Watch on YouTube: https://www.youtube.com/watch?v=yO5O_ngyA_8
Channel: Codie Sanchez
Testing Five AI Automation Bots With a $100 Budget: What Actually Worked
Codie Sanchez recently ran a 10-day experiment using Grok Bot (also referenced as Grockbot, Crockbot, or Rockbot) to test five different AI-driven money-making ideas. Each bot started with a $100 budget and a $1,000 revenue target. The results were mixed: two bots generated meaningful net returns, one lost money, one produced zero cash revenue but saved internal costs, and one delivered a modest profit. The exercise highlights a recurring theme in AI automation—automating a flawed process only accelerates losses.
What the Video Covers
The experiment spans five distinct use cases:
- Wolfie – A trading bot that scraped Reddit and X/Twitter for stock tips and executed trades via Robinhood.
- Alvin, Simon, and Theodore – A trio of bots that flipped free items from Facebook Marketplace, using Uber for logistics.
- Clipping Bot – An outreach bot that searched Upwork and Fiverr for video editing gigs, leveraging ElevenLabs for voice work.
- Finn – A re-engagement bot that built a Shopify landing page and Klaviyo email sequences to resell a book to past buyers.
- Dora – A cold lead generation bot that used Hunter.io, Apollo, HubSpot, Google, and LinkedIn to book calls for an Austin workshop.
Only Finn and Dora produced measurable net gains within the 10-day window. Wolfie lost money, the clipping bot secured no paying clients, and the reselling bots turned a small profit but required physical logistics.
Step-by-Step Walkthrough of Each Bot
1. Wolfie – The Trading Bot
Wolfie monitored social platforms for trending tickers and placed trades automatically through Robinhood. The premise was to capture retail sentiment signals in real time. In practice, the bot executed losing trades and was shut down. The takeaway: signal quality from public forums is noisy, and execution speed alone doesn’t compensate for poor alpha.
2. Alvin, Simon, and Theodore – Free-Item Flipping
These bots scanned Facebook Marketplace for free listings, arranged pickups via Uber, and relisted items for sale. Over 10 days they generated $130 in revenue. Cost of goods sold (COGS) came to $40.98, Grok Bot usage added $1, for a total cost of $41.98. Net profit: approximately $88. The model works but is labor-adjacent—physical pickup and listing effort limit scale.
3. Clipping Bot – Service Arbitrage
The bot applied to video editing jobs on Upwork (using a $21 premium account) and Fiverr, proposing AI-assisted clipping with ElevenLabs voiceovers. Zero paying leads materialized. However, when used internally to edit the channel’s own videos, it saved an estimated $50–$250 per video—roughly $600 per month in avoided freelance costs. The lesson: validate the service on your own workload before selling it.
4. Finn – Book Re-Engagement Bot
Finn targeted existing customers who had purchased a previous book. It built a Shopify landing page and a Klaviyo email sequence to cross-sell a new title. Results: $54.48 in revenue, $24 COGS, $17 for Klaviyo and Grok Bot usage, total cost $41, net ~$14.48. Modest in absolute terms, but it illustrates the “cheapest customer is the one you already have” principle.
5. Dora – Cold Lead Generation Bot
Dora sourced leads for a paid Austin workshop. Using Hunter.io and Apollo for verification, it warmed four inboxes from 50 to 200 emails per day. Outcome: 10 leads, 2 booked calls. Valued at $100 per response and $500 per booked call (based on Meta cost-per-lead benchmarks), the pipeline was worth ~$2,000. Hard costs: $37.26 COGS, $20 Grok Bot usage, total $57.26. Net value ~$1,943. This was the highest-leverage bot, though the $2,000 figure represents internal valuation, not cash collected.
Tools and Numbers at a Glance
- Platform: Grok Bot (Grockbot/Crockbot/Rockbot)
- Budget: $100 per bot, 10-day sprint, $1,000 target
- Trading: Robinhood, Reddit, X/Twitter
- Reselling: Facebook Marketplace, Uber
- Clipping: Upwork ($21 premium), Fiverr, ElevenLabs
- Re-engagement: Shopify, Klaviyo
- Cold outreach: Hunter.io, Apollo, HubSpot, Google, LinkedIn
- Email scaling: 50 → 200 emails/day across 4 inboxes
Caveats and Hype Check
If you point a bot at a bad process, you just lose money faster.
The cheapest customer you’ll ever get is the one you already have.
Use it yourself before you sell it to somebody else.
The $100-to-$1,000 headline is only partially realized. The ~$2,000 cold-lead figure is an internal valuation based on advertising cost equivalents, not recognized revenue. The trading bot lost money outright. The clipping bot earned $0 in external sales. The reselling bots netted $88 but require physical logistics. The re-engagement bot netted ~$14. The experiment’s real value is the framework: test internally, measure net economics, then decide whether to productize.
The video concludes with a call to action—comment “bot” to receive the Grok Bot templates—which functions as list-building for the channel. Treat that as a lead capture mechanism, not a guarantee of plug-and-play profitability.
Concrete Next Actions
- Audit your business for dead quotes, stale lead lists, or past buyers. Build a segmented re-engagement bot (Shopify + Klaviyo or equivalent) to test cross-sell offers.
- Run the clipping workflow on your own video library first. Quantify time and cost savings before pitching the service to clients.
- If you have recurring inventory or parts purchases, prototype a supplier price-negotiation bot that monitors vendor catalogs and triggers reorders at threshold prices.
- For cold outreach, verify every email with Hunter.io (or similar), warm multiple inboxes gradually (50 → 200/day), and track cost-per-booked-call against your customer acquisition cost benchmarks.
- Request the Grok Bot templates via the video’s comment/CTA if you want a starter scaffold, but plan to customize heavily for your specific process.
Bottom line: AI agents amplify whatever process they’re pointed at. Validate the economics manually first, then automate the winner.