Win/loss analysis · 12 min read · Updated 1 Aug 2026
Win/Loss Report Template (Free Win-Loss Analysis Report)
A blank win/loss report you can fill in today, plus the guidance for what belongs in each field. Structured around what buyers actually said rather than what the CRM close-reason field recorded, because those two are rarely the same answer.
Copy pastes straight into Google Sheets or Excel with the columns intact. Downloads are free with a work email.
The win/loss report template
This is exactly what you get when you copy or download. Blank fields are yours to fill in; each table ships with one example row to show the pattern, which you delete.
Report header
Fill this in first. Say how many deals you reviewed and how you chose them, or nobody can judge how much weight to give the conclusions.
- Reporting periodThe window of closed deals this covers, e.g. Q1 2026
- Deals reviewedHow many, split into won / lost / no decision
- How deals were selectedAll closed deals, or a sample, and on what basis
- Evidence usedBuyer interviews, call recordings, CRM notes, or a mix
- OwnerOne named person, not a team
- Date publishedThe date the findings below were compiled
1. Executive summary
Write this last, put it first. Give the numbers, then the one thing you want changed as a result.
- Win rate this periodWins divided by decided deals, with the raw counts
- Change since last periodDirection and size, e.g. down 6 points from 47%
- Top reason we wonThe most frequent reason, in the buyer's words
- Top reason we lostThe most frequent reason, in the buyer's words
- Decision needed from this reportThe specific change you are asking for
2. Deal-by-deal log
First row is an example, delete it. One row per reviewed deal. The reason column holds the buyer's words, not the rep's interpretation of them.
| Deal (segment, size) | Competitor | Outcome | Primary reason, in the buyer's words | Evidence source |
|---|---|---|---|---|
| ExampleNorthwind, mid-market, 140 seats | Acme Analytics Pro | Lost | "Your reporting could not match their exports" | Loss interview, 6 Mar |
3. Reason analysis
First row is an example, delete it. Group the deal log into a small set of reasons. Six to eight categories is the working limit before the analysis stops separating anything.
| Reason | Appears in wins | Appears in losses | Share of decided deals | What it tells us |
|---|---|---|---|---|
| ExampleReporting depth | 1 | 6 | 23% | Our biggest single loss driver in mid-market |
4. Head-to-head performance
First row is an example, delete it. One row per competitor faced. Win rate against a specific competitor is far more actionable than an overall win rate.
| Competitor | Deals faced | Win rate against them | Why we lose to them | Why we beat them |
|---|---|---|---|---|
| ExampleAcme Analytics Pro | 17 | 41% (7 of 17) | Reporting depth in ops-led evaluations | Time to value when the buyer has a deadline |
5. Where deals were decided
First row is an example, delete it. The stage a deal was lost at tells you which team can fix it. A deal lost at discovery is a marketing and targeting problem, not a closing one.
| Stage | Deals lost here | Most common reason | What it suggests | Whose problem it is |
|---|---|---|---|---|
| ExampleTechnical evaluation | 5 | Reporting depth | The gap surfaces once ops runs a real test | Product, with sales enablement support |
6. What buyers actually said
First row is an example, delete it. Verbatim quotes only. This section is what makes the report persuasive: paraphrase it and it becomes an opinion.
| Quote, verbatim | Deal | Outcome | Theme it supports | Source |
|---|---|---|---|---|
| Example"We liked you more, but we could not get the board comfortable" | Northwind | Lost | Perceived risk at our size | Loss interview, 6 Mar |
7. Movement since the last report
First row is an example, delete it. A single period tells you very little. The comparison is where the report earns its keep.
| Measure | Last period | This period | Change | Most likely explanation |
|---|---|---|---|---|
| ExampleWin rate vs Acme | 52% | 41% | Down 11 points | They shipped native alerting in January |
8. Decisions, owners and dates
First row is an example, delete it. Findings that nobody owns produce the same report again next quarter. Every row needs a person and a date.
| Finding | What we will do | Owner | By when | How we will know it worked |
|---|---|---|---|---|
| ExampleReporting depth drives 23% of losses | Ship scheduled exports and add a reporting proof point to the battlecard | Sam, Product | 30 Jun | Reporting appears in under 10% of loss reasons next quarter |
How to fill in your win/loss report
How to fill in the win/loss report header
The header is where a win/loss report earns or loses its credibility, because it tells the reader how much the findings can carry. Two fields matter most: how many deals you reviewed, and how you chose them. A report drawn from six losses and no wins will produce a confident, wrong conclusion, and without this block nobody can see that.
