WIN-LOSS ANALYSIS: HOW TO RUN ONE, PLUS A TEMPLATE

NOV 5, 2026

Win-loss analysis is the practice of studying closed deals, both won and lost, to find out why buyers chose you or someone else. Teams combine buyer interviews, CRM data, and call evidence to spot patterns in product, pricing, messaging, and sales execution, then act on them.

74% of B2B buyer teams show unhealthy conflict during the decision process, according to Gartner's survey of 632 B2B buyers. A closed-lost dropdown that offers "pricing" and a rival's name has no good place to put "nobody could agree."

A closed-lost reason is a star rating: quick and countable, with little detail. Win-loss analysis is the written review, where the buyer explains what they saw and why they chose as they did.

This guide puts the whole kit in one place: the six-step process, a copy-ready template, a reason-code list for your CRM, interview and survey questions, the win rate math with a worked example, dashboard setup, a monthly review agenda and a 90-day launch plan. The template comes right after the six steps.

What win-loss analysis tells you that a closed-lost field can't

Gartner's 74% counts conflicting goals, disagreement over the next step, and being overruled by outside decision-makers.

Your rep heard from the champion. The champion gave one version of a debate your rep never attended.

The cast is big, too. Forrester's State of Business Buying puts the typical decision at 13 internal stakeholders and nine external influencers. Each of them has a reason. Your CRM has one field.

Conflict among that many people can stall a deal with no competitor in sight. A B2B win-loss analysis that studies just the deals lost to named competitors never meets those buyers.

A good program answers six questions:

• Why did the buyer choose us, or someone else? In the buyer's words, not the rep's summary.
• Who was in the running, and which criteria decided it? Competitors, the incumbent, the spreadsheet, or doing nothing.
• Where did the deal actually turn? Sometimes earlier than the rep thinks it did.
• Which messages and proof landed with which stakeholders? Finance and IT may need different proof.
• Which seller behaviors show up in wins and go missing in losses? Discovery depth, follow-up speed, how many people the rep got in the room.
• What would move next quarter's numbers? A product fix, a pricing change, a sharper message, a coaching plan.

Each question has a natural owner (product, product marketing, sales leadership, enablement, RevOps), so the output has to speak to several audiences at once.

Where win-loss evidence comes from: four sources

Forrester's analysts describe three ways to capture buyer feedback for a win-loss program: post-deal interviews, automated surveys, and sales call transcript analysis. Add a fourth, the fields and behavior already sitting in your CRM, and you have the full evidence stack.

• Buyer interviews. The why, in the buyer's own words, plus how they saw each vendor. Works best when deal sizes justify 15 to 30 minutes of buyer time. Pairs well with CRM reason codes for sorting and counting.
• Post-deal surveys. Ratings on product, pricing and sales experience, plus one open answer. Works best when pipelines with enough volume that interviews can't cover every deal. Pairs well with interviews on a sample of the same deals.
• Call and meeting evidence. What was said, which objections came up, who joined and who went quiet. Works best when you want signal while the deal is still live. Pairs well with interviews, for the why behind the moments.
• CRM fields and deal behavior. Stage timing, loss codes, competitors, content engagement, activity patterns. Works best when you want a trend line that costs little to maintain. Pairs well with every other source.

A few notes on using each:

Interviews. Forrester suggests conversations of about 15 to 30 minutes with buyers from both won and lost deals, open-ended questions, an incentive for their time, and a third party where objectivity matters. The logic is simple: a buyer has less reason to be polite to someone who isn't the seller.

Surveys. Keep them under five minutes: a handful of ratings and one open question, sent within days of the close while the decision is fresh. A ready-to-use set is in the questions section below.

Call and meeting evidence. Recordings and transcripts show what happened rather than what anyone remembers. Look for objections, competitor names, and which stakeholders appeared in which meetings.

CRM fields and deal behavior. The cheapest source, and the one that makes the others countable. A required closed-lost field with a fixed reason list turns every interview into a data point.

