How To Write Great Surveys with Actionable Data Results

Writing a great survey isn’t easy. At first it might seem straightforward, but after banging out a few questions with your favorite survey tool (I use Wufoo most of the time), you start to see the complexity and intricacies involved.

More and more startups are creating surveys as a means of collecting data from early prospects and customers. I’ve seen a few examples of this implemented; there’s a landing page with an email sign-up, and once you’ve converted, a survey pops up. This is a great way of collecting user feedback. But only if the feedback you collect is useful.

Recently I’ve been helping a friend collect data through surveys for a project he’s working on. (Sadly I can’t go into more detail than that, but it’s definitely interesting!) I logged onto Wufoo, created a new survey, and started writing questions. As I looked back at the questions, I asked myself, “What’s the point of each question?”

Good question.

Actually, out of all the questions I wrote (or asked myself in my head) that was the best one. What’s the point?

In reviewing the questions, I wanted to be crystal clear on each one’s objective. It took me a fairly long time re-working the survey before I felt it was ready for action.

I’m far from an expert survey-writer. And below I’ve included a number of great resource links to help you learn more about writing great surveys. But I wanted to share what I’ve learned to-date:

  1. Take the necessary time to write a great survey. Don’t rush this step because you’re eager for customer data. And don’t rush this step because you “just know” that all the answers are going to be what you want. Take your time. Do the research. Think things through.
  2. Surveys don’t have to be static. Feel free to iterate on your surveys, changing them as you see data coming in, adding / removing / editing questions as appropriate. I wouldn’t take this step lightly, because it could make the analysis more complicated, but don’t assume you nailed the perfect survey right away. And I think it’s reasonable to try A/B testing your surveys too.
  3. Focus on actionable data and metrics. You need results that you can actually do something with. There have been some great posts on actionable metrics. I encourage you to read them all.
  4. Be strict with yourself. I found when writing surveys it’s easy to let things slide a bit. You throw in a question, even if you’re not sure it’s right. Or you don’t edit the language carefully enough, assuming people will understand questions clearly. You need to be ultra-vigilant. And you need to be unafraid to kill a question completely if it doesn’t meet the proper standards.
  5. Don’t worry too much about length. I’ve experimented a bit with various survey lengths, and haven’t found it’s made a huge difference in completion. I suspect this is similar to the debate between long and short landing pages — it’s not the length that matters as much as the quality and effectiveness of the content.
  6. Ask tough questions. You have to force yourself to ask tough questions. You can’t be afraid, otherwise you won’t get enough honest, raw and actionable results. The survey.io tool (for very simple but useful customer surveys) asks, How would you feel if you could no longer use the product? That’s a tough question to ask, because you might not like the answer.
  7. Stay open-minded to discovery. As I look at the survey results I’ve collected, I’ve noticed that there’s lots to discover that was unintentional. This may not be a great thing, but it’s still interesting. I have a couple of open-ended paragraph-style questions in the surveys and the results are absolutely fascinating. They lead me to think about new ideas, and also make me want to follow up with respondents to dig further. It’s just about staying open-minded to the possibility that you don’t know everything about everything (which is fairly easy in my case!)

I have a lot to learn about writing great surveys. But one thing is certain: You need to recognize the importance of collecting actionable survey metrics, testing surveys and putting a lot of effort into getting them right.

Here are some resources for writing great surveys:

October 27, 2009 Posted in Business by

  • http://newtermlifeinsurance.net/ term life

    Yeah, you are true, mate. A lot of effort must be needed to get the right surveys with great results. Keep blogging and thanks for sharing

  • http://newtermlifeinsurance.net/ term life

    Yeah, you are true, mate. A lot of effort must be needed to get the right surveys with great results. Keep blogging and thanks for sharing

  • matthewneedham

    Great post Ben.

    I get increasingly frustrated with the number of organisations that ask pointless questions – which give them no insight into what they should be starting, stopping or keep doing.

  • matthewneedham

    Great post Ben.

