Reading guides & resources

The genuinely free places to read or download books, and the specific rights that make each one free. Plus how to tell a real free edition from a stolen one in about ten seconds.

What copyright expiry actually does, why it is territorial, why a public domain book can still be sold, and what it does not mean for translations, editions and audio.

The categories of genuinely free ebook source, what rights each rests on, and the single question that separates all of them from a pirated copy.

EPUB reflows, PDF does not. That single difference explains nearly every practical difference a reader will ever notice between the two formats.

Why two ebooks of the same book can be completely different to read, and the specific production choices that separate a properly made edition from a dumped one.

The number on the cover of an ebook is not a measurement of the book. It is a conversion of a fixed-format assumption, and understanding how it is made explains a number of confusing comparisons.

Six reliable methods, ranked by how often they actually work, and why an algorithm will keep handing you the same book you already loved.

The three genres are separated by a single question, what do you already know, and mistaking one for another is why readers abandon books more often than for any other reason.

Two communities, two names, much the same shelf. The difference is not in the book, it is in which question the reader arrives with and how each community discusses it.

Bestseller lists measure sales of books with marketing behind them. Four methods that reach books which never appear on one.

A recommendation from someone you follow is a different kind of information from a ranking. What that does to which books get read, and what it costs the author.

Advance reader copies are one of publishing's oldest arrangements and one of its least regulated. What they are, who receives them, and what reading one actually commits you to.

Reader reviews are unusually good at spoiling books while trying not to. The vocabulary that betrays them, and the review formats that make the problem avoidable.

A box set can be a series collected, an anthology assembled, or several authors bundled together, and the three behave very differently once your shelf runs out of room.

Paid and gifted reviews are a normal part of the book economy. What disclosure is actually required, and the tells that remain when nobody follows the rules.

The advance copy system, the review-copy process at retailers, and why the number of reviews a book has in its first month is mostly a logistics question.

The two are not versions of each other. They occupy different points in the funnel, behave differently under budget, and fail in different ways.

Attention and purchase are different things the algorithm optimises for separately. Recognising the gap is most of the work of reading a campaign honestly.

Spark Ads put a real organic post into the paid feed. That sounds like a smaller version of a normal ad and is instead a different mechanism with a different cost and a different failure mode.

The path is longer than the click and has at least four steps. Knowing which one is leaking is most of the value of running this channel.

When amplifying an existing post makes sense for a book, what it actually costs you in money and in access to your own audience, and how to tell whether a video is worth backing.

The metrics in the reporting dashboard, what each one is optimised toward, and the two numbers that decide whether the campaign is working.

A lookalike is a modelled guess at who resembles your customers. That makes it an inference rather than a measurement, and the distinction decides how it should be used.

Two jobs with different economics, different failure modes, and a budget split that is almost always wrong in the same direction.

A demand-side platform is software that buys ad space automatically. Understanding the auction it enters explains most of what people find confusing about programmatic buying.

Return on ad spend is the most quoted and least understood advertising metric. It is a ratio of two attributed numbers, and both of them are estimates.

Every platform credits itself with sales it probably did not cause. The two ideas that make sense of this, and the experiment that settles it.

What platforms actually infer about people, where the signals come from, and why the phrase personalised is doing a great deal of work in most ad copy.

BookTok has no official chart, which is precisely why the word trending is used loosely. A practical account of what the signal is, who measures it, and why a single video proves nothing.

The mechanical explanations that recur across documented cases, and why genre momentum matters more than any individual video.

Wrap-up, POV, romantasy, comfort rewatch, trigger warnings and the rest. The vocabulary readers use, and what each term actually implies about the book.

A reusable method for evaluating any book promotion, rather than a list of offers that would be stale the day it was written.