How to do a win-loss analysis in six steps

To do a win-loss analysis, pick one segment and one question, sample both wins and losses, gather buyer and call evidence, code each deal against a fixed reason list, share the patterns with the teams who can act, and track the changes you ship. Here is the win-loss analysis process in working detail.

Step 1: Decide what the analysis is for

Pick a single question for the quarter. "Why do we lose mid-market deals after the demo?" is a question. "Learn from our deals" is a mood.

Name the three or four people who will receive the findings and what each of them can change. A finding with no recipient turns into a slide that gets admired and ignored.

Step 2: Choose the deals

Sample wins, losses and no-decisions from the same segment and time window, ideally deals closed in the last quarter, so memories are fresh and the product and pricing haven't shifted underneath you. Mix reps, deal sizes and lead sources.

Wins belong in the sample. They show you which behaviors to protect, and they keep the analysis from turning into a quarterly autopsy.

Then record the stage where each deal turned, using the stages in your B2B sales process.

"Lost" is an outcome. "Lost after the security review" is a finding.

Step 3: Collect evidence in layers

Start with what you already own: CRM fields, activity history, call recordings, the content that was shared. That layer is free and fast, and it tells you which deals deserve a conversation.

Then add buyer voice. Invite buyers from a neutral name where you can, keep the ask short, and offer something in return. An invitation you can adapt and question sets for every outcome are in the questions section below.

Step 4: Code every deal

Give each deal one primary reason and up to two secondary reasons from a fixed list. Separate what the buyer said from what you infer, and note your confidence: high when the buyer and the call evidence agree, lower when you are reading between the lines.

Have two people code the first batch independently and compare. The disagreements show you where your reason codes are vague, and fixing those early saves a painful re-code later.

Step 5: Find patterns, not anecdotes

Cut the coded deals by segment, competitor, stage of loss, deal size, lead source and rep tenure. A pattern is something that appears across several deals. A vivid story from one loss is a lead to investigate.

Write findings in a fixed shape: "In [segment], [X] of [Y] losses to [competitor] cited [criterion] as the deciding factor, and the pattern started at [stage]." Then add the recommendation. That is the skeleton of a one-page win-loss analysis report, and the template below includes one.

Step 6: Close the loop

Every finding gets an action, an owner and a due date, tracked somewhere people actually look. Next quarter, re-measure the same segment and report what moved. Tie it back to the numbers leadership already watches, as you would when you measure enablement impact.

Then tell the sales team what changed because of what buyers said. If your reps see their feedback become a new battle card or a fixed pricing page, they have a reason to log better data on the next deal.

The free win-loss analysis template

Here is the deal record the template uses. Fill one in per deal, ideally within a couple of weeks of the close. The same record works in a spreadsheet, a CRM object or a shared doc.

• Deal basics. Fields: deal name and ID; account; segment; deal size; sales cycle length; owner; lead source; close date. Capture the facts you can pull straight from the CRM.
• Outcome. Fields: won, lost or no-decision; stage where it turned. Capture where the decision actually changed, which may differ from the stage where the CRM was updated.
• Competitive context. Fields: competitors in the evaluation; incumbent; vendor chosen. Include "none" and "status quo" as valid answers.
• Buying group. Fields: roles involved; champion; economic buyer; blockers; stakeholders who went quiet. Capture who was in the room and who should have been.
• Decision criteria. Fields: buyer's top three criteria, in their order; how each vendor scored on them. Capture the buyer's ranking, not your feature list.
• Reasons. Fields: primary code; up to two secondary codes; the buyer's words. Capture a short quote, or a paraphrase marked as a paraphrase.
• Evidence. Fields: sources used (interview, survey, call, CRM); confidence of High, Medium or Low. This lets a reader judge how much weight a finding deserves.
• Action. Fields: insight; recommended action; owner; due date; status. This is the row that turns analysis into work.

To build it with an LLM, paste this into ChatGPT, Claude or Gemini and fill the brackets:

Act as a RevOps analyst. Build a win-loss analysis template in [Excel, Google Sheets or your CRM] for [team and segment]. Our deal stages are [list] and our competitors are [list].

