A Niche Site's Full Ledger: $24,205 in Six Years, and the $10,000 Month It Led To

In November 2025, a developer in Edinburgh who publishes as Pete Codes put his website up for sale and did something almost nobody does: he published the complete financial history of it. No CS Degree, a collection of more than 200 interviews with self-taught and bootcamp-trained developers, had earned $24,205.47 in total since 2019. In its final twelve months it took $796.97 and cost $1,532 to run, a loss of $735.03.

That is not a failure story, and it is not a success story either. It is the most useful thing a person running a niche content site can read, because it puts real numbers against a business model that is usually described only by its winners.

The full ledger

From his own sale post, the all-time revenue breaks down like this:

• Newsletter and sponsorship placements: $18,329.28
• Wes Bos course affiliates: $1,368.84
• ZeroToMastery affiliates: $1,048
• AlgoExpert affiliates: $788
• Job board postings: $396
• Beehiiv ads over 12 months: $275.35
• UK bootcamp sponsorships taken outside Stripe: about $2,000

And the audience behind those numbers: 9,477 email subscribers, a 38% open rate, roughly 5% click-through, 27,000 website visitors over the year, peaking at about 15,000 a month at its best, 7,835 followers on X, 366 YouTube subscribers, and a domain rating of 39.

Read those two lists side by side. Nearly ten thousand engaged subscribers, an open rate most publishers would envy, and six years of consistent work produced about $24,000 across the whole life of the project.

What that ratio actually tells you

Three uncomfortable and useful conclusions come out of this ledger.

Sponsorship is where the money is, and it is lumpy. Newsletter and sponsor placements produced roughly three quarters of everything. Affiliate links, the model most niche sites default to, produced about $3,200 across all providers in six years. The job board, the classic advice for a developer audience, produced $396 in total.

The revenue is a direct function of active selling. The final year collapsed to under $800 not because the audience vanished but because, in his words, "I simply don't have time to work on No CS Degree." He was busy with ghostwriting and SEO clients. Sponsorships do not renew themselves; someone has to sell them each time.

The costs never stop. Ghost hosting at $372 a year and email tooling at about $1,200 a year turned a quiet site into a loss-maker. A content business at this scale has a floor of fixed costs that a few hundred dollars of affiliate income cannot clear.

The part that did work

Here is where the story turns, and why it belongs in a series about ideas that earned. On his own about page, Pete says he has been bootstrapping full time since July 2019, when he quit his job to build No CS Degree, and that he hit $10,000 in monthly revenue for the first time in July 2023.

That $10,000 month did not come from the site's sponsorships. It came from what the site made possible: LinkedIn and X ghostwriting for founders and brands, freelance SEO work, a second newsletter called High Signal with a $39 a month membership, and courses. He also sold an earlier project, Remote Companies, in March 2022.

The site was never the product. It was six years of proof: 200 published interviews demonstrating that he could find people, ask good questions, write clean copy and grow an audience from nothing. That is an unusually credible portfolio for someone selling exactly those skills, and it is why the services command real prices.

This is the honest version of the argument that publishing is the moat. The publishing rarely pays well by itself at this scale. What it pays for is access, credibility and inbound demand for something else.

The format was cheap on purpose

It is worth noticing how the site was actually made, because the production model is the most copyable part of it. Each piece is an interview with a developer who got a job without a computer science degree. The site's own criteria are specific: a featured developer must have landed a software engineering job or built a product earning $1,000 a month or more.

That format solves three problems at once. The subject supplies the content, so writing time is mostly editing. The subject shares the finished piece, so distribution is built in. And the criteria create a filter that makes the archive credible rather than padded.

Two hundred interviews at that cost per unit is a realistic weekend-and-evening project. It is also why the archive kept pulling traffic for years after individual pieces were published, and why the email list kept growing without paid promotion.

How to read this if you run a small site

1. Do the per-subscriber arithmetic before you commit. Nine thousand subscribers in a professional niche produced a few thousand dollars a year from sponsorship. If your plan needs more than that, the plan needs a product, not more subscribers.
2. Decide early whether the site is the business or the marketing. Both are valid. Confusing them is what leads to six years of unpaid work with a vague hope attached.
3. Sponsorship revenue requires a salesperson. If that is you and you get busy, the revenue goes with your attention.
4. Affiliate income is a rounding error at this scale. Useful as a supplement, fatal as a strategy.
5. Track fixed costs against trailing revenue every year. The moment they cross, you are subsidising a hobby, which is fine as long as you know.
6. The asset is the email list and the archive. Both survived a year of neglect, which is precisely why an owned audience is worth more than platform reach.

The timing nobody controls

One more factor sits under the final-year numbers. Traffic to No CS Degree fell from a peak of around 15,000 visitors a month to about 2,300 in the last thirty days he measured. Some of that is neglect. Some of it is the broader shift Builtplain measured in the clicks that disappear when an AI summary appears, which has hit exactly this category: informational content answering questions a chatbot now answers inline.

A person starting a niche interview site in 2026 faces a search environment that is materially worse than the one this site grew in. That does not make the model dead. It makes the email list, the sponsors and the direct relationships more important than the pageviews, since those are the parts a summary box cannot intercept.

Why publishing the loss matters

Most sale listings for content sites are written to flatter. This one is a plain accounting of a project that earned $24,205 over six years, lost money in its last twelve months, and still built the reputation that led to a $10,000 month elsewhere.

Both halves of that sentence are true at once, and the second half only happened because the first half was done properly for six years. That is a more accurate picture of how small internet businesses actually work than any revenue screenshot, and Pete deserves credit for showing the whole thing rather than the flattering half.