--- name: deal-a3 description: "The same A3 discipline applied to a losing pattern in the pipeline instead of a shop-floor problem: a Pareto over loss reasons, a root cause chain that survives scrutiny, and countermeasures a sales leader can actually act on. Use when: A recurring, bounded loss pattern with real data behind it — a segment, a deal source, or a stage where win rate has a measurable gap against its own history. You can pull real CRM loss-reason and stage data, or review actual call/deal notes — not just recall what usually goes wrong. Not for: One specific lost deal with genuinely one-off circumstances (the buyer's company was acquired, budget was cut company-wide) — that's a debrief, not a pattern worth an RCA. You don't have — and can't pull — real loss-reason or stage data. A deal A3 built from impression, not data, inherits every bias in that impression, same as any other A3." --- # Deal A3 — win-rate root cause analysis The same A3 discipline applied to a losing pattern in the pipeline instead of a shop-floor problem: a Pareto over loss reasons, a root cause chain that survives scrutiny, and countermeasures a sales leader can actually act on. Family: Problem solving / root cause analysis · Format: .xlsx · Domain: commercial ## What it is A deal A3 is the same A3 — background, a Pareto-driven root cause chain, a tested countermeasure tracked against a TAGS metric — pointed at a sales problem instead of an operational one. The lean connection is direct, not decorative: a sales team that says "reps need to try harder" every quarter is doing exactly what a shop floor does when it blames operators instead of the process, and it fails for the same reason. A3 discipline forces the same question sales usually skips — what specifically, in the process, produces this pattern of losses — rather than settling for a vibe about the pipeline. ## When to use it Use it when: - A recurring, bounded loss pattern with real data behind it — a segment, a deal source, or a stage where win rate has a measurable gap against its own history. - You can pull real CRM loss-reason and stage data, or review actual call/deal notes — not just recall what usually goes wrong. - You want a record that survives past the QBR it was presented in, with countermeasures someone is actually accountable for. Not when: - One specific lost deal with genuinely one-off circumstances (the buyer's company was acquired, budget was cut company-wide) — that's a debrief, not a pattern worth an RCA. - You don't have — and can't pull — real loss-reason or stage data. A deal A3 built from impression, not data, inherits every bias in that impression, same as any other A3. - The real issue is product-market fit or ICP at the segment level, not a fixable process gap in how deals are run — that's segmentation/ICP work, not a deal-level A3. ## How to fill it in - **background** — State the pattern in plain terms and why it matters now — a win-rate trend, a segment, a deal source. One paragraph. If you can't say why this matters to the forecast in one paragraph, the pattern isn't bounded enough yet. - **problemStatement** — One or two sentences, separate from the background: the specific, bounded gap — which segment or source, what the win rate actually is versus its own standard. Not "we're losing too many deals." - **metric** — Standard is the win rate this segment has historically run at; Target is the goal you're committing to. Actual is a real time series pulled from the CRM, not a single quarter-end number — a trend is what tells you whether this is a blip or a pattern worth an RCA. - **pareto** — Rank loss reasons from real CRM data, and choose the axis on purpose: ranking by count of deals lost is the most common way this misleads, because a handful of large deals lost to the same cause can matter more than a long tail of small ones. Dollar value of lost pipeline is usually the right axis for a forecast-driving RCA. - **fiveWhys** — Start from the Pareto's vital-few loss category, not "we're losing deals" in general, and follow the strongest evidence — actual call notes, CRM stage history — at each step rather than the most convenient story. Stop when you reach something the process actually controls, not "the rep should have worked harder." - **countermeasures** — Each countermeasure must trace to the stated root cause in the sales process — a stage gate, a script, a qualification requirement — not a general exhortation to sell better. Name an owner and a due date; a countermeasure with no owner doesn't happen in sales any more than it does on a shop floor. - **effectConfirmationPlan** — Decide before you act which number moves first. A process change usually shows up in a leading indicator (stage conversion, a logged qualification field) before it shows up in win rate itself, which lags by a full sales cycle. Track both. - **followUpActions** — Capture what makes the fix stick — playbook updates, rep training, CRM field/stage-gate changes. This is what turns a good quarter into a durable process change instead of a one-time correction. ## What good looks like The example below is a full worked deal A3: a TAGS box and trend showing a real win-rate gap against its own historical standard; a Pareto over loss reasons weighted by lost pipeline value, not deal count — including a second-level stratification of the dominant loss category; a causal chain that ends in a specific, fixable gap in the qualification process rather than "reps need to try harder"; and countermeasures that each name the process gap they close. Full worked example, on-screen and as a downloadable .xlsx: https://www.katafacts.com/katas/deal-a3 ## Common mistakes - **Root cause lands on "reps need to work harder" or "reps aren't following the playbook."