Reporting period
The window of closed deals, not the window in which you did the interviews: "deals closed 1 Jan to 31 Mar". Deals closed in December but interviewed in February belong to the December period.
Deals reviewed
Give the split: "31 deals: 14 won, 13 lost, 4 no decision". Reports that only review losses systematically overweight product gaps, because the reasons you won are invisible to them.
How deals were selected
State the rule: "all closed-won and closed-lost deals above $20k" or "a sample of 20, weighted to mid-market". If reps chose which deals to submit, say so, because that selection is not random and the reader needs to know.
Evidence used
Name the sources and their mix: "9 buyer interviews, 12 call recordings, rest from CRM notes". These are not equal in quality, and the reader should be able to discount accordingly.
Owner
One named person: "Marie, PMM". Win/loss programmes die when the analysis is everyone's job and therefore nobody's calendar item.
Date published
When the findings were compiled. Win/loss ages faster than most people expect: a reason that dominated last year may have been fixed two releases ago.
How to write the executive summary of a win/loss report
Numbers first, then the single change you want. This block will be the whole report for most of its audience, so it has to survive being read alone. The most common failure is presenting a win rate with no comparison, which gives the reader a number they cannot interpret.
Win rate this period
Always with the raw counts: "45% (14 of 31 decided)". A percentage without counts hides whether you are describing a pattern or four deals, and four deals is the more common case.
Change since last period
Direction and size: "down 6 points from 51%". If this is your first report, say so and state that the number is a baseline rather than a result.
Top reason we won
The most frequent win reason, quoted: "you got us live before our renewal". Teams routinely skip this and end up cutting a strength nobody realised was the strength.
Top reason we lost
The most frequent loss reason, quoted, and honestly. If it is price, resist rewriting it as "value communication" before you have evidence for that reading.
Decision needed from this report
The specific ask: "Approve scheduled exports for Q3, or accept the mid-market loss rate." A win/loss report that ends in observations gets read and forgotten.
How to fill in the deal-by-deal log
This is the raw material every other section is derived from. The one discipline that matters: the reason column holds what the buyer said, not what the rep concluded. Those two disagree more often than any other pair of fields in a sales organisation, and the CRM close-reason field records the second one.
Deal (segment, size)
Enough to segment later: "Northwind, mid-market, 140 seats". Reasons vary enormously by segment, and a report that cannot split by segment will average two different problems into one meaningless answer.
Competitor
The named product they chose instead, or "no decision" where the deal died without one. If the buyer went with an internal build or a spreadsheet, write that: it is a real competitor and it usually wins more often than teams admit.
Outcome
Won, lost, or no decision. Keep no-decision deals in the log. They are excluded from the win rate calculation but they are frequently the largest group, and the reasons behind them are different in kind.
Primary reason, in the buyer's words
Quoted: "Your reporting could not match their exports." Not "product gap", which is your category for their sentence. The categorisation happens in the next section, where it can be seen and challenged.
Evidence source
"Loss interview, 6 Mar", "call recording", "CRM note". A structured buyer interview is worth several CRM notes, and marking which is which prevents a weakly-sourced row from carrying the same weight as a strong one.
How to fill in the reason analysis
This is where the deal log becomes a finding. You are grouping quotes into a small set of reasons and counting them. Two rules keep it honest: keep the categories few, and count wins as well as losses. A reason appearing in wins and losses equally is not a loss driver, however loudly it was said.
Reason
Six to eight categories, defined in plain language: "reporting depth", "price", "perceived risk at our size", "missing integration", "incumbent inertia". More than eight and each holds two deals, which separates nothing.