Use one row per closed deal and 20 to 25 columns covering deal basics, outcome, competitive context, buying group, decision criteria, reasons (one primary code, up to two secondary), evidence with a High, Medium or Low confidence, and action (owner, due date, status). Give the header row as tab-separated text.

Add drop-downs for outcome (Won, Lost, No decision), reason codes, confidence and status. Use only these reason codes: [paste list]. Write formulas for win rate, win rate including no-decision, and losses by reason. Name the column each formula reads, assume rows 2 to 500, and test each on five made-up deals. If a bracket is blank, ask me first.

A template nobody fills in is a nicely formatted blank page.

The one-page report that goes with it

Once a batch of deals is coded, summarize it on one page:

• Headline: the sentence a busy VP should remember
• Scope: segment, time window, deals by outcome, and the mix of evidence behind them
• Three patterns: each with counts, a buyer quote and a confidence rating
• What we're still learning: thin samples and places where the evidence disagrees
• Actions: each with an owner, a due date and the metric you'll re-check next quarter

A reason-code list you can paste into your CRM

Reason codes are the spine of a win-loss program. Free-text reasons are rich and impossible to count. A fixed list is easy to count and a little blunt. The sweet spot is a list specific enough to point at an owner, and short enough to use in seconds because reps pick a family first, then a code inside it.

This one has eight families and 34 codes. Each family maps to a team that can act on it. Reps choose one of eight families, then one of two to six codes.

• Product and fit. Capability coverage; Integrations; Usability and adoption; Performance and reliability; Roadmap confidence. Owner: Product.
• Price and packaging. Price level versus budget; Packaging or licensing structure; Discounting and terms; Total cost of ownership. Owner: Product marketing and Finance.
• Competition. Incumbent or status quo vendor; Named competitor (record which one); Build in-house; Spreadsheet or manual process. Owner: Product marketing.
• Messaging and positioning. Value clarity; Differentiation; Use case and persona fit; Category clarity. Owner: Product marketing.
• Sales execution. Discovery depth; Demo relevance; Follow-up speed; Multi-threading across the buying group; Proposal quality; Negotiation handling. Owner: Sales leadership and Enablement.
• Proof and trust. Case study relevance; Customer reference feedback; Security or compliance review; Vendor viability. Owner: Enablement and Marketing.
• Buying process. No decision (priority shifted); Budget freeze or reallocation; Champion change; Procurement or legal timing; Internal alignment in the buying group. Owner: Sales leadership.
• Timing. Implementation timeline; Contract cycle fit. Owner: Sales operations and Customer success.

Five rules keep the list healthy:

• One primary code, up to two secondary. The primary is the factor that tipped the decision. Secondary codes record what else mattered.
• Keep no-decision separate from competitive losses. They need different fixes. A lost-to-competitor deal calls for positioning work. A stalled deal calls for a business case and a stronger champion.
• Code the wins with the same list. Most codes are neutral topics, so "won on differentiation" and "lost on differentiation" both work. The Buying process codes mostly explain no-decisions and stalls.
• Allow "Other" with required free text, and review it monthly. Recurring "Other" entries are new codes waiting for a promotion.
• Don't rename codes mid-quarter. Trend lines depend on stable definitions.

Win-loss analysis questions to ask buyers

Good win-loss questions are open, neutral and boring to defend against. You are not there to rebut anything. You are there to learn what the buyer saw.

A short invitation you can adapt

Subject: Twenty minutes on how you chose a vendor

Hi [Name], thanks for considering [Company] during your recent evaluation. We are talking with a handful of buyers to understand how decisions like yours get made, whether or not they went our way. A 20-minute conversation, no sales pitch, and we will send a [thank-you] for your time. Would [two time options] work?

Opening, for every outcome

1. What was happening in your business that made you look at this category in the first place?
2. Who was involved in the decision, and what was each person's role?
3. What would have happened if you had done nothing?

