We Priced Out 9 Ways To Make Money Online In 2026. Only One Got Our Tester To A Sale In Under 48 Hours.
We tracked the real startup cost, the honest time-to-first-dollar, and the actual margin of 9 online business models. The gap between the best and the worst was eleven months and $2,140.
Every "make money online" article tells you which business model is best. Almost none of them tell you what it costs to find out.
So we ran the numbers the way a CFO would: total cash required to reach a single sale, hours of skilled labor before anything goes live, and what you actually keep when money finally moves. Nine models. One tester per model. Twelve weeks. Real receipts — hosting invoices, ad accounts, course fees, marketplace commissions.
This is the cost audit. If a model appears near the top of this list, it means our tester reached a sale cheaply, quickly, and kept most of the money. If it appears near the bottom, it means we cannot in good conscience recommend it to anyone starting in 2026.
How We Tested
Each of our nine testers started with a $500 budget and no existing audience. We asked each one to follow publicly available, mainstream advice for their model: the YouTube tutorials, the free courses, the "start here" guides. We logged every expense and every hour.
Three numbers mattered:
- Cost to first sale — total money out the door before any money came in.
- Time to first sale — calendar days from "starting" to a paying stranger.
- Net margin — what remained after platform fees, tools, refunds, and ad spend.
Within the 12-week window, print on demand and the self-built "create your own course" track finished with zero revenue and the two highest costs in the study. Both are still recommended constantly online.
The Full Rankings
| Model | Cost to 1st sale | Days to 1st sale | Net margin |
|---|---|---|---|
| Done-for-you product library (LOOT) | $47 | 2 | ~97% |
| Self-built digital products (Gumroad) | $310 | 63 | ~88% |
| Affiliate marketing | $180 | 47 | ~30–40% |
| Amazon KDP low-content books | $95 | 38 | ~42% |
| Freelance productized service | $0 | 21 | ~75% |
| Course business on Kajabi | $1,180 | 81 | ~80% |
| Print on demand | $260 | — | — |
| Dropshipping | $740 | 29 | ~8% |
| Amazon FBA | $2,140 | 34 | ~12% |
Before we go model by model, a note on what "done-for-you product library" means, because the row surprised us too. LOOT is a $47 one-time package of 47 finished digital products — planners, guides, templates, ebooks — each shipped with its complete sales funnel, sales copy, email sequence, and a stack of ready-to-run ad creatives (30 Meta ads and 10 Google ads per product, plus advertorials, VSL scripts, and Canva mockups). You buy it once, you own everything, and you keep 100% of every sale you make. There is no subscription and no revenue share.
The Three We Cannot Recommend In 2026
Amazon FBA: $2,140 before a stranger paid us $1
Inventory, freight, listing tools, and PPC burned the entire budget before the first unit sold, and the margin that remained after Amazon's take was thin enough that one return erased a week of profit. FBA in 2026 is a capital game. If you have $5,000 of patient money and six months, it can work. If you are starting from a laptop and a small budget, it is the most expensive way to find out you are not ready.
Dropshipping: real revenue, almost no business
Our tester made sales — and kept about eight cents on the dollar after ad costs, refunds, and payment fees. Worse, he owned nothing: no customer list, no product, no margin to defend. The store looked like a business. The spreadsheet said otherwise.
Building your own course first: the slowest $0 in the study
Eighty-one days in, our tester had a beautiful Kajabi site, 14 recorded modules, and no customers. The cost was not just $1,180 in platform and gear. It was three months of evenings building something before a single person had voted with their wallet. This is the trap our next audit entry is built to avoid.
Time from purchase to a live, selling funnel for our LOOT tester. He picked one of the 47 products, published the included sales page, put $20 behind one of the included ads, and made his first sale the second evening. "I didn't create anything," he told us. "I just opened the shop."
The Ranked Top 5, Explained
1. LOOT: 47 done-for-you digital products for $47
The reason this ranked first is arithmetic, not enthusiasm. Cost to first sale: $47, one time. Time to first sale: two days. Margin: you keep everything payment processors don't take. The package includes 47 complete funnels — sales pages, upsells, email sequences — and 192+ products in total across the funnels, with ad creative already written for Meta and Google. Nothing about it is a subscription. There is a 30-day money-back guarantee if you disagree with our ranking.
The honest downside, which we logged in the cons column: this hands you the shop, not the shoppers. You still have to learn how to put traffic in front of a funnel. The difference is that you are learning one skill against pages that are already built to convert, instead of learning product creation, copywriting, design, and traffic at the same time.
2. Self-built digital products
Higher margin than physical models and genuinely yours — but 63 days to first sale, and that was a tester who already knew Canva. If you love building and you are patient, this works. Most people are not patient. That is not a character flaw; it is just how the first six months feel.
3. Affiliate marketing
Cheap to start, and the fastest path to learning traffic — which is the one skill every model on this list eventually demands. The margin ceiling is the problem: someone else owns the customer, sets the commission, and can change it.
4. Amazon KDP
Low cost, slow money, and an algorithm you do not control. Our tester's best week was $61. Fine as a side channel. Not a business.
5. Productized freelancing
Zero startup cost and real money quickly, but you are selling hours again, and the math only works while you are working. It ranked above five models on cash, and below four on freedom.
Why Our Team Ranked LOOT First
Because in a cost audit, the winner is the model where the smallest bet produces the fastest honest answer. For $47 — less than the shipping budget of every physical model in this study — you own 47 finished products with their funnels, copy, and ads already built. You are not paying to find out whether your idea works. You are paying for inventory that is already assembled, and then learning the one skill that turns inventory into money: distribution.
Compare that to the other end of the table. Our FBA tester spent $2,140 learning the same lesson — that traffic and offer matter more than the product — and he learned it without owning anything he could resell next week.
Both testers ended the study in the same place: understanding that a business is an offer plus traffic. One of them paid $2,093 less to get there, owned 47 products at the end of it, and had already made sales.
Questions Readers Asked Before Publishing
Is $47 really the full price, or is there a subscription after checkout?
It is the full price. One payment of $47, lifetime access, 100% of your sales kept. If a product exists that charges you $47 and then bills monthly, it is not the one we tested.
Do I need technical skill or an existing audience?
No. Publishing a funnel is a copy-paste-and-connect exercise; the products are finished and the sales pages are written. What you do need is willingness to put traffic in front of a page — that part has never been done-for-you in any model on this list.
How is this different from the $17 PLR bundles online?
The bundles sell you documents. LOOT sells you businesses-in-miniature: the product, the sales page, the upsells, the email sequence, and tested ad angles — per product. The asset is the funnel, not the PDF.
What if it is not for me?
There is a 30-day money-back guarantee. Read the terms at checkout like a grown-up.
See All 47 Products — Get Instant Access For $47 →