** — Same failure mode as blaming operators on a shop floor — it's rarely wrong that more effort would help, and it's never actionable. Ask why the playbook wasn't followed, or why following it wasn't enough; that's usually a gap in the process itself, which is something a manager can actually fix. - **The Pareto is ranked by count of lost deals instead of lost pipeline value.** — A long tail of small lost deals can outnumber a handful of large ones without outweighing them — ranking by count routes attention to the wrong category. Choose the axis on purpose and say which one. - **Loss reasons come from the CRM's dropdown field, filled in from memory at deal-close time, not from the actual deal history.** — A rep closing out a deal under time pressure tends to pick the easiest plausible reason, not necessarily the real one. Where it matters, go back to call notes and CRM stage history rather than trusting the dropdown alone. - **Countermeasures are playbook-wide retraining with no link to the specific root cause.** — General retraining fixes the mood in the next sales meeting, not the process gap — and afterward there's no way to tell whether it actually moved the number that mattered. - **Win rate is tracked as a single number instead of a real time series with its own standard.** — One quarter-end percentage can't tell you if this is a new pattern or normal variation. The standard (historical baseline) is what makes the gap real instead of a guess. ## What it connects to Upstream: - **Ideal customer profile (ICP) canvas** (`icp-canvas`) — A deal A3's root cause often traces back to who's being pursued in the first place, not just how the deal was run once it started — a pattern of losses in a specific segment is sometimes an ICP signal, not a process gap. - **Discovery call standard work** (`discovery-standard-work`) — Feeds the deal A3's Pareto directly when the loss pattern concentrates at or after discovery — a standardized discovery process is what makes "which step failed" answerable from data instead of memory. - **5-Whys** (not in the catalogue yet) — The root cause section runs a 5-Whys inline, starting from the Pareto's vital-few category — same as canonical A3. Downstream: - **Mutual action plan** (`mutual-action-plan`) — When a deal A3's countermeasure is process-level (a new stage gate, a new qualification question), the mutual action plan is where that discipline gets applied deal-by-deal going forward. - **Pipeline management board** (`pipeline-management-board`) — The confirmation plan's leading indicator (stage conversion, a logged qualification field) belongs on the team's regular pipeline review, not just checked once at the end of the confirmation window. - **Sales funnel value stream map** (`sales-funnel-map`) — A deal A3 explains one loss pattern in depth; a funnel value stream map shows where WIP and conversion loss concentrate across the whole pipeline, which is often where the next deal A3's topic comes from. ## Where AI helps Judgement — stays with you: - Deciding which root cause is real, not just plausible, for this specific segment - Choosing the Pareto axis — count, dollar value, or another impact measure — for this specific pipeline - Committing to a target win rate and owning it - Judging whether a stage-gate or playbook change actually closed the gap, not just moved it Analysis — AI helps: - Pressure-testing a why-chain for gaps or unjustified leaps ("reps need to try harder" flagged as one step short) - Flagging when a loss-reason Pareto looks misleading for the stated axis - Drafting a first-pass background and problem statement from raw CRM export or call notes - Checking that every countermeasure traces to a stated root cause in the process, not a general exhortation Drudgery — automated: - Computing Gap from Target and the latest Actual win rate, every time the series updates - Formatting the page and keeping it to one sheet - Rolling owner and due-date fields into a tracked action list - Exporting to xlsx (and pptx for QBR report-outs) in the house format --- Deal A3 — win-rate root cause analysis is licensed [CC BY 4.0](https://creativecommons.org/licenses/by/4.0/). Attribution: Katafacts (katafacts.com). ## Using this skill Apply this method as part of whatever broader task, instructions, or deliverable you're already working on -- it's a method to use, not a standalone conversation to start. There is no generation tool for this catalogue yet, so draft the artifact's actual content yourself, following the fields above, the same way a person filling this out by hand would. The canonical guide page and a downloadable worked example (.xlsx) are at https://www.katafacts.com/katas/deal-a3.