Appears in wins / Appears in losses
Count both. A reason cited in six losses and one win is a real driver; a reason cited in six losses and five wins is a topic buyers discuss, not a reason they decided. Reports that count only losses miss this entirely.
Share of decided deals
Express as a share so the size is visible: "23%". Include the denominator somewhere, because 23% of 31 deals and 23% of 7 deals warrant different responses.
What it tells us
One line of interpretation, labelled as such: "Our biggest single loss driver in mid-market." Note the segment if it is concentrated in one, because that changes who needs to act.
Treat price carefully
Price is the most over-reported loss reason in every win/loss dataset, because it is the easiest thing for a buyer to say and the most comfortable thing for a rep to hear. Before accepting it, check whether the deals you lost on price were also the ones where you lost the value argument earlier.
How to fill in head-to-head performance
An overall win rate is close to useless for action, because it averages together fights you should win and fights you should not be in. Win rate against a named competitor is what changes behaviour: it tells sales which deals to invest in, and it tells product marketing which battlecard is failing.
Deals faced
The count you actually competed against them in. Below about ten deals, treat the rate as indicative and say so in the row rather than presenting a precise-looking percentage.
Win rate against them
With the counts: "41% (7 of 17)". Track this same figure every period. Its movement is the single most useful number a win/loss programme produces, because it responds to things you control.
Why we lose to them
The dominant reason in that subset, not the overall dominant reason: "reporting depth in ops-led evaluations". Different competitors beat you for different reasons, and merging them produces advice that fits nobody.
Why we beat them
Equally specific: "time to value when the buyer has a deadline". This column feeds directly into the battlecard, and it is the half most teams never fill in.
Feed it back into the battlecards
Every row here should change something in the corresponding battlecard. If your battlecard's "why we win" does not match what buyers say in the deals you won, the battlecard is a hypothesis and this report is the evidence.
How to fill in where deals were decided
The same loss reason means different things at different stages, and the stage is what identifies whose problem it is. A deal lost at discovery is a targeting or messaging problem. The same reason at technical evaluation is a product problem. Reports that skip this section deliver every finding to sales, including the ones sales cannot fix.
Stage
Use the stages your CRM already has, so the report can be reconciled against the pipeline. Do not invent a parallel taxonomy for this document.
Deals lost here
A count. Clustering is the signal: if half of a period's losses happen at one stage, that stage is the finding, ahead of any individual reason.
Most common reason
The dominant reason at that stage specifically. Reasons are not evenly distributed across stages, and the difference is usually the most actionable thing in the report.
What it suggests
The mechanism: "the gap surfaces once ops runs a real test". This is what turns a count into something someone can act on.
Whose problem it is
Name the function: product, product marketing, sales, pricing. A finding delivered to nobody in particular is the reason win/loss reports get read politely and change nothing.
How to fill in what buyers actually said
This section is short and does more work than any other. Executives discount summarised findings and they do not discount a buyer's sentence. Five verbatim quotes will move a roadmap discussion that five percentages will not, which is why paraphrasing here quietly destroys the report's power.
Quote, verbatim
Exactly as said, in quotes, uncorrected: "We liked you more, but we could not get the board comfortable." Tidying the grammar removes the thing that makes it convincing.
Deal and outcome
Attach each quote to a deal in the log above so it can be traced and checked. Anonymise to whatever level your team agreed, and be consistent.
Theme it supports
Link it to a row in the reason analysis: "perceived risk at our size". Quotes that support no theme are anecdotes, and one memorable anecdote has redirected more roadmaps than it should have.
Include wins, not just losses
A quote explaining why someone chose you is as useful as one explaining why they did not, and it protects a strength from being cut in the next prioritisation round.
Get permission and keep it clean
Follow whatever you told the buyer at the interview. Never attribute a quote to a named person or company if the interview was conducted on the basis that it would not be.
How to fill in movement since the last report
One period of win/loss data is a snapshot with a lot of noise in it. The comparison is where the value is: it tells you whether the thing you changed last quarter did anything. Without this section, a win/loss programme produces a series of unconnected observations rather than a feedback loop.
