About SplitTheWorld

SplitTheWorld is a weekly opinion poll with a single rule: you have to answer before you see the results. Every week, one carefully chosen question goes live. You pick your side. Then — and only then — the curtain lifts to reveal exactly how the rest of the world voted.

The result is almost always surprising. Questions that feel like they have an obvious answer turn out to be deeply divisive. Questions that seem niche turn out to be near-universal. That gap between what you expected and what the data actually shows is the whole point of SplitTheWorld.

WHY WE BUILT THIS

The idea came from a simple observation: most people are surprisingly bad at predicting how common their own opinions are. Ask someone whether pineapple belongs on pizza and they will confidently tell you their view is obviously correct — and that most reasonable people agree with them. They are often wrong on both counts.

This phenomenon has a name in psychology: the false consensus effect. We tend to assume that our beliefs, values, and preferences are more widely shared than they actually are. It is not a character flaw — it is simply how human minds work. We spend the most time with people who think like us, and we naturally generalise from that limited sample to the wider world.

SplitTheWorld was built to disrupt that assumption in the most entertaining way possible. Not with academic studies or carefully worded surveys, but with a single question, a real-time vote count, and the genuine, unfiltered result of how thousands of people actually answered when asked.

There is no agenda here. We are not trying to nudge you toward any particular view. We do not have a preferred answer on any question. SplitTheWorld is purely a mirror — held up to show the world how it actually thinks, on the things that matter a little, a lot, or not at all.

THE QUESTIONS

Building the question bank was the hardest part of creating SplitTheWorld. A question has to clear several hurdles before it earns a spot. First, it must be genuinely debatable — there should be no obviously correct answer that any reasonable person would immediately agree on. Second, it needs to be interesting enough that people actually care about the result. Third, it has to be answerable in seconds, without needing background knowledge or cultural context that not everyone would share.

The questions that work best tend to live in the gap between personal preference and something that feels like it should have a right answer. "Is a hot dog a sandwich?" is trivial on its face — and yet people feel strangely, almost angrily strongly about it. "Should you always finish a book you are not enjoying?" sounds like it has an obvious answer, but it does not — it depends entirely on your values around commitment, opportunity cost, and self-care.

Our current question bank covers more than a dozen categories, including food and drink, relationships and social life, technology and the internet, pop culture, philosophy, work and money, everyday habits and etiquette, travel, sport, and health. Questions are chosen to represent a wide range of topics so that no two consecutive weeks feel the same.

New questions are reviewed and added regularly. The goal is never to run out — and with 400 already in the bank, there are enough questions to run SplitTheWorld for nearly eight years without repeating a single one.

HOW THE DATA WORKS

Every vote cast on SplitTheWorld is recorded in real time using Cloudflare's global edge infrastructure. The percentages you see when you reveal your result are calculated directly from every vote ever cast on that specific question — not estimates, not projections, and not figures borrowed from other polls or research studies.

As more people vote throughout the week, the numbers shift. A question that starts on Monday at 55% / 45% might end on Sunday at 62% / 38% as different audiences discover the site. That live quality is intentional. SplitTheWorld is not a static snapshot — it is a running count of how real people are answering in real time.

Votes are completely anonymous. We do not collect names, email addresses, device identifiers, or any personally identifiable information when you answer a question. Your answer is stored only in your own browser's local storage, so you can return and see your result again. That data never leaves your device in any form that could identify you as an individual.

WHAT THE MINORITY / MAJORITY LABEL MEANS

One of the most interesting things SplitTheWorld reveals is not just what the majority thinks — it is where you personally sit relative to it. Finding out that you are in the 14% minority on something you considered completely obvious is a genuinely disorienting feeling. It prompts a kind of useful cognitive friction: were you wrong? Or are you just unusual?

The answer is that being in the minority does not make you wrong. It just means your position is less common. And being in the majority does not make you right — it just means more people agree with you. SplitTheWorld makes no judgement either way. But it does surface the data honestly, and invites you to sit with what that information means to you.

Some people find their minority status validating — proof that they think independently. Others find it prompts genuine reconsideration. Both responses are entirely reasonable, and both are exactly the kind of reflection that makes SplitTheWorld worth doing.

THE SHARE MECHANIC

Every result is designed to be shared. The combination of a bold percentage, a strong opinion, and the framing of majority versus minority creates a natural impulse to show people. "I am in the 23% who think cold pizza beats hot pizza" is a more interesting thing to post than a simple opinion — it carries real data behind it, and it invites others to respond with their own numbers.

Results from SplitTheWorld have been shared across X, WhatsApp, Facebook, Reddit, and Instagram. The conversations that follow are often more interesting than the original question — which is exactly what the best opinion polls have always done.

CONTACT US

If you have a question you think would split the world, feedback about the site, or anything else you would like to say, we would love to hear from you. Reach us via the contact page.

You can also browse past questions and their final vote counts in the Archive, read analysis of past splits in Insights, or see which questions are coming up next on the What's Coming page.