How do you build a win/loss analysis agent?

THE SHORT ANSWER

Run an agent every other Tuesday at 10 AM that reads every won and lost deal from the last two weeks. It pulls from Salesforce (deal size, close reason, sales notes), call transcripts, and win/loss interviews, then extracts seven categories: win patterns, loss patterns, feature insights, segment insights, competitive positioning, pricing insights, and three product recommendations. The point is to stop building features nobody asked for and start building the ones that move deals. One run might show: we win enterprise on API reliability and lose mid-market on price, and two fixes would convert 30% of mid-market losses.

You won a big deal last week. You lost one too. Your sales team knows why, but they are busy with the next deal, so the insight gets lost. Meanwhile product does not know: are you losing deals to a product gap, a pricing problem, bad UX, or just competitive pressure? So you keep building features nobody asked for while missing the ones that would actually close deals.

Read every deal, bi-weekly

The agent runs every other Tuesday at 10 AM and pulls three data sources. Win analysis reads recent closed-won deals from Salesforce, with deal size, segment, cycle length, and sales notes, and looks for patterns. "All our enterprise wins mention API reliability" tells you API reliability is a differentiator. Loss analysis reads recent closed-lost deals with the recorded reason, sales notes, and call notes. "We lost to competitor X because of feature Y we do not have." Interview data pulls your win/loss interview transcripts, the "why did you choose us" and "why not them" conversations that give you the real story.

To set it up, connect Salesforce for win/loss records, call recordings or transcripts from lost-prospect calls, your win/loss interview data, and rough competitive context on what competitors offer. It analyzes two weeks of outcomes each run.

Seven categories of insight

The output runs seven sections: win patterns (what is common across wins, which features you win on, which segments close easiest), loss patterns marked PRIORITY, feature insights (which gaps cost deals, which features you build that do not move deals), segment insights, competitive positioning, pricing insights, and three product recommendations.

Loss patterns is the section that matters most, and it is labeled PRIORITY for a reason. Understanding what is common across losses, which competitor keeps winning and why, and whether you lose to price or to product is the single highest-leverage input to roadmap decisions. The output of one run reads like this: "We win enterprise on API reliability and lose mid-market on price. Two product fixes would convert 30% of mid-market losses."

Why this beats the quarterly review

Quarterly win/loss reviews are retrospective and lean on sales memory. By the time the review happens, the context is stale and reps have moved on. Running it bi-weekly catches patterns while the data is fresh and gets them to product in time to act before sprint planning.

The result is a validated roadmap: you know which features would actually close deals, not just which customer shouted loudest. Sales gets data to sharpen positioning ("we win when we emphasize API reliability"). And you catch competitive threats early. Connect Salesforce and your call transcripts, run it, and read the loss patterns section first.

FAQ

What does the agent do? It reads every won and lost deal from the last two weeks and extracts seven categories of product insight, ending with three recommendations you can take into roadmap planning.

What data does it need? Salesforce win/loss records, call transcripts from lost deals, win/loss interview data, and rough competitive context.

SOURCES

THE LONG VERSION

RELATED ANSWERS

Last reviewed 2026-07-31 · 3 min read