What does a win/loss program include?
A win/loss program includes a defined interview cadence run on a recurring schedule, a target ratio of loss to win interviews weighted toward losses, a consistent research window relative to deal close, a structured questionnaire, synthesized findings rather than individual deal recaps, and a distribution mechanism that gets findings to the functions that need to act on them.
Each of these elements exists to solve a specific failure mode that shows up when a program is assembled ad hoc rather than scoped in advance.
Interview Cadence and Ratio
A program specifies how often interviews run, commonly quarterly or semi-annually, rather than deciding case by case whenever someone remembers to schedule one. It also specifies a target interview count and ratio for each cycle. A common structure is 20 to 30 interviews per cycle, weighted roughly two to one toward losses. Wins carry noise: timing, a weak competitor, a rep relationship that happened to click. Losses tend to produce a cleaner signal, since something specific usually went wrong, which is why loss-heavy programs surface more actionable patterns than programs that split interviews evenly.
Research Window
Programs define a timing window relative to when the deal closed, typically 30 to 180 days. Interview a buyer too soon and they’re still transitioning into whatever they chose. Interview them too late and memory has already compressed into a simplified, rationalized version of what happened. The window balances proximity to the decision against the reflection time a buyer needs to articulate what actually drove it.
Questionnaire and Deal Selection
A program uses a consistent questionnaire that gets refined across cycles rather than reinvented each time, covering not just product and pricing but process, internal politics, and perception, the areas where the most actionable findings tend to live. Deal selection matters just as much as the questions asked. Programs that let sales curate which deals get reviewed introduce a selection bias, since the deals flagged as not worth reviewing are frequently the ones where the real reason for the loss would be most revealing. The stronger pattern is executive buy-in over individual deal approval, with the deal list pulled directly from the CRM rather than filtered by the team whose performance the research might reflect on.
Synthesis and Distribution
A program’s output is a synthesized report, not a stack of individual interview summaries. The distinction matters because a report that connects patterns across a full interview set produces findings a team can act on immediately, while a folder of well-written but disconnected write-ups leaves the pattern recognition to whoever receives it. Distribution then completes the loop: findings reach cross-functional stakeholders in a readout format, translated into a specific action for each function, rather than emailed out as a document each team interprets independently.
A program missing any one of these elements tends to drift toward the activity-metrics trap, where running interviews on schedule gets mistaken for the program actually working.