Reader Survey Template

Find out who is actually reading, what they finish, and what they want next — the audience questions your analytics dashboard cannot answer.

Free · copies into your editor in one click
Live preview — try it, nothing is saved

You read the things we write — now we get to read something of yours. Six quick questions, and the editorial calendar takes notes.

press Enter ↵

Analytics tell you what readers clicked; they cannot tell you what readers wanted, skipped, or quietly gave up on. Pageviews count arrivals, not satisfaction — a listicle that disappoints ten thousand people outscores an essay that delights two hundred. A reader survey is how publications correct for that distortion, and this one is tuned to be answerable in ninety seconds by someone who was mid-scroll when they found it.

Why these questions. Reading frequency separates your devoted core from drive-by traffic — and you should read every other answer twice: once for everyone, once filtered to "every post" readers, because those are the people your publication actually belongs to. First-discovery data assembles your real acquisition story over time in a way last-click analytics never will ("a friend sent a link" is the growth channel dashboards systematically undercount). The topic votes, capped at three, force prioritization instead of encouragement. The length question is phrased as what readers finish, not prefer — aspiration says long-form, behavior says otherwise, and the wording nudges toward behavior. The star rating trends writing quality wave over wave, and the what-next question routinely produces better article ideas than the ideas meeting.

First-visit answers are a bonus study. Someone who found the survey on their first article and answered anyway is telling you how the publication reads cold: their discovery answer maps acquisition as it happens rather than as it is remembered, and their topic picks arrive unshaped by loyalty. The star rating is also worth reading as a shape rather than a number — a pile of fours means consistent quality, while a split between twos and fives means two different audiences are sharing one masthead, and the topic votes will usually name them.

What we left out. Demographics (ask separately if advertisers need them — mixing them here costs completions), reader-identity questions like job titles, and email capture. An anonymous survey earns blunter answers about your writing than one that might reply.

Who runs this. Independent bloggers doing an annual check-in, editorial teams planning next quarter's calendar, substackers deciding whether to double posting frequency, and content marketers proving the blog serves actual humans.

Make it yours. Replace the topic options with your actual beats — the placeholder list is deliberately generic. Embed the form inline at the bottom of a popular post where your real readers already are, or link it from your about page for a running pulse. Run it as an annual wave with a close date, or leave it always-on and compare quarters in the CSV export. Either way, publish two or three findings in a post; readers who see their votes shape the calendar answer again next year, and bring friends.

Frequently asked questions

Where should the survey live to reach real readers?

At the bottom of well-trafficked posts via the inline embed, where finishing an article is the qualification. A popup embed on exit works too, but expect more drive-by answers.

How many responses make it representative?

For editorial decisions, fifty-plus engaged readers beat a thousand random visitors. Filter answers by the frequency question in the CSV — your "every post" readers are the signal.

Annual wave or always-on?

Both work: close-dated waves give clean before/after comparisons, an always-on link gives a running pulse. Every response carries a timestamp in the export, so quarters are easy to split.

Do readers need to sign in to answer?

No — anyone with the link can respond, and the survey collects no identity fields, which is exactly why the writing critiques will be usefully blunt.