How much it costs us to process one property-owner lead end-to-end (LLM + Firecrawl scrapes), and what that means next to our $0.50 / line price.
Not a live dashboard. Numbers from real lab runs and production-shaped measurements. Not an invoice. PII stripped.
vs list price $0.50 / line → large gross margin on the marginal lead (fixed “shelf” costs separate below).
Product shape: batch job (upload list → process async → notify).
Concurrency today: 4 leads in parallel on the Cloud Run worker.
Not “answer in 2 seconds” — that would be a future tier with paid people-data APIs.
Firecrawl does not bill like ChatGPT tokens. We pay credits per scrape page.
| Call type | Credits | Notes |
|---|---|---|
| Sunbiz search / detail | 1 | Corporate registry |
| TPS name / address / detail (stealth) | 5 | Cloudflare-protected; retries cost again |
| Cache hit (lab only) | 0 | Production worker runs without disk cache between jobs |
Failed stealth attempts bill the same as success.
| Scenario | Credits / lead | When |
|---|---|---|
| Clean individual | ~10–20 | Address + detail, few retries |
| Clean entity | ~17–27 | Sunbiz pierce + TPS on officer |
| Operating average | ~20–24 | COGS table + FTL batch (real API delta) |
| Hard / partial / flail | 40–70+ | Bot blocks, thrash, weak names |
Lab gold-set (15 leads, mixed): ~18 credits / lead average (balance delta).
Example at 24 credits / lead:
| Plan (indicative) | ≈ $ / credit | ≈ $ FC / lead @ 24 cr |
|---|---|---|
| Hobby | $0.0032 | ~$0.08 |
| Standard (ops base) | $0.00083 | ~$0.02 |
| Growth | $0.00067 | ~$0.016 |
Takeaway: the plan tier moves Firecrawl $ more than the LLM does. Standard is the sensible default for production volume.
Same motor, different Camada 2 models:
| Model | ≈ LLM $ / lead | Notes |
|---|---|---|
| Gemini 2.5 Flash (Vertex) | ~$0.009–0.011 | Best cost; lab recommendation for cost-mode |
| Claude Haiku 4.5 | ~$0.02 | Fine quality; not on Vertex |
| Gemini 2.5 Pro | ~$0.035 | Overkill for cost |
| Gemini 3.5 Flash | ~$0.05–0.06 | More thrash / worse economics in lab |
| Claude Sonnet 4.6 (historical) | ~$0.055–0.058 | Gold quality baseline |
List prices used for Flash (Vertex): ~$0.30 / M input · $2.50 / M output (thinking billed as output).
PROD (as of 2026-07-30 flip): Camada 2 on Vertex · Gemini 2.5 Flash with guardrails (anti–registered-agent, richer Sunbiz paths, short-name address-first). Grok remains available as rollback if needed.
Method: 15 leads from the original Sonnet high-confidence enriched set (ground truth), scored name match OR shared phone.
| Result | Number |
|---|---|
| First pass (Flash, before guardrails) | 11 / 15 (73%) |
| 4 hard fails investigated | RA-vs-officer, incomplete path, scrape flakiness, 1 ambiguous gold |
| After code guardrails · retest those 4 | 3 / 4 recovered |
| Remaining disagreement | 1 lead where current Sunbiz managers ≠ namesake heuristic in gold |
Projected if we re-score the full 15 with guards: ~14 / 15 (93%) — meets the prior quality bar if that one gold row is treated as ambiguous or re-labeled.
Gold = prior Sonnet high-confidence set; one lead’s ground truth may lag current Sunbiz filings.
Guardrails shipped: never prefer Registered Agent / attorney over PT/MGR; force richer Sunbiz detail paths; short owner names go address-first.
| Line | Amount |
|---|---|
| Price (current commercial) | $0.50 / line |
| LLM (Flash lab) | ~$0.01 |
| Firecrawl @ 20–24 cr, Standard | ~$0.017–0.020 |
| COGS / line (variable) | ~$0.03 |
| Gross margin on variable COGS | ~94% |
Shelf / fixed (order of magnitude): ~$85–90 / month (Firecrawl Standard + light infra). Break-even is “a few hundred lines / month,” not thousands — exact break-even depends on plan mix and hit rate; this page is about per-lead variable cost.
The expensive leads are usually the bad ones (bot blocks, incomplete names, thrash). Without a brake, a rotten upload burns Firecrawl and still returns poor results. Goal: if quality will be bad, fail cheaper.
In production now:
not_found with 0 Firecrawl.SKIPTRACE_LEAD_FC_BUDGET=28) — thrash stops mid-lead as partial (not 50–70 cr).TPS_TRIES=2.Lab proof (4 bad leads, parallel, budget 20):
BREEN W → enriched under budget (~15 cr)LeadFcBudgetExceeded partial (~20–25 cr), not open-ended flailNext layers: job-level credit ceiling · two-pass lean→full · product copy “garbage → cheap not_found”.