Measure
Keep the same handful of measures every period: overall win rate, win rate against each main competitor, the top three loss reasons, and the no-decision share. Changing what you measure resets the comparison to zero.
Last period and this period
Both figures, in the same form, with counts. If the sample size changed materially, note it: a rate that moved because you reviewed twice as many deals has not moved.
Change
Direction and size: "down 11 points". Resist reading significance into small movements on small samples, which is the most common way these reports mislead.
Most likely explanation
One line, labelled as a hypothesis: "They shipped native alerting in January." Name what would confirm it so the next report can check rather than re-speculate.
Check last period's decisions
The strongest use of this section is to report what happened to the actions from the previous report. It is the cheapest way to turn a stream of documents into an accountable loop, and almost nobody does it.
How to fill in the decisions section of your win/loss report
Win/loss analysis has an unusually poor conversion rate from insight to change, and the reason is structural: the findings are handed to a room rather than to a person. This block fixes that. Three to five rows, each owned, each dated, each with an observable result.
Finding
Traceable to a section above, with its number: "reporting depth drives 23% of losses". A decision with no supporting row is a pre-existing plan looking for evidence.
What we will do
Concrete: "Ship scheduled exports and add a reporting proof point to the battlecard." Note that most win/loss findings have two actions, one for product and one for sales enablement, and teams routinely take only the first.
Owner
A named person per row, and note that the owner is frequently outside the team that wrote the report. Agree it with them before publishing rather than assigning it in the document.
By when
A real date. Where the fix is a roadmap item, the date is when it enters the roadmap, not when it ships, so the row can actually be closed.
How we will know it worked
Expressed in the report's own terms: "reporting appears in under 10% of loss reasons next quarter". This is what lets the next report grade the last one.
Three to five rows
A win/loss report generating fifteen actions has generated none. Pick the ones with the largest share of decided deals behind them.
Sourcing and upkeep: keeping a win/loss report honest
These rules apply to every section above. Win/loss reports fail in a specific way: they gather reasons from the people who were in the room rather than the person who made the decision, and everything downstream inherits that error. The others are sampling problems, and both are avoidable.
Ask the buyer, not the rep
The single highest-value rule. Reps report the reason they were given at the end, which is usually the most comfortable one. Buyer interviews surface the reason the decision actually turned on, and the two differ often enough that a report built only on CRM notes should say so in the header.
Review wins as carefully as losses
Loss-only analysis produces a list of things to fix and no list of things to protect. Half of what a win/loss programme is for is finding out which strengths are actually deciding deals, so they survive the next prioritisation.
Keep the no-decision deals
They are excluded from the win rate but not from the analysis. In many categories no-decision is the largest single outcome, and its causes, budget, priority and inertia, are different from competitive losses and need different responses.
Interview soon, and by someone neutral
Within two to four weeks of the decision, while the reasoning is still recallable. Buyers are measurably more candid with someone who was not in the deal, which is the main argument for a third party or at least for a non-account-team interviewer.
Watch for the price mirage
Price is the easiest answer for a buyer to give and the easiest for a rep to accept. Before you record a price loss, check whether the value case was made and understood earlier in the deal, and whether you won similar deals at similar prices.
Say what the sample can and cannot support
State the counts, and do not present a percentage from eight deals as though it were from eighty. Marking a finding as indicative costs nothing and protects the report's credibility the first time a number moves back.
Run it on a cadence and hold the measures
Quarterly suits most teams. Keep the same measures every period, because the trend section is where the programme actually pays for itself and changing definitions resets it.
How to roll out your win/loss report
- 1Copy or download the blank template. Use Copy to paste it straight into Google Sheets or Excel with the columns intact, or download the CSV, Notion or PDF version.
- 2Decide which deals you are reviewing, and write the rule down. All closed deals above a threshold, or a stated sample. Reps choosing which deals to submit is not a random sample, and the header should say so.
- 3Delete the example rows. Each table ships with one example row so the pattern is obvious. Remove it before you circulate the report.
- 4Interview buyers, not just reps. The reason a rep records and the reason the buyer decided on are frequently different. Use the free win/loss interview questions tool below for the question set.
