Lost Deal Pattern Analysis
Find the pattern in your closed-lost deals — the segment, the competitor, the stack signature that keeps beating you — from the data, not the post-mortems.
The segment-and-stack signature that keeps beating you — from the data
62% of losses: Datadog-installed
mid-market. 'Price' masked a fit gap.
| Pattern | Losses | Read |
|---|---|---|
| Datadog-installed MM | 24 | Displacement gap |
| Enterprise, no competitor | 6 | Lost on features |
| SMB price | 8 | Real price losses |
Overview
Analyze a pasted set of closed-lost deals by enriching each account with HG firmographic and technographic data to surface the structural patterns — which segment, which installed competitor, which stack signature — that correlate with losses, so the loss reasons go beyond anecdote.
Use cases
Loss reasons that survive scrutiny
Reps log 'price' because it's the easy answer. Enriching the lost accounts shows whether price losses cluster around a competitor or segment — usually revealing a fit or playbook gap that pricing can't fix.
Fix patterns, not anecdotes
One lost deal is a story. Thirty-eight lost deals with a shared stack signature is a pattern RevOps can route to enablement or product as a concrete gap.
View workflow prompt
# Lost Deal Pattern Analysis
## Parameters
- `{{lost_deals}}` *(required)* — Pasted closed-lost deals (deal, domain, segment, stated loss reason, competitor if known). Example: `Acme - acme.com - Mid-market - lost to Datadog - price`
- `{{category}}` *(optional)* — The category these deals competed in. Example: `observability`
## Purpose
Move beyond anecdotal loss reasons: enrich the {{lost_deals}} (competing in the {{category}}) with HG firmographic and technographic data to find the structural pattern — the segment, competitor footprint, or stack signature — that the losses share, so RevOps fixes a pattern, not a story.
## Process
1. **Parse** — read {{lost_deals}} with stated reasons and competitors.
2. **Enrich** — `company_firmographic` (segment, size, industry) and `company_technographic` (installed stack, competitor footprint) per lost domain.
3. **Find patterns** — cluster losses by firmographic segment, by installed competitor, and by stack signature. Compare the cluster frequencies to surface the dominant pattern.
4. **Validate against stated reasons** — check whether 'price' losses concentrate in a segment or competitor (often 'price' masks a fit or feature gap).
5. **Recommend** — name the highest-leverage pattern to address (e.g., 'losses concentrate in Datadog-installed mid-market accounts — that's a displacement-playbook gap, not a pricing problem').
## Output Format
Markdown with:
- `# Lost Deal Pattern Analysis`
- `## Loss Clusters` (table: pattern | # losses | example accounts)
- `## Stated vs. Structural` (where 'price' masks something else)
- `## The Dominant Pattern` (the one to fix)
- `## Citations`
## Quality Checklist
- Every cluster cites `company_firmographic`/`company_technographic`
- Stated loss reasons are tested against structural data, not taken at face value
- The dominant pattern is named with evidence, not asserted
- Sample size caveats are stated when clusters are small