    I get increasingly frustrated with the number of organisations that ask pointless questions – which give them no insight into what they should be starting, stopping or keep doing.

  • terrisorenson

    I find this post to be very informative and helpful when thinking abou the creation of an online survey. I do believe that length does matter in the online environment and that surveys should be kept to 15-20 minutes to reduce participant lag. Or maybe more creative questionairre design to keep respondents engaged. The survey tool does need a lot of thought and time before hand and I think you have outline the importance of this very nicely.

    As well, you commented “They lead me to think about new ideas, and also make me want to follow up with respondents to dig further. It’s just about staying open-minded to the possibility that you don’t know everything about everything (which is fairly easy in my case!)”

    You can always reach out to these thinkers after the quant study using online qualitative software such as a Online Focus Group or Online Bulletin Board.
    http://www.itracks.com/Products/BulletinBoardFG

    The intergration of quantitative and qualitative methodology can provide your clients with further insight.

  • terrisorenson

    I find this post to be very informative and helpful when thinking abou the creation of an online survey. I do believe that length does matter in the online environment and that surveys should be kept to 15-20 minutes to reduce participant lag. Or maybe more creative questionairre design to keep respondents engaged. The survey tool does need a lot of thought and time before hand and I think you have outline the importance of this very nicely.

    As well, you commented “They lead me to think about new ideas, and also make me want to follow up with respondents to dig further. It’s just about staying open-minded to the possibility that you don’t know everything about everything (which is fairly easy in my case!)”

    You can always reach out to these thinkers after the quant study using online qualitative software such as a Online Focus Group or Online Bulletin Board.
    http://www.itracks.com/Products/BulletinBoardFG

    The intergration of quantitative and qualitative methodology can provide your clients with further insight.

  • http://www.fidelitylocksmith.com/ Locksmith Services

    Hey Ben, Thats good post… It true that writing surveys is a great deal indeed and may factors need to be taken in to account in order to achieve desired result. Hope to see some more ideas on the same line i.e identify techniques in order to improve the end result of the survey.

  • http://www.fidelitylocksmith.com/ Locksmith Services

    Hey Ben, Thats good post… It true that writing surveys is a great deal indeed and may factors need to be taken in to account in order to achieve desired result. Hope to see some more ideas on the same line i.e identify techniques in order to improve the end result of the survey.

  • http://www.beblunt.com/ Jonathan Levitt

    Hey Ben. Great Post!
    #2. It's very important to know up front which metrics you'll want to trend over time. Changing survey questions can cause a loss of trend lines.
    #5. My Experience is that it's not about data points or number of questions, but about interactions by the respondent with the survey. Depending on how your survey is built, its important to make sure that you don't exceed 25 interactions with the screen (an interaction defined as a click, scroll, selection of drop down menu etc). We start to see exponential drop off after about 25 interactions.

  • http://www.beblunt.com/ Jonathan Levitt

    Hey Ben. Great Post!
    #2. It's very important to know up front which metrics you'll want to trend over time. Changing survey questions can cause a loss of trend lines.
    #5. My Experience is that it's not about data points or number of questions, but about interactions by the respondent with the survey. Depending on how your survey is built, its important to make sure that you don't exceed 25 interactions with the screen (an interaction defined as a click, scroll, selection of drop down menu etc). We start to see exponential drop off after about 25 interactions.

  • http://www.facebook.com/byosko Benjamin Yoskovitz

    Jonathan – Thank you for stopping by and commenting.

    I appreciate the specific points on #2 and #5 – very useful for me and others.

  • http://www.facebook.com/byosko Benjamin Yoskovitz

    Jonathan – Thank you for stopping by and commenting.

    I appreciate the specific points on #2 and #5 – very useful for me and others.

  • http://shearinglayers.com/ Nick Shepheard

    I think your third point is the killer.

    Asking questions that, whatever the answer, wouldn't change anything, is a waste of a good opportunity and good will. I've gone toe-to-toe on this point a few times with research people. My beef has always been that they like to ask “interesting” questions when what customers and company both need is something actionable.

