Home › Design Guides › Testing Book Cover Concepts With Reader Polls

Testing Book Cover Concepts With Reader Polls Before You Launch

Published July 6, 2026 — Design Guides

Posting two cover concepts to a Facebook group and asking "which one do you like better" feels like market research, but it usually is not. The votes you get back are shaped far more by which image is simply prettier in isolation than by which one would actually sell more books on a retailer shelf, and the people voting are rarely a fair sample of your actual target reader. That does not mean cover polling is useless — it means the useful version of it looks different from the casual version most authors run.

Ask the Right Question, Not "Which Do You Like"

"Which cover do you like better" measures aesthetic preference in a vacuum. What you actually want to know is which cover most effectively signals genre and makes someone want to know more, which is a different question. Better prompts: "Based on these covers alone, what genre do you think this book is?" tests whether your genre signaling is working before anyone has read a word of the description. "Which of these would make you stop scrolling?" gets closer to actual purchase behavior than a straightforward preference vote. Running both questions on the same set of covers, separately, tells you more than a single popularity contest ever will — a cover can win the popularity vote and still fail the genre-recognition test, and the genre-recognition test predicts sales far more reliably.

Where You Poll Determines What the Data Means

A poll posted to your own author newsletter or existing reader Facebook group measures your existing fans' taste, which is useful for confirming you have not alienated your current audience, but it says nothing about whether a new reader browsing a retailer page for the first time would be drawn in. A poll posted to a broader reader community for your genre — a subreddit for your specific subgenre, a genre-focused Facebook group you do not already run — gets you closer to a cold-audience read, but comes with the tradeoff of much less context about who is actually voting. Neither audience alone gives you the full picture; if you have the time, run the same poll in both settings and pay attention to where the results diverge, since divergence usually points at whether the cover works primarily on existing goodwill or on the image alone.

Sample Size and the Danger of Small Polls

A poll with 20 or 30 responses can flip based on a handful of enthusiastic votes from friends and family, and treating that as a real signal is one of the more common ways authors talk themselves into a weaker cover because it happened to win a small, biased sample. Aim for at least 100 responses from a source outside your immediate personal network before drawing any conclusion, and be skeptical of a result that is close — a 55/45 split with 100 votes is genuinely closer to a coin flip than a decision, and other factors, like which cover your designer and trusted beta readers independently preferred, should carry more weight than a narrow poll margin.

Poll Concepts, Not Near-Finished Files

Run the poll at the concept stage, with two or three genuinely different directions, rather than after a single concept has already gone through several rounds of revision. Polling minor variations of one design — a slightly different font, a shifted color — invites people to vote on noise, and by that stage you have usually already sunk enough design cost that a poll is more likely to be theater than a real decision point. If you want structured feedback specifically on a near-finished cover rather than a public vote, that is closer to the feedback process covered in our guide on briefing and reviewing designer concepts, which works better as a small group of trusted readers giving specific, directional feedback than as an open popularity poll.

What to Do When the Poll Contradicts Your Gut

  • Trust the genre-recognition data over the preference data if they disagree — a cover readers correctly identify as your genre will usually outsell a cover they merely find prettier.
  • Weigh comments over raw vote counts. A written comment explaining why a cover feels off-genre or confusing is more diagnostic than a tally.
  • Remember the poll is one input, not a verdict. Combine it with how the cover performs against genre convention benchmarks for your category before making a final call.

More Design Guides

Related: How to Brief a Cover Designer: A Template for Authors · When to Redesign Your Book Cover: Signs Your Current Cover Is Hurting Sales