How to Create a Survey That Doesn’t Suck (Huge Guide. Survey Question Examples Included)

Can’t live with ‘em, can’t live without ‘em.

Am I right?

Of course I’m right!

I know you know you should be surveying your audience on a regular basis but let me ask you…

Are you actually doing it?

If you are – awesome! (though – keep reading – because you might just be making a few costly errors in the way your survey is structured)

If you aren’t – I’m going to make it super easy for you.

Writing Survey Questions That Give You Insight Into Your Customer’s Minds

Surveys are something I help a LOT of my clients with because getting into your prospective customers’ heads is the BIGGEST piece of a creating an SEO strategy that actually works for your business.

One HUGE problem I see with a lot of surveys people send out to their audiences is that they’re just plain BAD surveys.

Marketing Research was one of my favorite classes in college so I can be a biiiiiiit of a survey nerd.

I’m constantly shouting at the TV when supposed news sites report the answers to questions that are clearly just BAD questions.

What makes a survey bad?

Here are a few common survey question mistakes:

Mistake #1 – Not Qualifying Your Responses

What the heck do I mean by this?

I mean you need to take off your assumptions as much as possible when writing a survey.

The big assumption a LOT of small business owners (and any survey writers) make is ASSUMING every single person who fills out their survey is already in their target market for the survey.

Here’s the truth though:

Not everyone on your list or who follows you on social media is necessarily in your target market.

Or maybe, someone else not familiar with you stumbles across your survey and fills it out.

Without including what’s called a Qualifying Question – you have no way of verifying that the people who actually filled out your survey are the people you WANTED to fill out your survey.

What does a good Qualifying Question look like?

It’s multiple choice, the response is required and your respondents can only choose ONE answer.

The question content? Is the most basic requirement of what it takes to be in your target demographic.

For example, if you target moms of young children, you would want to include the following:

Are you currently a mother with at least one child 5 years of age or younger?

( ) Yes, I currently have one child 5 years of age or younger

( ) Yes, I currently more than one child who is 5 years of age or younger

( ) No, I have children but none currently 5 years of age or younger

( ) No, I do not have any children.

( ) Other (please explain) ______________________

If you are targeting in-house graphic designers who work in companies with more than 50 employees, you might break this out into 2 questions:

Do you currently work for a company that has 50 employees or more?

( ) Yes, I currently work for a company that has 50 employees or more

( ) No, I currently work for a company that has less than 50 employees

( ) No, I am self-employed.

( ) No, I am not currently employed.

( ) Other (please explain) ______________________

 

Are you currently employed as a graphic designer?

( ) Yes, I am currently employed as a graphic designer

( ) No, I am currently employed but my job has nothing to do with graphic design

( ) No, I work as a graphic designer but I am self-employed.

( ) No, I do not work as a graphic designer and am self-employed.

( ) No, I am not currently employed.

( ) Other (please explain) ______________________

The trick here is break down your target market to the very minimum they need to reach to qualify as your target market. What is the single differentiating factor that everyone in your target market shares that makes them different from the rest of the world?

THAT makes them your target market. And THAT makes the perfect qualifying question.

Mistake #2 – Leading the Witness

We all make assumptions about our audiences. One of the points of surveying is to find out if those assumptions are correct or incorrect.

What messes up your survey answers?

Leading the witness.

What do I mean? I mean baking an assumption into your question.

Something like the following might be a bit more obvious:

We gardeners are a thrifty bunch. How much would you estimate you spent on gardening supplies in the last 6 months?

The assumption baked into this question is that ALL gardeners are thrifty. The result is that your responders may report lower than factual estimates.

How do you fix it?

Simply remove the first sentence. Presto! Leading Question Be Gone.

But sometimes, it’s a little less obvious.

Take, for instance:

How hard do you find training for your last marathon to be?
(followed with a scale of 1-10, 1 being super easy, 10 being super hard)

At first glance, this one might not look like a bad question but take a second look.

The assumption written into this question is that training is hard – this is going to skew your responses to be “harder” than they might be if the question was left neutral.

A better way to ask the same question?

On a scale of 1-10, with 1 being super easy and 10 being super hard, please rate how you felt about your most recent marathon training efforts.

See how we removed the assumption there?

Not sure if a question of yours is leading? Post it in the comments below and let’s take a look!

Mistake #3 – Double-Barreled Questions

One of the biggest flaws with your survey can be if your respondents are confused by the mere asking of your question.

One of the most common confusing questions?

The Double-Barreled Question.

These questions are *actually* asking two questions in one but don’t give the respondent a chance to answer both individually.