  • http://shearinglayers.com/ Nick Shepheard

    I think your third point is the killer.

    Asking questions that, whatever the answer, wouldn't change anything, is a waste of a good opportunity and good will. I've gone toe-to-toe on this point a few times with research people. My beef has always been that they like to ask “interesting” questions when what customers and company both need is something actionable.

  • http://www.boxing-classes.net/ Enzo

    Thanks for the post, for all the efforts it takes to create a survey, it is definitely worth paying a bit more attention to the questions asked so that the ''output' data is actionable. For me, point number 3 above is an important reminder to everyone who wants to more their company/service forward on the back of a survey. Cheers.

  • http://www.boxing-classes.net/ Enzo

    Thanks for the post, for all the efforts it takes to create a survey, it is definitely worth paying a bit more attention to the questions asked so that the ''output' data is actionable. For me, point number 3 above is an important reminder to everyone who wants to more their company/service forward on the back of a survey. Cheers.

  • amyblake

    Putting the time into preparing a good survey can also yield a more interesting survey—which means more people will want to start (and, most importantly!) finish. I think asking tough questions is a great way to get buy-in. If you are clearly looking for answers that might give you negative feedback, the survey has more validity. And YES to point 7: open-ended questions that invite further comments.

  • amyblake

    Putting the time into preparing a good survey can also yield a more interesting survey—which means more people will want to start (and, most importantly!) finish. I think asking tough questions is a great way to get buy-in. If you are clearly looking for answers that might give you negative feedback, the survey has more validity. And YES to point 7: open-ended questions that invite further comments.

  • http://www.instigatorblog.com Benjamin Yoskovitz

    Amy – Thank you for stopping by and commenting, much appreciated.

  • http://www.instigatorblog.com Benjamin Yoskovitz

    Amy – Thank you for stopping by and commenting, much appreciated.

  • http://www.tape4backup.com/lto-2-tapes.php LTO-2

    Liked the points on Num # 2 and Num # 3.All the points are useful for me and for others too.

  • http://www.tape4backup.com/lto-2-tapes.php LTO-2

    Liked the points on Num # 2 and Num # 3.All the points are useful for me and for others too.

  • http://www.articlessubmissionservices.com/ sach

    wow! great article . i do agree with you when you said that you have to take time to write your survey, it really helps to do a research first before you write your survey, just because you need customer data does not mean that you have to rush. Do take your time and then write a great survey,

  • http://www.articlessubmissionservices.com/ sach

    wow! great article . i do agree with you when you said that you have to take time to write your survey, it really helps to do a research first before you write your survey, just because you need customer data does not mean that you have to rush. Do take your time and then write a great survey,

  • http://blog.paulriley.me.uk paul riley

    Stumbled across your blog by accident, I have been asked to add a survey to another site I run and hope your pointers will make the process that much more effective.

    Great Blog

  • http://blog.paulriley.me.uk paul riley

    Stumbled across your blog by accident, I have been asked to add a survey to another site I run and hope your pointers will make the process that much more effective.

    Great Blog

  • http://www.instigatorblog.com Benjamin Yoskovitz

    Paul – best of luck preparing and launching the survey. Please let all of us know how it goes after.

  • http://www.instigatorblog.com Benjamin Yoskovitz

    Paul – best of luck preparing and launching the survey. Please let all of us know how it goes after.

  • http://www.b2bfax.net/ DThomas

    Thanks for the understandable information concerning surveys. As a small company we are always looking for ways to interact with our clients and to know how best to serve them. Surveys are an option we are looking at. Thanks and great post.

  • http://www.b2bfax.net/ DThomas

    Thanks for the understandable information concerning surveys. As a small company we are always looking for ways to interact with our clients and to know how best to serve them. Surveys are an option we are looking at. Thanks and great post.

