One Buyer’s Story Is an Anecdote. Twenty Is a Pattern
Every engagement I run starts with the same question: how many interviews is enough?
My answer is thirty. Ten wins, twenty losses. That ratio is intentional, and the volume matters more than most programs account for when they’re scoping a project down to fit a budget or a deadline.
Across dozens of programs, the pattern in how patterns emerge is remarkably consistent. Early themes start to surface from the first five win conversations and the first ten loss conversations. I start to hear the same competitive concern come up twice, then a third time. A demo objection that seemed specific to one deal turns out to be the same objection from a different buyer, described in different words. It starts to feel like the shape of something is forming.
But the back half is where the pattern locks.
Something mentioned casually in interview four takes on real weight when I hear it again, unprompted, in interview seventeen. A concern that looked like an outlier the first time becomes a signal the third time. A piece of feedback I almost filed as a one-off individual complaint turns out to be the thread that ties a dozen separate conversations together, and that reframing only happens because there were enough independent conversations to tie together in the first place.
This is where a lot of programs quietly shortchange themselves, and it’s worth checking whether yours is one of them. Running win/loss one deal at a time, producing a summary after each individual interview, misses the compounding effect entirely. It’s reading pages instead of the book. Running interviews serially over many months, a few here, a few there, creates the opposite problem: by the time there are enough conversations to compare, the earliest ones are six months stale, and the buyer’s memory, along with the market conditions, has moved.
Twenty to thirty interviews, gathered within a defined window, is what makes a finding durable enough to present without hedging. It’s the difference between telling a board “one buyer mentioned pricing felt confusing” and telling them “pricing structure came up, unprompted, in eleven of twenty loss interviews, and it never appeared as a concern in a single win.” The first is a data point someone in the room can dismiss. The second is a finding someone has to act on.
Volume also changes what a team can trust about the shape of the finding, not just its existence. A concern that surfaces in loss interviews but never once in the win interviews is a real, specific gap. The same concern showing up in both sets tells a different story, something buyers notice but ultimately tolerate, which calls for a different kind of response than a dealbreaker does. That comparison is only possible with enough interviews on both sides to have something to compare.
I’ve watched programs try to shortcut this with fewer, deeper interviews, on the theory that quality can substitute for quantity. It can’t, not for this purpose. A brilliant, detailed conversation with three buyers still leaves you with three data points, and three data points can’t tell you whether what you heard is representative or just what those three people happened to experience. A ninety-minute interview with one loud, articulate buyer is still an anecdote, no matter how well it’s documented. Depth answers what happened in one deal. Volume answers what’s true across your business.
None of this is an argument for running win/loss forever without a stopping point. Thirty is a target, not a floor that keeps climbing. Once the fifteenth or sixteenth loss interview in a row confirms what the tenth already told you, adding a thirty-first interview rarely changes the finding, it just adds cost. The discipline is in reaching the point where the pattern has actually locked, and stopping there, not in running interviews indefinitely for the comfort of more data.
Hitting thirty interviews also means planning for a much larger outreach effort than most programs budget for, and if your own program is scoped around a smaller number, this is why the participation math matters. Loss interview participation typically runs around five percent, which means getting to twenty completed loss conversations usually requires reaching out to somewhere near four hundred former buyers. Win interviews convert at roughly double that rate, so ten completed win conversations still means close to a hundred invitations. Programs that treat interview volume as the only variable, without accounting for the outreach it takes to get there, end up scoping a target they never actually reach, and then drawing conclusions from whatever smaller number showed up instead.
The real question about your own program is whether you’re running enough interviews to trust the pattern it’s telling you about, not just enough to produce a report.