Good business development questions: the kind that reveal a dying deal early

A pipeline review runs on questions, and most of them are the wrong ones. "Where are we on this?" produces a status narrative. "Is it going to close?" produces a number the person already knows you want to hear.

The questions below are chosen for one property: the answer is either a fact or a visible absence of one. Neither outcome requires anyone to be persuasive.

What to take away

Six questions about a live deal

What did the buyer do most recently, and when?

Not what you did. Their last act, with a date. A deal where the last buyer act is three weeks old is a deal with a problem, whatever stage it sits in. This one question replaces most of a status update.

Who else has to agree, and have you met them?

Two answers matter here and they are different. Not knowing who else signs off is early-stage and fixable. Knowing and never having met them is late-stage and dangerous, because your case is being argued in a room you have never been in by somebody who is not you.

What happens to them if they do nothing?

If the honest answer is "not much", you are competing with inertia and inertia usually wins. This is not a reason to abandon the deal. It is a reason to stop forecasting it at the same confidence as one where doing nothing has a cost.

What is the next scheduled event, with a date and a name?

An answer that is not schedulable is not a next step. Test it by asking whether it could go in a calendar without sending another message first.

What would have to go wrong for this to be lost?

People find this easier to answer honestly than its opposite, because naming a risk is not the same as predicting failure. The risks named are usually already known to the person and simply never asked for.

Has anything changed on their side since we started?

Sponsor moved, budget moved, priority moved, reorganization. Long deals accumulate these and nobody updates the original case. This is the question that catches a deal that died two months ago and has not been told.

Three questions that waste the meeting

"How confident are you?" A percentage that means nothing consistent between two people, and which the person will adjust to whatever keeps the conversation short. Ask what would have to be true instead.

"What do you need from me?" Well meant, but it makes the seller diagnose their own blockage and then ask for help, which most people find harder than the deal. Ask what the buyer last did, and the help you can give becomes obvious to both of you.

"Why is this slipping again?" It gets an explanation, and explanations are cheap. Ask instead what would have told you a month ago that this was going to slip, which produces something reusable.

Questions for a loss, asked of the buyer

Losses are the cheapest research available and nobody collects them, partly because the conversation is uncomfortable and partly because it is nobody's job the week after a deal ends.

Ask a short set, and ask them of the buyer rather than of your own team.

Rule one: the person who worked the deal should not run the conversation, because the buyer will be polite to them.

Rule two: record answers verbatim, not summarized into a category. A dropdown with six loss reasons turns everything you learned into one of six words, and that category is where information goes to die.

Reading the answers

An answer arrives with three properties worth noticing: how specific it is, how recent, and whether it describes the buyer or the seller. A specific, recent, buyer-side answer is evidence. Anything else is a starting point.

Watch for the reversed subject. "We are working on getting them the security review" describes your activity. "They asked for the security review on the ninth and their security lead has it" describes theirs. The same deal, two sentences, completely different information content.

Watch also for the answer that never varies. If every deal has the same blocker, the blocker is in your process rather than in the deals, and it belongs in a look at the operation itself rather than in a deal review. Where to look for that is set out in business development ops.

Turning the answers into something that lasts

A question asked once is a conversation. The same question asked of every deal for a quarter is data, and it does not need a system to become one.

Keep a plain record of two fields for each deal reviewed: the last buyer-caused event with its date, and the named risk. After three months, sort by both, and the pattern in the risks is usually more actionable than anything in the pipeline report.

Counting by hand for a fixed period, not building collection for it, is a general habit worth having. The mechanics are in operating processes metrics.

Be careful about what happens once these answers are attached to a person's assessment. A question used to run the work gets honest answers; the same question used to rank people gets prepared ones, and the effect is documented widely enough that it has a name in Goodhart's law.

The way to keep the two separate is the same argument made about people in management foundations, and the interview technique for getting a straight account of how work actually happens is in operating processes questions. Structured questioning of this kind is the same discipline that public sector guidance on performance management describes as ongoing monitoring rather than periodic assessment.

Common questions

How long should a deal review take per deal?

Four minutes for most, longer for the few that matter. The six questions take about that. A review that spends twenty minutes on each deal is a review that will be canceled within a quarter.

What if someone consistently cannot answer the buyer-act question?

Then the deals are not being worked, or the record is not being kept, and those need different responses. Find out which by asking them to tell you the last buyer act from memory. If they know it and it is not written down, the problem is the record.

Should losses be reviewed individually or in batches?

Individually for the buyer conversation, in batches for the reading. One loss tells you about one deal. Twelve read together tell you about your process, and the pattern is almost never what the individual explanations said.

Is it worth asking a buyer who chose us the same questions?

Yes, and it is easier to arrange. Wins have the opposite bias, though: people are generous about a decision they made and will credit you with reasons that had nothing to do with it. Read wins and losses together or not at all.