Short answer
Read the vote count before you read the percentage. A result built on a handful of votes is an anecdote wearing a number. Everything here comes from visitors who chose to vote, so it describes that self-selected crowd and nobody else. For what groups do when a real date and a real budget are involved, set the result beside the figures from our inquiry records, which measure behaviour instead of opinion.
How do you read a result?
Two numbers per answer, and the small one matters more.
Each answer shows its share of the vote and, in brackets, the raw count behind that share. Below the answers sit the total votes and whichever answer is currently ahead. At the top of a topic there is a summary line: how many polls it holds, how many votes it has collected, and how many of its polls have any responses at all. Within a topic the busiest polls come first, so the results resting on the most evidence are the ones you meet first.
Check the bracketed count every single time. Three votes out of four is a landslide on paper and means nothing in practice. A working rule: under about thirty votes you are reading a few people's opinions, and well into the hundreds you are reading something worth noticing. Shares are rounded one answer at a time, so a poll can add up to a point over or under a whole.

Is this a survey of Houston?
No, and it would be dishonest to dress it up as one.
Nobody was selected to answer. The people voting are the ones who found a group transportation website, opened the polls and decided a question deserved a tap. That makes them more likely to be planning something, more interested in party buses and limousines than the average person walking down Westheimer, and perfectly free to vote on topics they have never experienced. A college student can weigh in on corporate shuttle etiquette. A visitor from another state can vote on the Houston questions.
Timing bends things too. The first handful of votes sets the leading answer, and a poll that gets passed around one friend group can tilt toward that group's taste and stay there. None of which makes the results worthless. It makes them a description of the people who voted, which is how they should be read.
Nobody was asked. Everybody volunteered. That is the whole caveat.
How does this compare with your inquiry data?
They are measuring two different things, and the gap between them is often the interesting part.
A poll records what someone would like. Our inquiry records capture what people asked for with a real date, a real group and a real budget sitting in front of them. When the two agree you can lean on the answer. When they diverge, the distance usually maps the space between the night people imagine and the night they are willing to pay for.
Trip length shows it neatly. About 53% of Houston inquiries with a requested duration asked for three to four hours, and across all the markets in our records that rises to about 69%. If a poll on the ideal night out leans longer, nothing is broken. That is a wish standing next to an invoice. Group size behaves the same way: about 57% of Houston inquiries that mentioned a party bus and a group size were for 11 to 20 people, which is why the 20 passenger party bus is the first size to check regardless of what a poll on the perfect party size says.

Which one wins when they disagree?
Taste to the poll, logistics to the data.
A vote can tell you the music, the dress code, the kind of stops most voters enjoy. It cannot tell you how many hours your route needs or which vehicle your eighteen people fit in. For those, the behaviour figures are the better guide, and the planning tools turn them into numbers that belong to you rather than to the internet.
Vehicle style makes the split obvious. The Limo Sprinters topic can be full of enthusiasm while only about 7% of Houston inquiries with a quoted vehicle on file mentioned a Sprinter van. The poll is not wrong. It is telling you a Sprinter is a genuine option for a smaller group, and the 14 passenger limo sprinter is worth a look if your list is under fifteen. To find out how many hours the plan really needs, lay it out in the event timeline builder.
Let the votes pick the playlist. Let the data pick the bus.
What will a poll never tell you?
Four things, and they happen to be the four that cost money.
It cannot tell you a price, whether a vehicle is free on your date, what a specific vehicle includes, or whether a company will turn up when it said it would. For price, the Houston party bus price guide explains what moves a quote. For judging a company, start with the reviews guide and the questions in our safety planning guide.
And if you want to use the polls as an actual conversation with your own group rather than a scoreboard, the polls guide covers which topics to open at which point.