Win-Loss Pattern Analysis

Paste your closed-won and closed-lost notes. You get the patterns your CRM reason codes hide: what actually separates the wins, which stated loss reasons are cover stories, the stage where deals really die, and the three changes with the best evidence behind them.

A Markdown win-loss analysis containing: (1) win and loss counts for the sample; (2) 'What the wins share' with supporting deal labels; (3) 'What the losses share' with supporting deal labels; (4) a stated-versus-likely loss reason section; (5) 'Where deals actually die'; (6) segment patterns only where supported; (7) three to five ranked changes with their evidence and expected effect. No percentages, no significance claims, and every inference explicitly marked as one.

# Win-loss pattern analysis — 4 wins, 6 losses

Ten deals is a small sample and everything below is a pattern with a count, not a finding. Read it as a list of hypotheses ranked by how much the notes support them.

## What the wins share

**A champion with budget authority or direct access to it (W1, W3).** W1's champion was an ops director; W3's was a regional manager. Both sit above store level. No win in this sample had a champion below that line.

**An external forcing event (W2, W4).** W2 came inbound after a compliance fine. W4 came inbound and closed in five weeks — [INFERENCE: the notes do not state a trigger for W4, but a five-week close in this market almost always has one. Worth asking the rep.]

**Prior category experience (W1).** The champion had used a competitor before, so no time was spent establishing that the category exists. This appears in only one win, so treat it as a weak signal.

**Smaller estates closed faster.** W4 at 15 sites closed in five weeks; W3 at 40 sites took four months. Two data points, and the direction is unsurprising, but it is consistent with the loss pattern below.

## What the losses share

**No access to the economic buyer (L1, L5, L6).** L1 ran six months without ever meeting the CFO. L5 never got past the ops manager. L6's champion was a store manager with no budget authority. This is the single most common factor in the losses and it is present in no win.

**Larger estates lost more often.** L1 at 100 sites, L3 at 80, L5 at 50. The largest win in the sample is W1 at 60 sites. [INFERENCE: above roughly 50 sites you are meeting a procurement process and an incumbent relationship you are not currently equipped for.]

**Champion fragility (L3).** The champion left in month three and the deal died with them. One deal, but it is the clearest illustration of single-threading in the sample, and single-threading is visible in L1, L5 and L6 too.

## Stated loss reason versus likely loss reason

| Deal | Stated | Do the notes support it? |
|---|---|---|
| L1 | Too expensive | **No.** Six months with no CFO contact means price was never actually tested against a budget. 'Too expensive' here means 'nobody senior enough ever agreed it was worth buying'. Recorded as a pricing loss; it is an access loss. |
| L2 | Went with incumbent | **Partly.** The real fact — an existing enterprise agreement — surfaced late. That is not a competitive loss, it is a qualification failure, and it was knowable in week one. |
| L3 | No budget this year | **No.** The champion left in month three. Budget is what the next person says when they inherit a deal they did not start. |
| L4 | Missing features (native payroll export) | **Yes.** The only stated reason in the sample that is specific, verifiable and plausibly the actual cause. |
| L5 | Timing | **No.** Never got past the ops manager. 'Timing' is the most common polite decline in this sample and it carries no information. |
| L6 | Too expensive | **No.** A store manager with no budget authority cannot find anything expensive; they can only fail to get it approved. Access loss, recorded as price. |

**The headline:** three of six losses are recorded as price or timing, and in the notes none of the three is. Your CRM reason codes are currently telling you to change your pricing. The notes are telling you to change who you sell to.

## Where deals actually die
Recorded stage of death is late — month three, month six, negotiation. The notes suggest the determining moment is much earlier, in every case at the point where the champion is identified.

L1, L5 and L6 were all effectively decided in the first two calls, when the deal became single-threaded to someone without budget authority. Everything after that was a well-run process on a deal that could not close. L2 was decided before the first call, by a contract nobody asked about.

[INFERENCE — but it is the strongest inference available from this sample.]

## Segment patterns

**Hospitality inbound closes fast (W2, W4).** Both inbound, both hospitality, both quick. Two deals is not a pattern, but it is worth checking against the full pipeline because it would change where marketing spends.

**Retail above 50 sites is currently unwinnable for you.** L1 at 100, L3 at 80, L5 at 50 lost; the largest retail win is W1 at 60. On four data points this is a hypothesis, not a rule — but it is the one worth testing next, because it determines your target account list.

## Changes, ranked

**1. Add a hard qualification gate on economic buyer access.** No deal advances past discovery without a named budget holder and a date to meet them. Evidence: L1, L5, L6 — the three cleanest losses in the sample, all with the same cause. Expect fewer deals in pipeline and a higher win rate on what remains, and expect the pipeline reduction to be unpopular.

**2. Ask about existing enterprise agreements on the first call.** Evidence: L2 alone, but the cost was a full sales cycle and the question takes ten seconds. Cheapest change on this list.

**3. Stop accepting 'too expensive' and 'timing' as loss reasons.** Require the rep to record who declined and at what level. Evidence: four of six losses have a reason code the notes contradict. Until this changes, every subsequent analysis inherits the same distortion.

**4. Multi-thread every deal by the second meeting.** Evidence: L3 directly, and it would have limited the damage in L1 and L5. Cheap, and standard practice.

**5. Decide about native payroll export deliberately.** Evidence: L4 only. One deal is not a roadmap case, but it is the only genuine product gap in the sample. Check how often it appears in deals not in this set before building anything.

_What this cannot tell you: everything here comes from notes written by the rep who lost the deal. The reason a buyer gives a rep is not always the reason. Five buyer interviews — especially L1 and L5 — would be worth more than another twenty deals of notes._