@angus.sewell
Instagram · Greenscreen Talking Head
@angus.sewell Sometimes it’s nice to not try and scale something ngl #ai #agency #startup
Trending short videosCreator ranks four AI business models
The edit1 shot · 5 text
Shots Text1:03
Performance
- Views
- 147.6K
- Likes
- 3.8K
- Comments
- 129
- Engagement
- 2.7%
- Outlier
- 3.6×
- Length · 21 May 2026
- 1:03
Video #30326 · Greenscreen Talking Head
Creator ranks four AI business modelsfrom @angus.sewell
[01]
Overview
The creator critiques four different AI business models, starting with the most popular and, in his opinion, the worst: the AI agency. He then evaluates AI consulting and AI software, pointing out their respective challenges, before concluding with the 'Fractional AI Partner' model, which he has found personal success with by embedding himself as a part-time expert in multiple companies.
[02]
Simple outline
- 1State contrarian thesis
- 2Debunk common approaches
- 3Confess personal failure
- 4Reveal successful alternative
[03]
Related videos
110.8K2.1×Comparing expensive AI tools to DIY@angus.sewell · Breakdown
111.7K3.8×Testing AI to build business apps@angus.sewell · Rapid Fire Listicle
79.2KSimple breakdown of AI agents@itsmariahbrunner · Opportunity Explainer
644.1K100.3×AI problem and council solution@jasoncooperson · Breakdown
309.9K15.2×Reverse-engineer a coach's business model@angus.sewell · Tutorial / How To
2.9M29.1×Rapid fire AI tool ranking@rony · Rapid Fire Listicle
284K11.5×AI sales tool teardown tutorial@angus.sewell · Tutorial / How To
336.3K17.7×AI playbook for buying businesses@angus.sewell · Playbook
222.1K6.7×Debunking AI trading bot hype@angus.sewell · Breakdown
206.8K3.3×Comparing good vs bad AI marketing@shwinnabegobrand · Good vs Bad
389K3.1×Split screen AI news breakdown@rony · Case Study Breakdown
530.1KExplaining AI tool with animations@opusjake · Opportunity Explainer[04]
Classification
Format
Concepts
Elements
Account
[05]
Hook analysis
What stops the scroll in the first 3 seconds
Title hook · 55%
$28k for 16 days
Verbal hook · 40%
Bro, it turns out the worst AI business model to start is kinda also the most popular one.
Visual hook · 5%
A picture-in-picture layout with the creator's face in a small box overlaid on a stock photo of a corporate meeting.
[06]
Hook template
Steal this for your own video
The most popular is actually the worst one. Here's what made me $ in .
- — The most popular dropshipping model is actually the worst one. Here's what made me $14k in 3 weeks.
- — The most popular real estate investing strategy is actually the worst one. Here's what made me $52k in 30 days.
- — The most popular freelance pricing model is actually the worst one. Here's what made me $9k in 11 days.
[07]
Hook and payoff
The promise the opener makes, and how the video pays it back
Hook
A contrarian statement that pits a popular belief against a negative outcome ('the worst... is also the most popular'), creating immediate intrigue and a knowledge gap.
Payoff
The final beat provides a clear, actionable solution backed by a relatable success story, resolving the initial problem posed by the hook and rewarding the viewer for watching through the failures.
[08]
Narrative framework
The structural pattern this video runs on
Framework
The Personal Failure Gauntlet
Systematically debunking popular approaches to a problem by using personal failure as proof, before revealing a lesser-known but successful alternative. The list is presented out of numerical order to build a narrative arc.
- 1
The 'Good on Paper' Trap
Introduce and critique the first common approach, highlighting a theoretical flaw that prevents success in practice (e.g., a great plan that nobody implements).
- 2
The 'Fun vs. Reality' Conflict
Introduce and critique a second common approach, contrasting its appealing, creative aspects with its difficult, unglamorous realities (e.g., building is fun, selling is hard).
- 3
The Personal Failure Confession
Identify the most popular approach (as promised in the hook) and provide a detailed, vulnerable account of personal failure with it, solidifying credibility and emotional connection.
- 4
The Winning Formula Reveal
Present the final, recommended approach as the solution. Explain its mechanics, why it overcomes the flaws of the others, and provide personal proof of its success.
Themes
Techniques
Trends
Post caption
Sometimes it’s nice to not try and scale something ngl #ai #agency #startup
Text overlays
title · beginning
$28k for 16 days
informational · beginning
BUSINESS MODEL - #2 AI CONSULTING REVENUE One-time engagement fees CYCLE Discover -> Roadmap -> Present -> Handoff CLIENTS Enterprise & large mid-market YOU SELL THE STRATEGY. THEY DECIDE IF IT GETS BUILT.
informational · middle
BUSINESS MODEL - #3 AI SOFTWARE REVENUE Monthly subscriptions (SaaS) CYCLE Build -> Launch -> Acquire -> Iterate CLIENTS Niche vertical end-users RECURRING REVENUE. UNTIL A FUNDED COMPETITOR SHIPS YOUR FEATURE.
informational · middle
BUSINESS MODEL - #1 AI AGENCY REVENUE Project fees per engagement CYCLE Brief -> Build -> Deliver -> Repeat CLIENTS SMB & mid-market YOU GET PAID PER PROJECT. NOT PER OUTCOME.
informational · middle
BUSINESS MODEL - #4 FRACTIONAL AI PARTNER REVENUE Monthly retainer per client CYCLE Onboard once -> Deliver -> Compound CLIENTS 4-6 companies simultaneously YOU SKIP MONTHS 1 AND 2 FOR EVERY SINGLE CLIENT.
Transcript
Bro, it turns out the worst AI business model to start is kinda also the most popular one. This one, on the other hand, not terrible. You land, like, a $20,000 engagement. You deliver a good map. Nobody really implements it at all. Starting an AI software is super fun when you're building it. Then you have to go and sell the software and get people to use it. Oh, there's the hard part. This is the popular one that I was referring to. People usually start this and see how easy it is to close deals because everybody does want AI. Then they usually miserably fail to fulfill, which I've certainly been in that boat, my guy. The hard part is, like, understanding a client's context, especially if you've, like, never worked in their industry. Maybe this is like a skill issue on my part. Bro, I was so bad at running an agency. I would rip deals all day, bro. But then I would share to everybody after like a month. This is the first model where I've actually genuinely been success. Basically, you're like a part time employee for a bunch of different companies. You charge each of them, I don't know, anywhere between 8 to, like, $15 a month. And the thing is I get to, like, really embed myself and super learn about how their companies work. And I'm actually not fucking this one up, guys. Like, I'm not churning clients at all. Like, yeah. I got to capacity in, like, a week and a half, but it feels a lot better to not, like, be terrible at your job.