  • http://www.b2bfax.net/ DThomas

    Thanks for the understandable information concerning surveys. As a small company we are always looking for ways to interact with our clients and to know how best to serve them. Surveys are an option we are looking at. Thanks and great post.

  • http://www.lose-boobs.com/chest-fat.html Chest Fat

    i`ve always wanted to create a survey, but i couldn't find anyone to help me. Your article gave me a few valuable tips about what should i do. Thank you very much.

  • patrick5000

    nice article. i can't remember how many survey forms i've set up for clients. and they always lack in follow through. i'm more on the tech side and do webpages but can tell the clients don't put in the effort on the metrics and hardly ever change add or delete the questions on the form. i'm gonna forward your article to all my clients who do survey forms in their websites. Thanks.

  • http://www.autodialerforsalespeople.com pThomas

    nice article. i can't remember how many survey forms i've set up for clients. and they always lack in follow through. i'm more on the tech side and do webpages but can tell the clients don't put in the effort on the metrics and hardly ever change add or delete the questions on the form. i'm gonna forward your article to all my clients who do survey forms in their websites. Thanks.

  • http://www.instigatorblog.com Benjamin Yoskovitz

    Thanks for the kind words Patrick. I hope your customers find the post equally valuable.

  • http://www.instigatorblog.com Benjamin Yoskovitz

    Thanks for the kind words Patrick. I hope your customers find the post equally valuable.

  • anna

    hey, really great post:) I have a question.. I would like to write survey connected with blended learning for students, some questions for them to match-a,b or d:) maybe you know some webs where I can find how to write it , or with some examples of this kond of survey.. I need help:(

  • anna

    hey, really great post:) I have a question.. I would like to write survey connected with blended learning for students, some questions for them to match-a,b or d:) maybe you know some webs where I can find how to write it , or with some examples of this kond of survey.. I need help:(

  • http://www.instigatorblog.com Benjamin Yoskovitz

    Anna – I'm not sure where to find this kind of information. My first instinct would be to search Google using combinations of keywords that might get you the response you're looking for. I also know, if you do some digging, you can find some experts on Web surveys; and there may be some worthwhile connections to be made there. Good luck!

  • http://www.instigatorblog.com Benjamin Yoskovitz

    Anna – I'm not sure where to find this kind of information. My first instinct would be to search Google using combinations of keywords that might get you the response you're looking for. I also know, if you do some digging, you can find some experts on Web surveys; and there may be some worthwhile connections to be made there. Good luck!

  • anna

    ok:) I will try:) thank you very much.

  • anna

    ok:) I will try:) thank you very much.

  • http://www.instigatorblog.com Benjamin Yoskovitz

    Anna – I'm not sure where to find this kind of information. My first instinct would be to search Google using combinations of keywords that might get you the response you're looking for. I also know, if you do some digging, you can find some experts on Web surveys; and there may be some worthwhile connections to be made there. Good luck!

  • anna

    ok:) I will try:) thank you very much.

  • Pingback: Customer Validation Really Starts with In-Person Interviews

  • Pingback: Reminder: Don’t forget your audience!! | MarketingMaven GlobalSpec

  • Pingback: Can your user complete his task? | Stuart Silverstein – UX Strategy, Design And Process

  • Pingback: A Roundup of Survey Design Resources | actionable data

  • Pingback: A Roundup of Survey Design Resources (Cross-Post with Actionable Data) « Evaluspheric Perceptions

Ben Yoskovitz
I'm VP Product at GoInstant (acq. by Salesforce).

I'm also a Founding Partner at Year One Labs, an early stage accelerator in Montreal. Previously I founded Standout Jobs (and sold it).

My bio »

or

Follow me on Twitter

Get updates and special content
When I publish new content, get it directly in your inbox. Subscribers will get special stuff as well not available on the blog (but I promise it will be infrequent + high quality.)
Get the Lean Analytics Book!
Awesome Jobs
Check out the job opportunities at my portfolio companies.
Startup & Investor Resources
Find Stuff
My Photos