- 5Log the buyer's words before you categorise them. Fill the deal log with verbatim reasons first, then group them in the reason analysis. Categorising while collecting bakes in your assumptions.
- 6Count reasons in wins as well as losses. A reason appearing equally in both is something buyers discuss, not something they decided on.
- 7Split by competitor and by stage. Win rate against a named competitor tells sales where to invest; the stage a deal died at tells you which team can actually fix it.
- 8Finish with three to five owned decisions. Each with a named person, a real date and an observable result, then open the next report by reporting what happened to them.
Win/loss report FAQ
What is a win/loss report?
A win/loss report is a periodic analysis of why you won and lost the deals that closed in a given window. It combines a deal-by-deal log of the reasons buyers gave, a grouping of those reasons into a small set of drivers, win rates against each named competitor, and a set of owned decisions. The distinguishing feature of a good one is that the reasons come from buyers rather than from the CRM close-reason field.
How do you do a win/loss analysis?
Define the window and which closed deals are in scope, then gather the reason each deal turned on, ideally from the buyer within two to four weeks of the decision. Log each deal with the buyer's words verbatim, group those into six to eight reason categories, and count how often each appears in wins as well as losses. Split by competitor and by the stage the deal was decided at, compare against the previous period, then write three to five owned decisions with dates.
How do you conduct win/loss interviews?
Reach the actual decision-maker rather than your champion, within two to four weeks of the decision. Use someone who was not on the account team: buyers are measurably more candid with a neutral interviewer, which is the main argument for a third party. Ask open questions about how the decision was made and what the alternatives offered, and resist the urge to defend anything. Our free win/loss interview questions tool, linked below, gives you a tailored question set to run these conversations.
What is win/loss data?
Win/loss data is the structured record of closed-deal outcomes and the reasons behind them: the outcome, the competitor faced, the stage the deal was decided at, the reason the buyer gave, and the source that reason came from. The critical distinction is between data collected from buyers and data collected from reps. Both are useful, but only the first tells you what the decision actually turned on, and a report should say which mix it is built from.
What is the win/loss analysis framework?
There is no single standard framework, and lists presenting one as canonical are usually a vendor's methodology. The structure that consistently works has four parts: collect reasons from buyers rather than reps, categorise them into a small stable set so periods can be compared, segment by competitor and by deal stage because the two identify different owners for the fix, and close with owned actions. This template implements exactly that structure.
How do you calculate win rate?
Win rate is wins divided by decided deals: wins ÷ (wins + losses). Deals that ended in no decision are normally excluded from the calculation, because including them measures two different things at once, though you should report the no-decision count separately since it is often the largest group. Always publish the raw counts alongside the percentage, since 45% of 31 deals and 45% of 4 deals justify very different responses.
What is the formula for calculating loss rate?
Loss rate is losses ÷ (wins + losses), the complement of win rate over decided deals. A separate and more revealing figure is the no-decision rate: no decisions ÷ all closed opportunities. In many categories the largest competitor is inertia, and a report that only tracks wins against losses will never show that, because the deals that died without a decision have quietly left the denominator.
What is a win/loss ratio, and what does a 2.0 win/loss mean?
The win/loss ratio is wins divided by losses, expressed as a single number or as a pair. A ratio of 2.0, also written 2:1, means two wins for every loss, which is a 66.7% win rate over decided deals. A 1:1 ratio is 50%. The ratio and the rate carry the same information; sales teams generally find the percentage easier to compare across segments, while the ratio is more common in sports coverage, which is where most searches for the term originate.
What is a good win/loss ratio?
There is no credible universal benchmark, and we would rather say so than repeat a number with no methodology behind it. Win rates vary enormously by segment, deal size, whether inbound or outbound, and how disciplined a team is about qualifying out early, so a figure that is excellent in one context is poor in another. Three comparisons are genuinely useful: your own rate over time, your rate against a specific named competitor, and your rate by segment. A team improving from 38% to 45% against its main competitor has learned something real; a team comparing itself to an industry average has not.
How do you average wins and losses across a period?