Example:

How satisfied are you with your most recent copywriter’s rates & service?

( ) Not At All Satisfied

( ) Somewhat Satisfied

( ) Really Satisfied

Do you see it?

The two questions?

The two questions here are:

1. How satisfied are you with your most recent copywriter’s rates?
2. How satisfied are you with your most recent copywriter’s service?

Keeping these two questions smushed together means your answers are going to be inaccurate.

Some of your respondents are going to answer more about the rates, while others will focus on the service.

Splitting the question in two is all you need to do to fix it.

So What Makes a Survey Good?

We’ve just covered a lot of the common mistakes people make when writing survey questions. But what makes a good survey?

Elements of a *good* survey include at least the following:

Qualifying question

(but of course you already knew that – see above)

Psych-Friendly Question Order

Starts with multiple choice and “eases” into open-ended questions (it’s a psychological thing. Think baby steps of time commitment)

As short as possible.

If you’re going over 10 questions, huge red flags should be waving in your head. (If you use something like SurveyMonkey’s free plan, they won’t even let you go over 10). Me? I like a 5 question & under survey. The longer the survey, the less likely your respondents are to actually complete it.

Give a reason your audience should complete it

It doesn’t even matter what the reason is. Research shows just adding the word “because” can get more people to accept your request.

So simply saying something like

“Hey guys! If you are (insert your qualifying attribute), I would love it if you would take just 2 minutes to fill out this survey. Why? Because these answers help me shape what I create and offer to better serve you. Thanks!”

is going to get way better results than

“Hey! Survey! Click here to fill it out. Please.”

Include questions that actually get inside your audience’s heads

Want some ideas of what questions to include in your survey outside of your awesome qualifying one?

People are notorious for not knowing what they want.

“If I’d asked my customers what they wanted, they’d have said a faster horse.” – Henry Ford [Click to tweet that]

So your survey shouldn’t be designed to ask them what they want – it should be designed to get at the real gold – what problems do they need solving.

Here are a few of my favorites:

• From Tara Gentile
  • What frustrates you most about trying to…?”
  • What skills do you believe would improve your ability to…?”
  • How do you normally spend your weekends? Why?”
  • If you could wave your magic wand and change 1 thing about…, what would it be?
• From Derek Halpern
  • What are you struggling with?
• From Melissa Cassera (for those surveying *just* their existing audience)
  • What’s something you’ve always wanted to ask me?”
  • “What feels ‘blah’ about your life right now that I might be able to help with?”
• From KISSmetrics
  • What’s your biggest challenge in___?
  • What are your most burning questions about____?

Should you ask ALL of these questions? Heck no. Ask 1 or 2 that speak to you and how you communicate with your audience.

Create Your Survey

Once you’ve got your questions lined up, it’s time to create your survey.

I recommend using a service like SurveyMonkey (mostly free) or a Google Form (free) to pull it all together and make reading your answers a *lot* easier.

My go-to choice these days is Google Forms.

Here’s a quick tutorial I recorded for how to setup your first survey using Google Forms:

Hello? Anyone Out There? Where to Distribute Your Survey

Armed with your “because” reason (see above), it’s time to distribute your survey to your audience.

Here are a few places to start:

• Your email list
• Your Facebook page
• Your Twitter followers
• Any other social media followers
• Announce it on your blog
• Put a notification bar at the top of your website (like Hello Bar) with a link to the survey
• Individually email people you know are in your target audience

Following not that big yet? Don’t know anyone? Here are a few other places to distribute:

• Find a Google Plus Community where your people are hanging out
• Find a Private Facebook Group where your target market is
• Ask people you know to forward the survey to people they know in your niche
• Use a Facebook ad to target people in your target market (paid option)

How many responses do I need?

When I work with clients to put together these surveys to get some intel before developing an SEO strategy that works, we aim for at least 20 responses. The more the better.

Keep in mind that not everyone will actually take the time to fill out your survey. For every 10 people you ask, you might get only 3 people to fill it out. So you might have to ask close to 60 people before you hit at least 20 responses.

Nerd Alert: Is 20 responses “statistically significant” aka is it a large enough # of responses to represent what the rest of your community wants? Truthfully no, but that’s why it’s important to make surveying a part of your annual marketing process.

As your audience grows, the more responses you’ll get and the more themes you’ll be able to draw from your ever-so-growing statistically significant responses. Win!

Feel Like a Better Survey Creator Already?

Did you find this helpful? Have a question? Have another tip to share about using surveys? Let us know in the comments.

Think you know someone who could use this post as they start to create a survey of their own? Go ahead and share this with them – because you know I love to share the marketing nerd wealth