About this book

Most growth books are written backwards. Someone succeeds, then constructs a framework that explains the success, then presents the framework as the cause. The reader gets a theory that has never been tested against a case where it didn't work, because those cases were edited out.

This book is written from a single complete dataset: every day of search data from the first day a product was indexed to the day it passed a quarter of a million users, plus the behavioural analytics for the same period, plus a dated log showing exactly what was shipped and when.

That lets me do something unusual: check the advice against the numbers, including where the numbers contradict the advice. They do, in three important places, and those three chapters are the most useful in the book.

The evidence base

The case is a browser-based document tool called ihatepdf, built alone, launched 24 December 2025, at 249,472 unique users on 17 September 2026. Every figure quoted comes from one of three sources:

Acquisition spend across that period: zero. No ads on any platform, no paid links, no sponsorships, no agency, no paid directory placements.

What this book is not

It is not a promise that you will get the same result. It is one case. A single case can prove that something is possible and can show you mechanisms, but it cannot tell you how often those mechanisms work. Where I am generalising beyond what one dataset supports, I say so.

It is also not a monetisation book. This case reached a quarter of a million users and very little revenue, and Chapter 30 is honest about why those are different problems and how the choices in this book make the second one harder.

How to read it

Part I is about what to build. Part II is about how to make it findable. Part III is about reading your own numbers, and is the part most people get wrong. Part IV is about the ceiling you will hit. Part V is the bill. Part VI is what came next, including the book re-running its own findings on nine months of data and reporting which ones survived.

If you only read three chapters, read 3, 4 and 14. They are the ones where the data disagreed with what I believed.

Every chart in this book is drawn from the exports listed in Appendix A. Where a chart would have been prettier than the truth, you get the truth. There is a chapter whose central number is nine sessions, and a chapter that declines to draw its own obvious conclusion because the data does not support it.

Three ways to read this

The book is ordered as an argument, front to back. But most people arrive with a specific problem, so here are three shorter routes. Each is self-contained.

"I have no traffic at all."

Chapters 1 → 2 → 6 → 7 → 11 → 14 → 38. Is there demand you can reach; what the first ninety days actually look like; how to decompose your product into entry points; the page specification field by field; why your pages may be invisible; why you must not judge any of it for a quarter; and the whole page on one graded list. Then Appendix B, and start.

"I have traffic but I'm stuck."

Chapters 17 → 23 → 24 → 25 → 28. What high impressions with no clicks really means; the diagnostic that separates an on-page problem from an authority problem; why you get linked; what actually earns links versus what only moves a vanity score; and the outreach mechanics. Run the audit tool first. It will tell you which of these applies to you.

"I have traffic that doesn't convert."

Chapters 18 → 19 → 20 → 16. The half of the funnel search data cannot see; the dead-click finding; the device split; and why your falling click-through rate is probably arithmetic rather than a problem.

And three chapters worth reading whatever brought you here: 3, 4 and 14. They are the ones where the data contradicted what I believed, and they are the reason the rest of the book is shaped the way it is.

About this book

About this book is part of How to Grow Your SaaS to 250K Users Without Spending on Ads, a playbook on taking one product from zero to 250K+ users on search alone, with nothing spent on advertising. Every claim in it is checked against the real data, including the three findings that contradicted the author.

Continue reading

Free tools that implement this book