Evaluation

1. Which options did you consider, and how did you find them?
2. What criteria mattered most, in order of importance?
3. How did each vendor compare on those criteria?
4. Which content, demos or references helped you most? Which fell flat?
5. Was there a moment when your opinion of a vendor changed?

The decision

1. What was the deciding factor?
2. Who had the final say, and did anyone disagree?
3. What nearly changed the outcome?

Experience with our team

1. How did our team compare with the others you met?
2. Did you get what you needed, from the right people, at the right time?
3. What should we have asked you that we didn't?

Extra questions for losses

1. What would have had to be true for us to win?
2. What would you tell our product team?
3. What did the winning vendor do that we should know about?

Extra questions for wins

1. What almost made you choose someone else?
2. What are you most hoping, or worried, about in rollout?
3. Which parts of our process did you tell colleagues about?

Extra questions for no-decisions

1. What changed inside the company after the evaluation started?
2. What would need to happen for this to restart?
3. Who still needs convincing, and what would convince them?

Interviewer habits that pay off. Record with permission. Aim follow-ups at moments (what happened in that meeting?) rather than opinions (what did you think of the demo?).

Ask "can you give me an example?" more than you ask anything else.

A five-minute win-loss analysis survey

When volume makes interviews impractical, send this within days of the close. Questions 1 to 5 use a 1 to 5 scale.

1. Overall, how would you rate your experience with [Company] during the evaluation?
2. How well did the product meet your needs?
3. How did our pricing and packaging compare with your expectations?
4. How well did our team understand your business?
5. How helpful were the content, demos and references we shared?
6. Which factor mattered most in your final decision? (Offer the eight reason-code families as choices.)
7. What is one thing we should have done differently?
8. May we follow up for a 20-minute conversation? (Yes or no)

Question 6 matters most: it hands you a buyer-chosen code to set beside the one your team assigned.

How win rate and win-loss ratio are calculated

Win rate and win-loss ratio are cousins that get mixed up. These are the definitions most teams use:

• Win rate = won deals ÷ (won deals + lost deals)
• Win rate including no-decision = won deals ÷ (won + lost + no-decision)
• Win-loss ratio = won deals : lost deals
• Value win rate = value of won deals ÷ (value of won + value of lost)
• Stage conversion = deals that reached stage N+1 ÷ deals that reached stage N

Here is the arithmetic on an illustrative quarter with 100 closed opportunities: 30 won, 45 lost and 25 no-decision.

• Win rate: 30 ÷ (30 + 45) = 40%
• Win rate including no-decision: 30 ÷ (30 + 45 + 25) = 30%
• Win-loss ratio: 30 : 45 = 2 : 3
• Value win rate (won $1.5M, lost $2.5M): 1.5 ÷ (1.5 + 2.5) = 37.5%

Same quarter, two win rates: 40% or 30%, depending on whether the 25 stalled deals count. Neither is wrong. They answer different questions. The first measures how you do when a buyer decides. The second measures how many buying efforts end in a signed deal. Pick one as your headline, label it, and stay consistent.

What is a good win-loss ratio? There is no universal number. Win rates swing with segment, deal size, lead source and, most of all, how your team defines a qualified opportunity. A team that qualifies tightly can show a higher win rate than one that opens an opportunity on the first call. Benchmark against your own history by segment, and treat published averages as loose context.

A worked win-loss analysis example

Every number below is illustrative and the customer is a composite, so you can see what a filled-in record and a quarter-end roll-up look like.

One deal record

• Deal: Mid-market software company, $48,000 annual contract, 62-day cycle
• Outcome: Lost, turned after the proposal stage
• Competitive context: Evaluated two vendors; chose a competitor already used by a sister team
• Buying group: Champion in marketing ops; economic buyer in finance joined late; IT never attended a meeting
• Decision criteria: 1. Integration with the existing stack. 2. Time to launch. 3. Price
• Primary code: Sales execution: Multi-threading across the buying group
• Secondary codes: Product and fit: Integrations; Buying process: Internal alignment in the buying group
• Buyer's words (paraphrase): Finance and IT never saw a plan that addressed their concerns, so the safer, already-approved option won
• Evidence and confidence: Interview plus call recordings; High
• Action: Add a stakeholder map to the discovery stage; build an IT-facing integration brief; owner: enablement lead; due in 30 days

