Have you used Online Polls on Your site?

A few practical notes based on the thread: several members (@InsightsDigital, @muta, @MktgRob) report low take-up. That pattern is common — polls are cheap to add but often give low, self-selected samples unless targeted. Treat them as a way to gather directional feedback or drive a small interaction, not as a statistically representative study.

When a poll is worth keeping: the question should be single, short, and relevant to frequent visitors (not generic). Better placement is inline with content readers came for, or inside a newsletter or members-only area where the audience is already engaged. Social-network quick polls (LinkedIn/Twitter) reach a different audience and can get higher raw votes, but they still reflect the platform’s demographics.

Tactics to raise useful signal: limit answer choices (3–5), display results immediately to reward voters, run the poll for a defined short window, and instrument it as an event in analytics to measure vote rate (votes / unique visitors). For more reliable insight, send the same question to an email list or logged-in users — that reduces duplicate and bot voting and increases relevance. If preventing repeat votes matters, track by account or unique cookie rather than IP alone.

How to interpret results: assume selection bias. Small totals are directional only; combine poll data with behavior metrics (time on page, click-throughs, signups) before acting. If the aim is market research, use a proper panel or randomized survey method rather than a site poll.

A lightweight test plan: pick one clear question, place it in two locations (homepage vs inline article), run each variant for a fixed period, track event conversions, and compare both vote rate and downstream behavior. If votes stay near zero after those tests, reallocate the space to something that directly drives conversions or content.