A B2C SaaS for seafarers: test preparation and technical reference content in English. About 30,000 monthly visitors, a large question bank, and a small content team. The product was free and monetized only by context ads.
In January 2026, the owner was ready to shut it down. Revenue was about $100/month and did not cover costs. The team was at risk. I took over the product with no external PM guidance.
Problem
The product had an audience but no business model. Free content attracted traffic, but ads could not cover the cost of content and engineering. Users came to pass certification exams; the product gave them content, but not a reason to pay.
The question was not “which feature should we build?” It was: can this product become a subscription business without destroying the organic traffic that keeps it alive?
Evidence
- Revenue: ~$100/month from context ads.
- Audience: ~30,000 monthly visitors; no paid acquisition.
- Monetization: none — no paywall, no subscription, no paid archives.
- User behavior: support conversations and search behavior showed that users came for test questions and answers, not for ads or generic articles.
- Cost base: a small content team and hosting; the product was not covering its costs.
Options considered
- Keep ads and cut costs. Reduce the team and accept a small ad-supported product.
- Sell one-off PDFs only. Low-ticket, no recurring revenue.
- Put everything behind a paywall. Maximum short-term revenue, but high risk to SEO and to users who had never paid.
- Freemium subscription plus paid PDF archives. Free value first, paid depth at the moment of need.
- Paid acquisition. Spend on ads to grow out of the problem.
Decision
I chose freemium: free content and a calibrated free tier, with premium subscriptions and paid PDF archives. No paid acquisition.
Why:
- Users do not pay for a pig in a poke. They need to see value first.
- The free tier had to be designed around exam behavior: 5 free questions and 10 free searches per day. That is enough to feel the product, but it runs out at the moment when a user needs more.
- A full paywall would have damaged the organic traffic that made the product valuable.
- Paid acquisition was not affordable and would not fix a broken business model.
- PDF archives gave a low-ticket entry point; subscriptions gave recurring revenue.
Execution
- Negotiated a two-month reprieve with the owner: the closure was postponed from March to May.
- Rebuilt the monetization: premium subscriptions, PDF archives, limited free search.
- Implemented an SEO-safe paywall with schema markup. The goal was to monetize without triggering a Google cloaking penalty; organic visibility and bounce rate stayed stable.
- Added payment rails for markets where PayPal was not enough: crypto for Turkey, Anypay for Russia.
- Handled unpaid orders personally: 1–2 of 5–7 daily unpaid orders were converted through direct follow-up; about 15 clients were brought to sale in the first program. Later trained a team member to run the same playbook.
- Led a small team: three editors at the start; later two editors and an SMM trainee. One editor left voluntarily; no layoffs during the turnaround.
- First sale: March 27, 2026, $10 via PayPal.
Sidebar: the card-payment outage
In August 2026, users were writing “I can’t pay by card.” The “Debit or Credit card” button opened a grey, unclickable form in every browser; PayPal worked. Support and the team initially assumed user error.
I checked the network requests: our side was clean. I disabled our custom scripts: not the cause. The root cause was an open bug in the WooCommerce PayPal Payments plugin (GitHub issue #4524, known since July 14, 2026). The plugin’s script was blocking the card form.
I did not replace the plugin — that would have been a large, risky migration for a problem that could be isolated. I shipped a CSS hotfix that unblocked the card fields.
After the fix, card payments became a major payment path: roughly 40–50% card, ~40% PayPal, the rest crypto/Anypay. The business effect was not “a CSS fix”; it was a restored revenue path.
Result
| Metric | Before | After |
|---|---|---|
| Monthly revenue | ~$100 | ~$967 by July 2026; ~$2,171 total revenue by October 2026 |
| Subscription revenue | $0 | $879/month (October 2026) |
| Gross margin | Not meaningful | 54.3% |
| Profit | Negative / not covering costs | ~$1,179/month (October 2026) |
| Paid acquisition | None | None |
| Monthly traffic | ~30,000 visitors (baseline) | ~43,804 visits in October 2026; 52.5% organic |
| Team | At risk of closure | Retained; one voluntary departure |
Total revenue includes subscriptions, PDF archives, lifetime sales, and advertising/partner placements. It is not MRR. Subscription revenue was $879/month; the rest was one-off and partner revenue.
The product moved from “shutdown planned” to a sustainable subscription business. The owner stopped talking about closing it.
What I did not have: repeat-purchase tracking, cohort retention, and LTV. I instrumented what I could and made decisions with the data I had. Building a proper metrics layer was the next step, not something I pretend I already had.
What I learned
- A business model is a product decision, not a finance afterthought.
- The free tier is the product. If it is not designed around the moment of willingness to pay, the paywall is just a wall.
- SEO and monetization can coexist, but only if the paywall is implemented with the search crawler in mind.
- The hardest part was not building the paywall. It was deciding what not to give away.
- What I would do differently: instrument repeat purchases and cohorts earlier; run pricing experiments before scaling; set kill criteria for new products before they start.