One quarter, rolled up

The 45 losses from the earlier math, coded by primary reason:

• 12 losses. Messaging and positioning: Differentiation
• 10 losses. Sales execution: Multi-threading across the buying group
• 6 losses. Competition: Incumbent or status quo vendor
• 6 losses. Price and packaging
• 5 losses. Product and fit
• 3 losses. Proof and trust
• 3 losses. Other
• 45 losses in total.

The 25 no-decisions split as 11 priority shifted, 8 internal alignment issues and 6 champion changes.

Three findings fall out of that roll-up.

• Messaging and multi-threading together account for 22 of the 45 losses, nearly half, and both are fixable without touching the product.
• The no-decision split points at the buying group again. Internal alignment and champion changes explain 14 of 25.
• Price is the primary reason in 6 deals, and the incumbent or status quo wins another 6. Both are real, and together they still trail messaging plus multi-threading.

The roll-up turns into three actions: a differentiation rewrite owned by product marketing, a stakeholder-mapping play owned by enablement, and a business-case template for champions owned by sales leadership. Each has an owner, a date and a metric to re-check next quarter.

Win-loss analysis in Salesforce and HubSpot, plus the dashboard to build

You don't need new software to start. Most CRMs, Salesforce and HubSpot included, let you add what a win-loss program needs.

Fields to add to the deal or opportunity record

• Primary reason code (picklist from the list above) and up to two secondary codes
• Competitor (picklist, with "none" and "status quo")
• Stage where the deal turned (picklist matching your stages)
• Decision criteria (short text, buyer's top three)
• Evidence source and confidence (picklists)
• Win-loss summary (long text, two sentences maximum)

Make it stick. Require the reason fields when a deal moves to a closed stage, using a required-field rule or a validation rule. Require the same fields on won deals. Keep the picklists short enough to scan in a few seconds.

The win-loss analysis dashboard

Six views cover most of what leadership wants to see:

1. Win rate trend by segment, with your chosen definition labeled on the chart
2. Loss reasons by segment, as counts stacked by family
3. Win rate by competitor, including "status quo" and "none"
4. Stage of loss, to show where the deal turned
5. No-decision trend, separate from competitive losses
6. Won versus lost cycle length, because stalls leave tracks in the calendar

Review the dashboard monthly and read the numbers beside the interview notes.

Counts tell you where to look. Buyer quotes tell you what you're looking at.

Reading live deal signals with pf1.ai

Interviews explain a decision after it lands. Signals from inside the deal show you the decision forming. pf1.ai, from the house of Paperflite, is an AI revenue enablement platform where agents work across content, deals and coaching, and two of them feed a win-loss program directly.

Deal agent reads each deal while it is open. Opportunities get an AI win probability with a trend line, built from CRM data, calls, emails, calendar activity and how buyers engage with shared content. It also shows whether each person on the buying committee is engaged, cooling or has gone dark. For a win-loss file, that gives you the "when" and the "who" before you book an interview: the probability line dipped the week the economic buyer stopped showing up, so that is where your questions start.

The Deal Agent handles the follow-up questions. Reps ask it in plain language about a deal's health, risks and next steps, and managers can query the pipeline the same way. Each answer points to the call, email or content engagement behind it. A manager prepping a loss review can ask about a deal's health and risks instead of scrolling through an activity log.

When your roll-up says multi-threading cost you ten deals, the Coach agent is where the fix gets practiced. It annotates the moments of interest in calls, scores them against the coaching framework your team uses, and generates practice scenarios from the gaps it finds, with AI buyers built around the real deal context. Reps get a place to rehearse bringing finance and IT into the room.

