Surveys: many people use them.

But few feel they get the kind of information they truly need to improve their products, their services, and their businesses.

That's a problem the folks at Zapier, a tool that lets you easily connect apps and automate repetitive tasks, has worked to solve with their free ebook, The Ultimate Guide to Forms and Surveys. (You can download a free copy here.)

Here's their advice for how to use surveys to your advantage.

No entrepreneur can afford to say, "I don't know." You absolutely must know what your customers are thinking, how they're using your service, and what they want you to do next... and you needed to know all of that yesterday.

It's easy to reach for new tricks by mining your site analytics and hoping the data will point you in the right direction. And it might. Your best option, though, might be to do the old fashioned thing and just ask people for their opinions with a survey.

Surveys are decidedly low-tech, simple tools to get feedback. Write a quick survey--even just on a piece of paper--and you can get feedback from real people before 5 p.m. But if you're not careful, your survey questions can introduce bias into the answers, confuse respondents with compound questions, and wind up generating feedback that only gives you what you want to hear.

Don't do that. Here are some quick tips to help you make the most of your next survey and get feedback that is actionable and useful for your business.

1. Start With the Answers

It's not the questions that matter in the survey--it's the answers, the things people tell you in response to your questions. So, before you start writing questions for your next survey, think about the answers you need.

Start off by writing down your survey's hypotheses--the idea you'd like to prove or disprove with this survey. Perhaps you think people want more features in your app, or that they'd like banana flavored burgers on your menu.

Write that down and then list the data you'd need to prove or disprove your hypothesis. Think through the type of answers you'll need for each item; perhaps you'd want to know a percentage of people who want a new feature, or would like suggestions of other menu items.

Now, you can write the questions that will get the answers you need. Write them carefully, as you want to avoid bias and framing in your questions. Make sure your question is asked in a simple, understandable manner that doesn't emphasize either answer--you don't want fans to say something simply to agree with you.

Finally, pick the survey question type that'll get the answers you need. A yes/no answer is perfect for binary questions, while a multiple choice question might be better for more nuanced answers, and a textbox answer gives respondents the most flexibility.

2. Use a Survey Builder App

Paper's fine for a simple survey, but if you want to effectively survey thousands of people, a survey app is a necessity. Don't just use any app that can build a survey--after all, you could create a survey in a form builder app--but pick an app that's specifically designed for making a survey.

There are dozens of apps designed for building surveys, and even Microsoft's free Excel Online lets you make surveys inside a spreadsheet. Just about any survey app could work. The important thing is that your survey app includes the question types you need for your survey, and summarizes your survey results in easy-to-read charts and tables.

As an extra advantage, most survey apps include pre-made question banks that you can use in your survey. Those questions will be written specifically to avoid survey bias and framing, and you can use them as the basis for your own customized questions. They may save time in building a survey, but most importantly they will help ensure your survey gets the unbiased feedback you need.

Survey apps are generally simple to use--if you've ever built an online form or put together a PowerPoint presentation, the drag-and-drop editors in most form apps will be simple to figure out. All you'll have to worry about is getting your answers into the app and sharing your survey to gather feedback.

3. Find the Perfect Audience

Sharing your survey may be as difficult as writing out your original survey questions. It looks easy enough--build a survey, copy the share link, and send your survey out into the wild--but to get quality feedback from your survey, it must be shared with a diverse audience.

Share your survey only with your friends or team members, and you'll likely get positive feedback. Send a survey to your competitor's customers, and you'll get negative feedback. Ask a random person on the street, and you'll get silence.

Sampling bias, which results from survey results from respondents who are not representative of the population or intended focus group of a survey, can skew your results enough that they're entirely inaccurate. It's what made the 1948 U.S. presidential election get called incorrectly by newspapers, and it's what leads countless surveys off course.

There's no point in building a survey only to ask the wrong people, so you must consider how you share your survey. People who visit your website are different from existing customers, and those who answer email surveys are different from your Twitter followers. Beta testers and conference attendees, your highest-paying customer and those who asked for a refund: each represent a specific set of wants, need, and opinions, and you need to hear from them all to get a well-rounded feedback.

Try to share your survey in a number of channels, and consider the bias of each channel as you think about the answers they give. Make sure you have enough respondents to have statistical variation--survey a few people, see what variation you get, then estimate the audience size you'll need to survey for quality feedback. And never survey just the group of people you know will give the answers you want--there's no reason to do a survey in the first place if you don't want real, unbiased results.

Tip: Many survey apps let you export a PDF of your survey to print out, and some include mobile apps to gather survey responses offline. And any survey app can work with phone surveys--just read the questions off the screen, and type in the speaker's answers yourself. Don't let your app limit you to only collecting survey responses online.

4. Analyze Your Data Appropriately

Now that you've built a survey and have the results in hand, it's time to pull out the spreadsheet app and start crunching numbers.

Wait. Not so fast. Have you ever seen a terrible chart, ones that obscure data, make unimportant differences seem drastically important, and are filled with chartjunk? Your data deserves better than that.

The types of tables and graphs you should use depend on your answer types. Categorical data is best displayed in a frequency or contingency table, perhaps with highlighting to show the least or most popular answers. Ordinal data might seem easy to summarize with a simple percentage, but a frequency table is a much more honest way to display the data, perhaps accompanied by a graph that shows both the low and high ends of the answer ranges. And interval data must always be graphed with equal sized intervals.

Everything doesn't need a graphic, either. Tables are the best way to showcase precise numbers. If there are few things you can compare in the answers, don't turn them into a graph--show them together in a table so you can take each answer at face value. If you have numbers you can easily compare, though, a graph can be an easy way to visualize the answer distribution. Choose whether to use graphs or tables based on the answers, not just on which looks nicer in a presentation.

5. Only Go So Far

Your survey responses can fill detailed spreadsheets and make great boardroom presentations, but they're still not absolutely authoritative. You can use your survey to help shape your business plans, but only if you're sure you can trust your survey data and its interpretation.

To make your survey data trustworthy, try to take several different surveys over a period of time prior to acting on the surveys. See if the feedback seems similar every time, or if it varies enough to make each result feel uncertain. Don't be afraid to throw away your work if your thesis doesn't end up working out--that's why you did a survey in the first place.

Then, don't focus on the smallest points in the survey. Focus instead on the high points, the trends and largest differences, and not just the most unusual findings. Always remember that some margin of error will always exist in any survey method, and that the smallest differences may not actually be important at all. And don't make your survey data seem more accurate than it is--respect the limits of precision, and round percentages to clean numbers.

Gathering data in a survey--or any type of form--can be difficult. With a few guidelines and careful work, though, you can still make a survey that will give you actionable feedback, whether or not it gives the answers you wanted to hear.

After all, that's what is most important.

For more tips on creating great forms and surveys, the best apps to use, hidden features that can speed up your work, ways to automate form and survey data with Zapier, and much more, check out Zapier's new free eBook, The Ultimate Guide to Forms and Surveys. You'll learn more great tips to help you gather data better, along with free spreadsheet templates to help you analyze and graph your survey data.