Service Businesses Half-true — works only if you do the unspoken work

By the Income Reality Check Editorial Team · Updated 2026-07-11 · 7 min read

7 Claude AI side hustles that “pay more than a 9-5”: what the pitch leaves out

Verdict: Half-true — works only if you do the unspoken work. The seven business models are legitimate, but each one is a sales job wearing an AI costume.

Patrick Dang’s video “7 Claude AI Side Hustles That Pay More Than a 9-5” has racked up more than 57,000 views by promising beginners a path from zero to $10,000 a month “in a matter of weeks.” The seven ideas he lists are real. What the video compresses into a single sentence — finding and closing the clients who pay you — is the part that decides whether you earn $10,000 or $0.

What the video actually claims

Dang, who says he has 390,000 subscribers and seven years building online businesses, frames Claude as a once-in-a-lifetime opening. His seven “side hustles”: become an AI creator, an AI ghostwriter, a content-distribution operator, an AI-assisted UGC creator, an AI sales/lead-gen freelancer, an AI SEO (and “GEO/AEO”) specialist, and an AI ad producer. Each gets scored on ease of starting, speed to $10k/month, profit margin, and how little tech you need.

The proof is a run of client stories. Sandy, he says, knew nothing about Claude six months ago and now has 45,000 subscribers and makes “over $40,000 a month.” Brandon, a former construction employee, hit “$10,000 per month” in 30 days ghostwriting LinkedIn posts for construction firms. Another client, he says, cleared $10,000 a month producing UGC ads while still in college. The pricing guidance is consistent across all seven: monthly retainers from roughly $500 to $5,000 per client, with margins near 100% because “Claude does the work.”

To his credit, Dang is more honest than most. He openly calls other creators’ side-hustle ideas unrealistic, and he admits the AI-only sales-rep dream isn’t ready — a human still has to “finesse” the actual conversations. That candor is exactly why the missing piece stands out so sharply.

What the method actually requires

Here’s the thing every one of the seven ideas has in common: they are client-services businesses. You don’t earn until a real company agrees to pay you a retainer. Claude can draft the LinkedIn post, cut the clip, or build the target list — but it can’t make a stranger trust you with their brand and their budget.

So how do clients actually show up? For most beginners, through cold outreach, and the numbers there are humbling. Belkins’ 2025 analysis of thousands of B2B campaigns pegs a solid cold-email reply rate at 5–10%, with only top-quartile senders hitting 15–25%. Dang’s own instruction — “reach out to 50 companies” — implies maybe two to five replies, and a reply is not a booked call, and a booked call is not a signed retainer. To land three clients (Brandon’s number), you’re realistically contacting several hundred prospects, writing tailored samples, and running follow-up sequences for weeks. That’s the job. Claude makes the drafting faster; it doesn’t shorten the funnel.

The tools cost real money too, even if it’s modest. A working setup runs Claude Pro at $20/month or Max at $100–$200/month per Anthropic’s pricing page, plus outreach software, a scheduling tool, and often editing subscriptions. Margins are high, but “close to zero cost” isn’t accurate once you’re sending email at volume.

Then there’s the regulatory backdrop for anyone selling these income promises. In January 2025 the U.S. Federal Trade Commission proposed expanding its Business Opportunity Rule to cover “money-making opportunities” like business coaching, and would require sellers to hold written substantiation for any earnings claim they make. That’s directly relevant here: Dang’s video routes viewers to a paid coaching program and a “book a call” funnel. The FTC has teeth on this — in October 2024 it settled with Lyft for $2.1 million over deceptive claims about how much drivers could earn. (U.S. readers: none of this makes Dang’s program illegitimate; it means “clients make $10k in weeks” is exactly the kind of statement a regulator now expects to see backed by typical, not peak, results.)

Who actually wins this game

Look closely at the success stories and a pattern emerges. Sandy’s breakout wasn’t a beginner’s fluke — she landed her first client on Upwork by sending a Loom video, then paid someone else roughly $1,000 to build the delivery system while she managed it. Brandon picked a niche he already understood (construction). Dang himself is selling coaching to an audience of 390,000 he spent seven years building. The distribution — the audience, the network, the ability to get a stranger on a call — came first. The AI came second.

Who wins, then? People who already have a sellable skill or a warm network, early movers in genuinely new lanes like AI SEO, and operators willing to treat client acquisition as the actual product. The video’s own framing quietly confirms this: Dang repeatedly says the hardest part isn’t building anything — it’s distribution, “getting people to even want to try it.” That’s the whole ballgame, and it’s where the beginner is least equipped.

What you’d realistically earn

The $10,000-in-30-days stories are real for the people named. They are also the visible tip of a much wider distribution. U.S. Bureau of Labor Statistics data on alternative work arrangements found independent contractors earned median weekly pay of $949 in 2023 — below the $1,132 median for full-time traditional employees. Freelancing, on average, does not beat a 9-5; a minority of freelancers clear well above it.

The creator idea faces the same math. As CNBC has reported, income for social-media creators is volatile and, for most, modest — a large share earn a few thousand dollars a year, not a month, and full-time creator income typically lands well below traditional employment. Sandy’s $40k/month exists. It is not the median outcome for someone who started six months ago.

A grounded expectation for a beginner doing this seriously: roughly $0 for the first one to three months while you build samples and outreach, a first client somewhere in the $500–$2,000/month range if your pitch lands, and a plausible climb to $3,000–$8,000/month after a year of consistent selling — if you’re good at sales and you stick with it. That’s a genuinely good outcome. It’s just not “this weekend.”

Who this is (and isn’t) for

This works if you can spare 10–15 hours a week, you’re comfortable pitching strangers and hearing “no” fifty times, and you have a skill or niche you already understand well enough to sound credible on a call. A few hundred dollars for tools and a tolerance for a slow first quarter help. It does not work if you expected Claude to replace the selling — if the phrase “reach out to 50 companies” made you wince, the mismatch is the point, and this is the wrong path.

What to remember

Every one of these seven ideas is a real way to make money, and Dang is more grounded than most creators in this niche. But the label “side hustle” hides a service business, and the engine of a service business is client acquisition — the one task Claude can assist with and never do for you. Judge the opportunity by how you feel about sending the hundredth cold message, not the first AI-drafted paragraph.

For related breakdowns, see our reviews of 5 Claude AI side hustles that “pay more than a 9-5” and 10 Claude AI side hustles that can pay a full-time income.