of those places are really late disqualifications wearing a stage name.
One operational point. For any of this to be visible, the pipeline view the team looks at has to show deal age and last buyer action, not just stage and value. What that actually requires, and what "real time" is worth when the underlying decision is weekly, is covered from the build side in rivl's note on a real-time dashboard for a sales team.
We are not going to quote a figure for how much disqualification improves forecast accuracy or win rate. The numbers in circulation for this come from vendor material without a published method or a control group, and a number like that becomes a target, which is how you get a team disqualifying to hit a hygiene metric. The mechanism is what holds up: a forecast built only on deals someone can describe a next step for is wrong less often than one built on everything that was ever entered.
The documented behaviour above is Dynamics. Salesforce, HubSpot and Pipedrive each handle this differently, and the configurable parts are configurable, so check your own instance before relying on the specifics. What transfers is the shape: the record survives, the reason is a field worth using properly, and the disqualified set has downstream value.
Last thing, and it decides whether any of this holds. Disqualification only happens if a shrinking pipeline is safe to report. If the culture punishes it, reps will keep dead deals alive at forty percent and you will have a tidy process producing fiction. That is a management problem and no framework survives it.
Send three things: how many open deals have had no buyer-initiated contact in thirty days, how many have moved their close date more than twice, and how many disqualifications you recorded last quarter. We will come back with which ones we think should be closed, which are worth one more attempt, and what the forecast looks like after.
Book a Free Consultation