Status: idea candidate, not evaluated, not built Captured: 2026-08-05 by Charlie Origin: a ChatGPT thinking session Annette ran (undated). She surfaced the PDF export on 2026-08-05 after Charlie found zero record of the idea anywhere in her project files, memory, or Claude Code session history. Source PDF: C:\Users\annet\Downloads\download (2).pdf. This file is the record. Do not let it live only in a Downloads folder again.

---

The concept

A Fiverr-style two-sided marketplace that pairs inexpensive personal trainers and workout buddies in lower-cost countries (India, Mexico, the Philippines) with clients in the US, UK, and other high-income countries. Live virtual sessions, not content.

Decisions Annette already made

Decisions made 2026-08-05, after the capture

These override anything earlier in this file and any recommendation Charlie made:

• She is building it. Her words: "I don't care if it doesn't make money. I'm definitely building it." Profitability is not the deciding test here. Do not re-argue the business case, do not open with revenue risk, and do not treat this as an open question. It is settled.
• No manual concierge test first. Charlie recommended hand-matching 10 clients before building any software. She declined. Going straight to building.
• Supply comes from existing hiring marketplaces, Fiverr first, then other low-income-country hiring sites. Fiverr doubles as her market check: she looked, and the fitness supply is thin.

What this means for how to help her: the useful work is de-risking and sequencing, not evaluation. The question is no longer "should this exist," it is "what kills this, in what order, and what does she do about each one."

Open questions

1. Payments and KYC. Paying individuals in India, Mexico, and the Philippines makes her a payment facilitator. Stripe Connect handles it, but per-provider onboarding and India's rules are real friction. Not discussed in the ChatGPT thread at all.
2. Liability. Uncertified people leading exercise for retirees and GLP-1 patients with chronic conditions. Waivers, scope-of-practice limits, insurance. Not discussed.
3. Competition. Not researched. Fiverr already has fitness categories.

Charlie's read (2026-08-05)

The strongest thing here is the GLP-1 angle. Millions of people are losing weight fast on these drugs and losing muscle with it, and almost nobody is delivering cheap, live, human accountability into that gap. Annette is the customer: she's on Mounjaro, she's 57, she writes about protein and strength for exactly this audience, and she already owns the publications (SmartStrongAlive, ProteinFirstRecipes) that reach them. Demand side is her actual strength.

The weakest thing is that it's a two-sided marketplace, which is the hardest business model there is. Both sides have to show up before either side has a reason to. She has real revenue urgency and 80 project folders already, so a cold-start marketplace build is the wrong shape of bet right now.

Existing assets that matter: TrainerBooked and PersonalTrainerPlaybook already give her trainer-side relationships and credibility. That's a genuine head start on supply.

Recommended next step (not yet approved)

Skip the platform. Run it by hand first, the "do-it-first" way she likes: recruit 5 providers manually, hand-match 10 clients pulled from her own newsletter list, take payment with a plain Stripe link, deliver over WhatsApp, and see whether people book a second session. If matching works and retention is real, then software is worth building. If it doesn't, she's out a few weeks instead of a year and a Sharetribe subscription.

Tool claims from the ChatGPT thread that do NOT hold up

Verified by web search 2026-08-05. Flagging because she would otherwise act on them:

• "Kreezly" (ChatGPT's #2 pick and one of its two "start here" recommendations): no evidence it exists. It does not appear in any 2026 roundup of freelance marketplace platforms. Treat as fabricated until she sees a working product herself.
• "Sharetribe Flex or Sharetribe Go": stale. Flex and Go merged into a single product simply called Sharetribe in October 2023. Current pricing runs roughly $39 to $299/month plus per-transaction fees, and real marketplaces report far higher all-in monthly costs once Stripe Connect and add-ons land.
• The rest of the list (Bubble, Webflow + Memberstack, AppGyver + Xano) was not verified. AppGyver in particular changed hands and should be checked before anyone relies on it.

Consistent with her standing rule that outside AI drafts get checked before they're trusted.