Sample and preview

Review a sample opportunity map before you share data.

Open a sample row, inspect the scoring logic, and decide whether the output is worth connecting your own sources.

Sample deliverable

Fictional fixture data — your map uses the same format with your uploads. Select any row to inspect scoring components.

Opportunity map · fictional sample Not your graph

Fictional sample only. Your map is built from sources you upload after reserving.

#	Account	Intro path	Source evidence	Score factors	Conf.
1	Northwind Analytics	You → Alex Kim (customer of)	Target account tier 1 improves fit.; Explicit target-account flag from upload.; Strongest relationsh	Relationship strength 0.61Account fit 0.90Timing 0.90Actionability 0.95	0.82
2	Harbor Metrics	You → Sam Chen (met with)	Target account tier 2 improves fit.; Explicit target-account flag from upload.; Strongest relationsh	Relationship strength 0.60Account fit 0.83Timing 0.63Actionability 0.85	0.82
3	Quartz Security	You → Reese Walker (customer of)	Target account tier 1 improves fit.; Explicit target-account flag from upload.; Strongest relationsh	Relationship strength 0.58Account fit 0.90Timing 0.51Actionability 0.95	0.82

Row inspector

Northwind Analytics

Confidence 0.82

Relationship strength: 0.61
Account fit: 0.90
Timing / relevance: 0.90
Actionability: 0.95
Confidence: 0.82
Multi-source corroboration bonus: 0.20
Single-source penalty: 0.00
Weak-relationship penalty: 0.00

Evidence Target account tier 1 improves fit.; Explicit target-account flag from upload.; Strongest relationship: customer of (strength 0.61).; Timing signals present: active customer, crm deal stage, product u

Inspect score — #1 Northwind Analytics (confidence 0.82)

Relationship strength: 0.61
Account fit: 0.90
Timing / relevance: 0.90
Actionability: 0.95
Confidence: 0.82
Multi-source corroboration bonus: 0.20
Single-source penalty: 0.00
Weak-relationship penalty: 0.00

Source evidence: Target account tier 1 improves fit.; Explicit target-account flag from upload.; Strongest relationship: customer of (strength 0.61).; Timing signals present: active customer, crm deal stage, product usage, target account flag.; Corroborated across 8 source types.; Multi-source bo

Preserved source record IDs: 0056d6de-400, 0461f291-5d8, 2e470fb2-512, 3cf0f101-e37, 51bcb396-e93 +8 more — evidence kept per upload, not regenerated text.

Fictional fixture row — scoring logic is real; inputs are sample data.

Fictional fixture row

Proof you can verify before you pay

Inspect a sample opportunity map and score breakdown.
Use upload fallback or public preview before you reserve.
Review connector permissions and data boundaries before sharing sources.

The paid build is a working opportunity map, not a report

The buyer-facing artifact is the same shape as the preview: ranked rows, source evidence, confidence gaps, intro asks, and a manual review checklist.

CSV export 25 ranked warm opportunities with intro path, why-now context, confidence, and missing-data notes.

Inspectable brief Readable HTML/Markdown summary that explains the highest-signal accounts and the evidence behind them.

Review checklist Verification steps before outreach: confirm titles, check source recency, and adapt the suggested intro ask.

2 business days Service-style fulfillment after validated intake and sources, with published refund terms.

Why this is more than pasting exports into Claude

WarmPath keeps source rows, penalties, and corroboration attached to each account. More sources improve coverage instead of creating another one-off prompt transcript.

DIY prompt path Paste CSVs, ask for rankings, then manually recreate source IDs, confidence logic, gaps, and repeatable export format.

WarmPath workflow Connect or upload sources progressively, preserve evidence per row, inspect score components, then export the map.

Upload fallback: preview WarmPath on sample files

If you are not ready to connect a source, upload contacts CSV and target accounts CSV (required minimum). We build a ranked mini-preview from your relationship data — locally in test mode, no deposit, no credentials, no bought lists.

Fallback preview, not final deliverable. Max 3 files, 500 rows per CSV. Optional CRM, LinkedIn export, or calendar ICS improve corroboration.