Do not average the periods' percentages. Sum the underlying counts and calculate once: total wins ÷ total decided deals across the whole window. Averaging percentages gives every period equal weight regardless of how many deals it contained, which lets a quiet month with three deals distort the result. If you want the trend, plot each period's rate with its counts visible.
What are three common mistakes companies make when doing win/loss analysis?
First, asking reps instead of buyers: the reason a rep records is the one the buyer found comfortable to give, and it is disproportionately price. Second, analysing only losses, which produces a list of things to fix and no list of strengths to protect, so a genuine advantage gets deprioritised in the next planning round. Third, ending in findings rather than owned actions with dates, which is why so many win/loss programmes produce four quarters of consistent insight and no change.
Why is price the most reported loss reason?
Because it is the easiest answer for a buyer to give and the least uncomfortable for a rep to hear, so it survives both ends of the conversation. Before accepting a price loss, check two things: whether you won comparable deals at comparable prices in the same period, and whether the value case was made early enough to be weighed. If both are true, the deal was more likely lost on perceived value or on risk, and recording it as price sends the finding to the wrong team.
Should you include no-decision deals in a win/loss report?
Yes, in the analysis, but not in the win rate calculation. No-decision is frequently the single largest outcome, and its causes, budget disappearing, priority shifting, or a buyer deciding the problem was tolerable, are different in kind from losing to a competitor and need different responses. Report the count and the reasons separately so the biggest category does not become invisible.
How many deals do you need for a win/loss report to be meaningful?
Enough to see the same reason more than once, which in practice means roughly fifteen to thirty reviewed deals per period for most B2B teams. Below that, run the report anyway but label the findings as indicative and lead with the verbatim quotes rather than the percentages. A quote from one buyer is honest evidence; a percentage derived from six deals presented to two decimal places is not.
How often should you produce a win/loss report?
Quarterly suits most teams: it accumulates enough closed deals to see a pattern while staying close enough to the decisions to act. Interviews themselves should happen continuously, within two to four weeks of each close, because recall decays quickly. The report is the periodic assembly of a continuous collection process, not a quarterly scramble to remember what happened.
Who should own win/loss analysis?
Product marketing usually owns the analysis and the report, sales operations supplies the deal data, and the interviews are best run by someone outside the account team. Ownership of the resulting actions almost always sits elsewhere, typically with product or pricing, which is why the decisions block names an owner per row and why those owners should agree to their row before the report is published.
What is the difference between a win/loss report and CRM close-reason data?
Close-reason fields record what the rep selected from a dropdown when the deal closed, chosen quickly and often from a list that does not fit the real answer. A win/loss report records what the buyer said when asked properly, in their own words, with the stage and the competitor attached. The two disagree often enough that treating close-reason data as win/loss analysis is the most common reason a programme produces no useful findings.
What does win/loss mean in Excel, and is it the same thing?
No, and it is a genuinely confusing overlap. In Excel, a win/loss chart is a type of sparkline: a compact in-cell chart that shows each value as an identical up or down marker, used for streaks such as sports results. It has nothing to do with win/loss analysis in sales. If you want to chart this report in Excel, a simple column chart of reason counts and a line chart of win rate by period will serve you far better than a sparkline, which deliberately discards magnitude.
Can you run a win/loss report in Excel or Google Sheets?
Yes, and for most teams starting out it is the right tool. Use Copy above to paste the full structure into a sheet with the columns intact, or download the CSV, with the deal log on one tab and the summary sections deriving from it. The constraint appears at scale: the interviews, not the spreadsheet, are the expensive part, and a sheet will not tell you that a competitor changed their pricing three weeks before your win rate against them dropped.
Do you need a win/loss analysis tool?
Not to start. A structured question set, a calendar habit and this template will take a team a long way, and the discipline matters more than the software. Tools earn their place at two points: when interview volume outgrows what one person can schedule and transcribe, and when you want deal outcomes read against what competitors were actually doing at the time, which is a competitive intelligence problem rather than a survey one.
Build win/loss reports on live competitor data
Flares tracks what each competitor changed, so every loss reason is read against current facts rather than guesses.
Discover Flares14-day free trial · 30-second setup