# 12 of 30 ChatGPT answers (40%): How Cydel Giraudel runs The AI Receipts on daily AI routines

Cydel Giraudel is the solo founder of The AI Receipts, a site that shows which AI businesses make money using payment-verified revenue. More than ten scheduled AI routines run the business daily. They cover data refresh, SEO publishing, answer pages, ChatGPT citation checks, newsletter builds, sponsor outreach and follow-ups, Quora answers, funnel experiments and a daily KPI check-in. He sets strategy and approves anything that goes public. As of Nov 9, 2026, he reports that ChatGPT cited the site in 12 of 30 tracked buyer questions, up from 10 of 30 on Oct 30, and that a first sponsor signed up less than 90 minutes after his pitch landed.

**Try this step:** <strong>Make proof the product.</strong> A verified revenue leaderboard ranks 64 AI businesses. It is refreshed daily from Stripe data via TrustMRR. Each of the 77 case studies includes a "how we verified this" methodology and an evidence label: Verified, Public data, Self-reported or Estimate.


| | |
|---|---|
| **Who** | Cydel Giraudel, solo founder of The AI Receipts (theaireceipts.com), published by Bambana LLC. X @cydelmg and LinkedIn, linked from cydel.com, verified 2026-11-18 |
| **Problem** | Building trust and distribution for a new media and data product in a space "flooded with unverifiable income claims" |
| **Stack** | More than ten scheduled AI routines (AI tool or model not named), Astro on Vercel (serverless API routes, Blob storage, Speed Insights), Stripe subscriptions with trials, Beehiiv, Resend, TrustMRR revenue data, Satori/resvg chart images, Google Search Console API, IndexNow |
| **Confidence** | Medium (first-person, his own venture; figures are first-party and a snapshot as of Nov 9, 2026) |
| **Last verified** | 2026-11-18 |

## The problem
He says the "make money with AI" space "is flooded with unverifiable income claims, screenshots and guru content." He writes that founders, indie hackers and readers want to know what really works. He also writes that AI answer engines like ChatGPT "need trustworthy sources to cite." His answer was "to make proof the product."

## The blueprint (as he describes it)
1. **Make proof the product.** A verified revenue leaderboard ranks 64 AI businesses. It is refreshed daily from Stripe data via TrustMRR. Each of the 77 case studies includes a "how we verified this" methodology and an evidence label: Verified, Public data, Self-reported or Estimate.
2. **Hand the daily work to scheduled routines.** More than ten scheduled AI routines run the business daily. They cover data discovery and refresh, SEO publishing, answer pages, ChatGPT citation checks, newsletter builds, sponsor outreach and follow-ups, Quora answers, funnel experiments, and a daily KPI check-in.
3. **Publish for Google and answer engines.** A new post or refresh goes out every weekday. Posts carry Article, ItemList and FAQPage structured data and are pinged through IndexNow. The site has 65 blog posts and dozens of answer pages "built to answer high-intent questions in the exact format AI answer engines cite." The sitemap lists 276 URLs.
4. **Benchmark ChatGPT on fixed questions.** He tracks a fixed set of 30 buyer questions against ChatGPT. The first benchmark ran on Oct 30.
5. **Ship a newsletter three times a week.** A Beehiiv issue goes out every Monday, Wednesday and Friday. It includes "Receipt of the week", weekly movers, a sponsor slot and auto-generated charts.
6. **Pitch sponsors and let them buy without a call.** He reports ~195 personalized sponsor pitches plus ~78 follow-ups in under two weeks. They went out from a warmed, authenticated domain (SPF, DKIM and DMARC pass; mail-tester 9.3/10). The sponsor page has a live traffic panel, a $29/mo starting seat with the rate locked for 6 months, a self-serve claim form and Stripe checkout links.
7. **Work the smaller channels.** A daily verified-numbers Quora answer has run since Oct 17 (20+ answers). A Reddit launch wave ran Oct 26 to 29. He posts founder-led breaking-news and data posts on X. The site was submitted to 15+ AI tool and newsletter directories.
8. **Measure with his own analytics.** A privacy-light tracker records first-touch and session attribution and filters data-center traffic. A public live-stats API powers the sponsor page.
9. **Keep a human gate.** In his words: "Every number is verified before it ships, and nothing goes public without my approval."

## Stated results
All figures are as of Nov 9, 2026.
- **ChatGPT citations:** cited in 12 of 30 answers (40%), up from 10 of 30 (33%) on Oct 30. He says "ChatGPT readers are the stickiest traffic source."
- **Traffic:** ~3,000 monthly visitors in the first 3 weeks. Week-over-week visitor growth was +35% (775 vs 574). The record day was 154 visitors on Nov 6.
- **Search:** 2,442 Google impressions and 33 clicks in the first weeks, average position 10.4.
- **Sponsors:** the first outside sponsor (Structura) signed up less than 90 minutes after the pitch landed and went live the same day, Nov 9. A second inbound sponsor inquiry (Videco) came from outbound pitching.
- **Newsletter:** 8 issues sent, best open rate 37.5%, zero unsubscribes.
- **Reddit:** the launch wave brought ~95 visitors and the first organic newsletter signups.
- **X:** @cydelmg "passed 250K verified impressions in 90 days."
- Sponsor revenue and the reply rate on the ~195 pitches are not stated.

## Caveats
- This is a case study on his personal site about his own venture, published by Bambana LLC. He describes it as "not client work."
- The AI Receipts sells sponsorships, with a starting seat of $29/mo.
- Figures come from the product's own first-party analytics, Google Search Console, Beehiiv and his own ChatGPT benchmark. They are not independently verified.
- The 30 benchmark questions are his own fixed set. The source does not list them or describe how citations are checked.
- The source does not name the AI tool or model behind the routines. It does not describe prompts or how the routines are scheduled.
- The $2.82M combined MRR on the leaderboard belongs to the tracked businesses, not to The AI Receipts.
- The 250K X impressions cover 90 days, which is longer than the product's three weeks. The source does not say how much of that came from posts about The AI Receipts.
- Handles: X @cydelmg and LinkedIn in/cydel-micah-giraudel-0b9123134 (both verified: linked from cydel.com). No verified Bluesky, Threads, Instagram or TikTok profile was found.

## FAQ
### Which AI model runs Cydel Giraudel's routines?
The case study does not say. It describes more than ten scheduled AI routines but does not name the tool or model.

### Does anything publish without human review?
No, according to him. He writes that every number is verified before it ships and nothing goes public without his approval.

### How does he measure ChatGPT visibility?
He benchmarks a fixed set of 30 buyer questions against ChatGPT. Citations went from 10 of 30 on Oct 30 to 12 of 30 as of Nov 9, 2026. The source does not describe how each check is run.

### How much sponsor revenue has it made?
None stated. The source reports one sponsor live (Structura), one inbound inquiry (Videco) and a $29/mo starting seat, but no revenue total.

## Sources
- https://www.cydel.com/case-ai-receipts (primary: first-person founder case study on his own site, snapshot as of Nov 9, 2026)

---
*Want agents like these working your distribution? [Get It Built](/get-it-built). Get new teardowns weekly: [subscribe](/subscribe) (double opt-in).*