After preview: inspect the sample deliverable · public-data preview (secondary) · reserve full build (test mode, after preview).

Or try a public-data preview (no uploads)

Enter your website, ICP, and offer. We generate a small ranked preview from deterministic public-style fixture data — no deposit, no web scraping, no paid APIs. Reserve the full build only if the preview looks useful.

Preview, not final deliverable. No promised meetings, replies, or revenue.

Secondary path — uses deterministic fixture data. Prefer connecting one source or the upload fallback for your relationship data.

Inspect how WarmPath scores

Rules, evidence structure, and honest category comparison — same inspector view as the deliverable.

How WarmPath scores opportunities

Every ranked row is a weighted sum of inspectable components — not a black-box “AI score.” Expand any preview row to see the numbers and missing-data notes.

Multiple independent sources beat one. Corroboration from CRM, calendar, contacts, and target-account uploads adds a bonus; a single source applies a penalty.
Warm relationship path beats generic fit. A real connector (customer-of, met-with, LinkedIn export) outweighs “this account looks like your ICP” alone.
Missing data lowers confidence. Weak relationships, missing domains, or single-source rows get explicit missing-data notes — not hidden.
Preview is bounded. Free preview uses public-style fixture data; your full build uses your uploads with the same scoring logic.

What you get beyond “paste into Claude”

A one-shot Claude prompt can draft ranked accounts and intro wording. WarmPath adds structure you can verify before you act — without promising outcomes we cannot prove.

Output property	Claude + exports (DIY)	WarmPath structured map
Ranked warm paths	Prose list or table — regenerated each prompt; hard to diff across runs	Fixed CSV + brief rows with stable rank, intro path, and confidence per account
Source evidence	Paraphrased from pasted exports — no preserved record IDs	Contributing source types + preserved source record IDs per row
Multi-source corroboration	Depends on prompt quality; no explicit bonus/penalty rules	Inspectable bonuses and single-source penalties in score breakdown
Missing data	Often smoothed into confident-sounding copy	Missing-data notes and confidence penalties listed on the row
Score transparency	Single opaque “fit score” unless you re-prompt for rubric	Expandable component scores: relationship, fit, timing, actionability, corroboration

Preserved source record IDs — each row links back to specific upload rows, not paraphrased memory.
Cross-source corroboration — bonuses and penalties when only one source supports a path.
Missing-data penalties — gaps are listed on the row, not smoothed away.
Inspectable confidence — component scores you can challenge; not a single opaque percentage.

This does not guarantee better meetings or replies — it makes the map auditable so you decide what to trust.

How WarmPath compares — honestly

Preview-first, inspectable relationship-evidence map for founder-led outbound prep — not a full GTM platform, enterprise team graph, or live integration suite.

Wedge: ranked warm paths with preserved source evidence, cross-source scoring, missing-data penalties, and free preview before deposit — one buyer's uploads per build.

Category	Examples	Where they excel	WarmPath scope
GTM data & orchestration	Clay, Common Room	Enrichment marketplaces, signal platforms, AI agents, CRM sync, outbound orchestration at team scale	Narrow: relationship-evidence map from your exports — ranked paths and inspectable scoring, not a 150-provider waterfall
Live team relationship graphs	Connect The Dots (CTD), Affinity, Introhive	Auto email/calendar capture, firm-wide graph merge, intro orchestration, MCP/API, enterprise proof	Earlier & manual: preview-first upload/export, sandbox OAuth only, single-buyer deliverable — not a live team graph
DIY Claude + LinkedIn	Founder workflow	Fast $0 brainstorm — export connections + CRM → ranked ideas and draft asks	Structured evidence: source record IDs, corroboration bonuses/penalties, missing-data notes, expandable component scores

What we do not claim

No enterprise or team-wide graph — one buyer's sources per build.
No production live email, calendar, or CRM integrations — upload/export and sandbox OAuth only.
No customer logos or “beats Clay/CTD” superiority — category comparison only.
No outbound sending or promised meetings — you review and act from your own tools.

Grounded in docs/market-analysis/2026-06-26-warmpath-competitive-teardown.md.