Customer Feedback: Persuasion and Usability Matter

Perhaps one of the most sensitive calls to action (aside from asking for credit card information) is the request for customer feedback (especially when the request is made through the intrusive and long-despised pop up window.)

You’re asking the time starved customer to generously donate time to fill in the survey for nothing in return. And asking for feedback within seconds of the customer’s arrival on site may mean a quick bounce.

Calls to feedback should be treated like any other call to action – with careful attention to persuasion (copy) and usability (design and delivery).

I’ve come across a lot of these ForeSee surveys lately:

If you can’t see the image, this is the English copy…

Thank you for visiting Best Buy Canada!

Upon leaving our website, you may be selected to take part in a customer satisfaction survey. This survey is conducted by an independent company, ForeSee Results.

The feedback obtained from this survey will help us to enhance our website. All results are strictly confidential.

…wIth 2 calls to action: “Continue” and “No Thanks”

Is this optimal? I can think of at least 3 ways to improve:

1. The copy could use a little massaging to warm it up.

“The feedback obtained from this survey will help us to enhance our website.”

How about moving from company-centric to customer-centric messaging? Perhaps “Your feedback is valuable and helps us understand how we can improve your shopping experience”?

2. Calls to action need clarity.

Does “Continue” mean “Continue Shopping?” or “Yes, I’d be happy to take your survey?” I’d prefer to see clearer labeling: “Okay” and “No Thanks.” A green “Okay” button and red “No Thanks” may help clarify visually.

Let’s face it, most people won’t even read the copy and will click a button to close. If a visitor click the wrong button in haste and receives a pop up when leaving your site, that customer is now annoyed at you and feels you have not respected his or her wishes or privacy.

3. Timing and selection.

“You may be selected” – why interrupt customers (and risk them fleeing your site in disgust) if they may not be selected? Why accost them on the first page they land on? I would recommend triggering this prompt for customers who have clicked at least 2 pages in addition to the landing page (browsing or otherwise engaging in the site). I say this as one who has opted in to several Foresee surveys (more out of analytical curiosity than altruism). Often I am faced with over 20 questions, and must select N/A for many of them because I never got far enough to look at a product image, read a description or interact with a shopping cart.

Interruptive pop ups are not the only way to listen to the “Voice of the Customer” (VOC), as Bryan Eisenberg points out there are at least 2 other methods including passive and action/behavior triggers that are worth exploring and testing.

Related Articles

16 Responses to “Customer Feedback: Persuasion and Usability Matter”

  1. Your point about “Calls to action need clarity” is excellent.

  2. great post, Linda. I think your second point is the most important.

    The first thing any visitor will do when something like that pops up is look at the buttons. Switching “Continue” for “Take Part In Our Survey” means they can just instinctively hit ‘yes’ or ‘no’ rather than be bothered by the rest.

  3. Denny says:

    Great Post! I think that making the message customer-centric provides a much stronger incentive to take the survey.

    Also, the estimated time needed to complete the survey would be worth mentioning. If the survey takes 5min to complete I wouldn’t mind taking it. I don’t want to spend 30min of my time on it though.

  4. Good points! Especially in light of the debate there was last week in the Yahoo! Web Analytics Forum, where many commentators said they were very annoyed with the invitation pop up. I think they were exagerating, but it’s true that anything that will optimized the rate of visitors who decide to take the survey is important.

  5. Thanks for all your comments, and I agree, there should be a time estimate also provided. Some of these surveys are 25 questions long!

  6. You came up with very interesting points in this post. I think the third tip is very important. A customer can not give their honest opinion on a site if they have yet to go through the first page. When optimizing online surveys the number of questions should be minimized. Anytime I click on a survey and see there are too many questions, I’ll up clicking the close button.

  7. Cynthia says:

    Thanks for breaking it down and serving up some great pointers!

  8. Eric Head says:

    Thank you very much for your critic and analysis of the ForeSee survey device – these are great points for us to take to heart!

    A couple of points to consider:
    1) The “welcome” text is completely configurable by the client; while we will make recommendations on wording, it is ultimately up to the client to write the copy. Also, many of our clients do in fact state expected time to complete…usually in the 3 minute range on average;
    2) The survey trigger options are also very configurable in terms of where and how the survey presents to the user; these also are controlled on the client side;
    3) The client has good flexibility in terms of number of total # of questions on the survey – we will have a blend of scientifically-based model questions + any custom questions the client desires; any given survey is typically 20-30 questions in length and our average response rate is 7-9%.

    Thank you again for initiating the dialog – please don’t hesitate to let me know if there are other questions on the matter.

    Eric Head
    ForeSee Results

  9. Great post, we have been thinking along the same lines with our website

    We have made a free survey using Survey Monkey and sent it to our subscribers and also linked to it from several pages including the contact us page.

    This is the wording I used on the link:
    “We value your opinions and would be grateful if you would complete our anonymous quick 7 question survey.” This addresses several friction points at once. Firstly will I get spammed?, secondly how long is this gonna take?, and thirdly to reinforce how quick it is I have listed how many questions there are?.

    Take a look, all feedback is valuable.


  10. Good article, and yes, I definitely agree that there are other ways to configure a survey. In many ways, many of the questions I’ve found on online surveys can be more easily (and effectively) dealt with via usability testing and A/B testing.

  11. @Eric,

    Thank you for the clarification :)


    I couldn’t find the Survey Monkey on your site but found it on your blog. I think the survey length is great and the Survey Monkey tool is fine, the only thing that concerns me is if I exit the survey I’m taken to a Survey Monkey landing page rather than return to your blog or site — is there a way to change that?

  12. The Survey Monkey survey doesn’t seem to give me the option of taking people back to my site if participants exit the survey. Perhaps that is an option for one of the paid membership packages.

    I have also put a link on the order conformation page to remind customers to take the survey after they make a purchase.

    • @Simon,

      Surveys like ForeSee use popups which are great because they don’t hijack folks from the site, (and often they are triggered after customer has left the site) but the tradeoff might be that people don’t like pop ups.

  13. Rob says:

    I would have to agree with Dan Barkers comment in regards to #2. Way too often visitors skim content, and don’t always go(with their eyes) where they are supposed to. Having every call to action as specific as possible can reduce a ton of headaches for customer service.

  14. pay careful attention to persuasion (copy) and usability (design and delivery). that is right.

  15. Nice. Thanks for posting this. It is always awesome to see someone educate the public.

Leave a Reply

© 2014 Get Elastic Ecommerce Blog. All rights reserved. Site Admin · Entries RSS · Comments RSS