Bot-native income streams
Bid Scout / RFP scout (Experiment 1) is killed. Cold trust + competing with free/cheap BidNet-class alerts made Walter’s time EV bad. Do not run the archived RFP experiment. Next: doc-to-data Experiment 2 (CAREFUL).
CRO verdict: Best bot businesses are bounded event loops
(watch → enrich → decide → act → escalate), not content farms or vague
agencies. The RFP / tender scout ranked first on bot leverage and still
failed the sales test. Wholesale real estate is weaker on bot leverage
than the remaining ops loops.
Converted from ChatGPT Deep Research (CRO-approved structure). The raw Drive
export was not in content/_inbox/ for this PR — citations stay TBD.
Executive takeaways
- A bot earns its keep when the work is a repeatable event loop with a clear watch surface, a decision rule, an action, and a human escalate path.
- Top five (bot leverage): RFP scout (5), public-data signals (5), Google review ops (5), doc-to-data (4.5), inbound receptionist (4.5).
- Kill: faceless YouTube farms, SEO spam, KDP floods, a generic AI content agency, and a vague automation agency.
- Wholesale real estate can use a bot for research and memos. Phone and trust are still the business (~2/5). Do not treat it as the lead bot-native bet versus these loops.
- This month: Bid Scout is dead. Run one ops loop. Start with doc-to-data if a named warm partner exists; otherwise do not start.
Ranked ideas
Scores are CRO bot-leverage ratings from the Deep Research takeaway. No market sizes, conversion rates, or deal economics on this page.
| Idea | Bot leverage | Kind | Why it ranks |
|---|---|---|---|
| RFP / tender scout | 5 (killed) | Pre-award signal | Bot leverage was real. Killed: cold trust + BidNet-class firehoses. |
| Public-data signal feed | 5 | Pre-award signal | Same loop on permits, licenses, filings, and other public events. |
| Google review ops | 5 | Ops | New reviews are a bounded stream; draft, route, escalate. |
| Doc-to-data | 4.5 | Ops | Unstructured docs → schema → write-back; humans take low-confidence rows. Experiment 2 (CAREFUL). |
| Inbound receptionist | 4.5 | Ops | Missed calls / forms / chat: qualify, book or FAQ, hand off. |
| Wholesale real estate | ~2 | Relationship | Research is automatable; closing is not. |
Pages
This month
- Do not run the RFP / tender scout — Bid Scout is killed.
- Run doc-to-data only with one named warm partner.
- Do not start a content farm, SEO mill, KDP flood, generic agency, or a cold Document AI product.
Sources
- ChatGPT Deep Research export — TBD from Google Drive
Research/Inbox(expected local name:content/_inbox/2026-09-11-bot-native-income-streams.md). - CRO-approved page structure for this conversion (no invented numbers).
- Wholesale real estate — existing CRO takeaway (~2/5).
- Document-to-data — Experiment 2 deep-dive (CAREFUL).