Reynard does not generate covers, titles, blurbs or a single word of your book, and never will. It reads what a stranger receives from the assets you already made. Reynard is not affiliated with any retailer or publishing platform.
It is that the cover is not to market. It screams sci-fi and the book is epic fantasy. It reads as cosy when the book is grimdark. The art is good, the typography is fine, and a stranger scrolling past cannot tell in one second what kind of book it is.
What that costs you is not taste points. It is that the store does not know who to show it to. When the cover says one genre, the title says another, the blurb opens on a third and the categories and comps point somewhere else again, nothing lines up and the also-boughts fill with the wrong books. Your ads get shown to people who were never going to buy, which reads back to you as a bad cover, a bad blurb or a bad book.
What genre does this read as to a stranger is a decodable question. It has an answer that does not depend on taste. That is the one thing a simulation can honestly do here, and it is the thing this page is about.
Scoring one asset in isolation is useless, because a reader never sees one asset in isolation. The whole bundle has to point the same way.
Conflating these is how a tool like this gets run out of an author group. They are different objections and they deserve different answers.
That view is close to consensus in indie publishing and we are not here to relitigate it. Reynard generates nothing. No cover, no title, no blurb, no chapter, no line of your book. You upload or describe what you or your designer already made, and it comes back as a read, not as a draft. Nothing you submit is turned into art or copy for you or for anyone else.
Fair, and the risk is real. A run is private. It produces a report you read, and nothing from it is published, watermarked, credited or visible to a reader. Nothing touches the manuscript, the cover file or the listing. If you would not want it on the book, it is not on the book — because it was never anywhere near it.
This is the one aimed at us and it is the correct objection. Authors have shown the same chapter scored as annoying or as hilarious depending on whether they asked for a fair review or a critical one, with everything coming back as praise. So Reynard never grades quality. It is not asked whether the cover is good. It is asked what genre it reads as, who it looks like it is for, and where the signals contradict each other — and the segments disagree with each other, which is the entire point of reading them separately.
A general-population panel is the wrong instrument for genre fiction and always has been. A cross-section of the population is not a panel of romantasy readers, and authors have been saying exactly that about general-public polls for a decade. It is a real limitation, not a caveat, and we are not going to dress it up.
What it does instead is narrower and still worth something: it tells you what signal a stranger receives from the cover, the title and the first line of the blurb. That signal is what decides whether your ad, your category placement and your also-boughts ever put the book in front of a genre reader in the first place. Getting read correctly by a stranger is upstream of being judged by the right reader.
A premade cover runs from around $35; a custom cover from an indie designer runs into the hundreds and up. Against that, a $50 poll is a real slice of a debut's entire cover budget. Reynard is a free first check with no account, then $12, $25 or $39 per run, one-off, no subscription. In euro markets, €9, €19 or €29.
Your own price is a live decision and it is a signalling decision as much as a money one: 99c reads as a debut nobody knows, $2.99 to $4.99 is the normal indie band, and above that a stranger who has never heard of you reads it as expensive. A run tells you which band a stranger thinks the cover and blurb belong in. It does not tell you the optimal price and nothing can, before you have sales.
Read this part before you pay for anything.
Which genre a cover direction signals, and who misreads it.
Read the comparisonTitles, subtitles and series names: what they make a stranger expect.
Read the comparisonNo accuracy figure has been published, and this page explains exactly why not.
Read the comparisonThe first check costs nothing and needs no account. Write the thing you are testing, pick the market, read the report.
Read the comparisonOne check costs nothing and needs no account. You see the format of the output before you decide whether it is worth paying for a full run.
Start the free check