SALES OPERATIONS // FREE LIVE DEMO
Win-Loss Analyst
Deals close or die and nobody formally captures why — the same avoidable loss reason repeats quarter after quarter.
WHAT IS WIN-LOSS ANALYST?
An AI win-loss analyst takes a short summary of what happened on a closed deal and classifies the likely win or loss reason against a standard taxonomy — price, product fit, timing, competitor, champion loss, or no decision — with an honest confidence rating, then drafts a short interview script to confirm the real reason directly with the buyer instead of guessing from incomplete CRM notes.
HOW IT WORKS
Most 'loss reason' fields in a CRM are filled in by a rep's best guess, not a confirmed answer from the buyer — which means the same avoidable pattern (say, a recurring pricing objection) never gets caught because it's mislabeled as 'timing' or 'no decision' deal after deal. This agent is prompted to classify against a fixed, standard taxonomy so results are comparable across deals, and to state a confidence level honestly — a CRM note that says 'went quiet' supports a much weaker classification than one with an explicit stated reason, and the agent flags that difference instead of presenting a guess as certainty.
The interview script exists because classification from secondhand notes is never as reliable as asking the buyer directly. The four questions are structured to separate the stated reason from the real one, since buyers often cite price as an easy excuse when the actual issue was a stalled champion or a missed requirement.
You are a revenue operations analyst who runs win-loss analysis for a B2B sales team. Given a deal summary and outcome, classify the likely reason against a standard taxonomy (price, product fit, timing, competitor, champion loss, no decision), state a confidence level for that classification since CRM notes are often incomplete, and draft a short 4-question interview script to confirm the real reason directly with the buyer. Output ONLY a JSON array of typed blocks matching the OutputBlock union — no prose outside the JSON.USE CASES
Quarterly loss-pattern review
Run every lost deal from the quarter through the agent to see which taxonomy category dominates before assuming it's always 'price.'
Post-loss buyer outreach
Use the generated interview script to reach out to a lost prospect for real feedback instead of letting the deal go cold with an unconfirmed CRM note.
Win analysis, not just loss analysis
Run closed-won deals through the same process to confirm what actually tipped the decision, so the winning pattern can be repeated deliberately.
Sales leadership reporting
Aggregate confidence-rated classifications into a QBR-ready summary that distinguishes confirmed loss reasons from rep guesses.
RESULTS & BENCHMARKS
Time to draft a win-loss interview
projected (modeled)
Loss reasons confirmed vs. assumed from CRM notes
projected (modeled)
LIVE DEMO
Try it right now
FAQ
Common questions
How accurate is the loss reason classification?
It's only as accurate as the CRM notes it's given, which is why every classification comes with an honest confidence rating rather than being presented as fact.
Why does it draft interview questions instead of just giving an answer?
Secondhand CRM notes are unreliable — the agent's real job is prompting you to confirm the reason directly with the buyer, since that's the only source that settles it.
What taxonomy does it use?
A standard six-category framework — price, product fit, timing, competitor, champion loss, and no decision — so results are comparable across every deal in a pipeline.
RELATED AGENTS
Win-Rate Improvement Planner
Win rate is declining and leadership wants a plan, but 'sell better' isn't a plan — it's a wish.
QBR Data Aggregator
Building a quarterly business review means pulling numbers from five different tools and stitching them into a deck by hand, every single quarter.
Competitor Factual Debunker
Prospects repeat a competitor's exaggerated claim in a call and the rep either agrees, panics, or argues — none of which closes the deal.
Built by Hammad Yousuf — AI Marketing Automation Engineer, 540K+ YouTube subscribers.
See the production missions these agent patterns run in, or hire the whole system.