How this sits beside interview programs. Specialist providers run buyer interviews that capture the buyer's voice at depth, with a neutral third party asking the questions. pf1.ai shows the when, the who and what happened next while the deal is live. Interviews add the buyer's own why. Teams that want both use the signals to decide which deals deserve an interview. To see those signals on your own pipeline, book a demo

Turning loss patterns into coaching, content and battle cards

A finding becomes useful when it has a build item and a practice item attached. Use the pattern in your data to pick the lever.

• Losses cluster after the demo. Lever: demo relevance. Build: a demo story per persona and use case. Practice: demo run-throughs with objection handling.
• Losses to one competitor. Lever: positioning against that vendor. Build: a battle card and a fair comparison page. Practice: role-play against that buyer's priorities.
• Price named often. Lever: value framing. Build: rOI proof and a packaging explainer. Practice: negotiation and value-conversation scenarios.
• Single-threaded deals. Lever: reaching the whole buying group. Build: a stakeholder map for discovery. Practice: calls that bring finance, IT and legal in early.
• Rising no-decision rate. Lever: champion strength. Build: a business case template. Practice: discovery that quantifies the cost of doing nothing.
• Proof named as missing. Lever: evidence library. Build: case studies by segment, security pack. Practice: telling a customer story in under two minutes.

Competitive losses deserve their own asset. A well-kept library of battle cards gives reps the buyer-tested version of the story in the moment, and every new win-loss finding is a reason to refresh one.

For the research side, start with this competitive intelligence template and feed it what buyers tell you.

On the people side, match the practice to the pattern instead of assigning one generic course to everyone. This roundup of sales coaching tools covers the options for call review, role-play and learning paths.

Who owns win-loss analysis

Pick one accountable owner. Product marketing often fits well because the findings touch positioning, competitors and messaging. Where product marketing is thin, RevOps or an enablement lead can take it. The other teams play defined parts:

• Product marketing or the program owner: runs the interviews, codes the deals, writes the report
• RevOps: keeps the fields, picklists and dashboards clean
• Enablement: turns findings into content, training and practice
• Sales leadership: owns the behavior changes and answers the "reps will hate this" objection
• Product: takes roadmap and packaging inputs and answers back

Fold the output into your wider sales enablement strategy so findings arrive where planning already happens.

A 45-minute monthly review and a 90-day launch plan

The monthly review

• 15 minutes: Top three patterns from this month's coded deals, with evidence
• 10 minutes: One win and one loss, walked through end to end
• 10 minutes: Decisions: what changes, who owns it, by when

The 90-day launch plan

Days 1 to 30: set the foundation. Agree the definitions of win rate and no-decision. Publish the reason codes. Add the CRM fields and make them required at close. Pick the first segment and its single question. Code the last quarter's closed deals from CRM and call evidence so you have a baseline.

Days 31 to 60: collect and code. Send survey requests, book the first batch of interviews, and code each deal as the evidence arrives. Have two people code the first batch independently and reconcile the differences.

Days 61 to 90: report and act. Publish the first one-page report, ship at least one action from it, hold the first monthly review, and tell the sales team what changed because of what buyers said.

Win-loss analysis best practices that keep a program alive

Most programs fade for ordinary reasons. These habits keep yours going.

• Interview beyond the champion. A champion gives you one view of the decision. Ask to speak to the economic buyer or the person who raised concerns.
• Don't let reps code their own losses unreviewed. Reps coding their own losses may lean toward factors outside their control. A second reader catches that.
• Weigh recency honestly. A fresh, vivid loss is the loudest voice in the room. Count before you conclude.
• Resist the tiny-sample overreaction. Three deals make a theory, not a pricing change.
• Keep the report short. One page that gets read beats ten that get filed.
• Close the loop with buyers and reps. Tell interviewees their input shipped something, and tell reps the same.

Start with one segment and one quarter of closed deals

Win-loss analysis pays off fastest when it starts small and specific. This week, do three things: agree your win rate definition, add the required reason fields to your CRM, and code the last quarter of closed deals in one segment. You will have a baseline in days and a short list of questions worth asking buyers.

From there, add interviews, add call evidence, and let the monthly review turn patterns into work. The first real pattern in your own data does the convincing.

Frequently asked questions