Occam's Razor by Avinash Kaushik

4Q – The Best Online Survey For A Website, Yours Free!

100 User Comments

1. Joe Teixeira Says:
   March 4th, 2008 at 06:35

   This looks really really really really cool. I will work my way up the ladder and see if we can get this bad boy installed on our website.

   Is that "Yes or No" page text editable?

2. Beat Says:
   March 4th, 2008 at 06:48

   Well, I tried to sign up. All I get is: "We are currently experiencing technical difficulties.
   Please try again shortly."

3. Avinash Kaushik Says:
   March 4th, 2008 at 08:51

   Beat : Birthing pains!! :) Please try again. I just signed up again (for the blog this time!) and I was able to get it done. Thanks.

   Joe: I am thrilled you watched that far into the video. :)

   Now others will have to watch it as well in order to get the joke!

   Of course the text is editable.

   -Avinash.

4. Jonathan Levitt Says:
   March 4th, 2008 at 08:51

   Beat,

   Tried to reproduce the problem on this end – but to no avail. Can you tell me if you are still having a problem?

   Thanks.

   Jonathan

5. S.Hamel Says:
   March 4th, 2008 at 09:22

   Great idea! I registered and it's now on my blog (will also want to put it on wasp.immeria.net but it seems we are limited to one site).

   I also added 4Q to the list of tools that WASP can detect :)

6. Alex B Says:
   March 4th, 2008 at 09:32

   Avinish, this looks great and I'll implement it on a few sites within the month. For what it's worth, for those of us who have read your book and/or know what your opinions on analytics are, the video above could be shortened in half. I already know why you think surveys are useful, so for me it would have been much more helpful had you done a quick intro than simply started with explaining 4Q– perhaps two videos? :)

7. Tim Peter Says:
   March 4th, 2008 at 10:40

   Avinash,
   This is fantastic! Long overdue for someone to put this into the marketplace – and I'm thrilled to see iPerceptions do this with you. Nice job. I'll be posting this to my website over the next day or two and I'll let you know what I think.

8. Beat Says:
   March 4th, 2008 at 11:39

   Avinash, Jonathan,
   It's working fine now, thanks.

9. Jonathan Levitt Says:
   March 4th, 2008 at 11:41

   Beautiful. Thanks Beat. Enjoy.

10. Kat Says:
    March 4th, 2008 at 11:56

    Is there an option of adding an additional question? I would like to be able to distinguish between our representatives and our clients.

11. Dennis Gorelik Says:
    March 4th, 2008 at 13:52

    Avinash,

    If I install 4Q script it would ask exiting [from my web site] users if they want to fill the survey, right?
    What if user came to my web site second time?
    What if user opened several windows?

    Would it still ask user to complete the survey?

    In one of your videos I noticed that it's the source of your [minor] disagreements with Google — Google doesn't want to bother users too much with survey questions.

12. Will Kriski Says:
    March 4th, 2008 at 14:17

    This looks great. Thanks for providing this free tool! After implementing analytics I found that it doesn't really help that much to know what people intended to do or wanted to find.

    My only concern is the general dislike of popups but hopefully people will understand I'm doing it to improve their overall experience. I will be doing it for both my guitar lesson site http://willkriski.com and my IT consulting blog http://thewealthyworker.com

13. John W Ellis Says:
    March 4th, 2008 at 14:28

    It looks good.

    If I want to use this on multiple sites, do I need to create separate accounts?

14. Stacy Hyde Says:
    March 4th, 2008 at 22:53

    Avinash this is fantastic! Great work. I say this on behalf of all your readers that we appreciate all your efforts. You walk the talk.

    4Q is going up on our corporate site.

    Stacy

15. Anthony Says:
    March 4th, 2008 at 23:43

    Thanks Avinash, perfect timing, I was just trying to find a small & free onsite survey provider to implement your 3 questions.

    Is anyone else a little concerned about the way in which the invitation takes over the whole browser window? Wouldn't a smaller window over the top of the page be a better idea?

    Don't mean to criticise – after all, this is free!

    Thanks again Avinash.

    Anthony

16. Avinash Kaushik Says:
    March 5th, 2008 at 00:24

    Anthony : Aw come on, give us a break! Our baby is only a day old! :)

    On a serious note. . . .

    Good point on the invite Anthony (in addition to yours, I also have one other suggestion for improvement that I have sent to the team).

    I believe in putting things well built out there, learning, iterating and making things better with customer input.

    That is my hope with 4Q, that we are active listeners and your feedback will help make it perfect. In that sense you are a valuable part of the ecosystem, please keep all the feedback coming as it is deeply appreciated.

    John : Sadly you have to sign up for different accounts. I am sorry. This is one of the fixes the team will do, even I have five sites to sign up and don't want to have that many logins.

    It is one of the small compromises the team had to make.

    Will : This should present a dhtml layer to the visitors who land on your site and in as much won't be a pop-up. Then when they leave the site they'll be presented the survey.

    I am sorry it was not clear on the blog.

    Dennis : The survey will pop up when the Visitor closes all browser instances or leaves the website in all the instances. Ditto for tabs. I am getting the team to 1000% confirm this for me.

    The reference in the Authors@Google video about surveys was a question from the audience about using surveys on g.com. The purpose there is to get you off as fast as possible, hence the hesitation. But you'll find surveys on many Google sites, such as the AdWords help center.

    Kat : So sorry but no options to add any more questions.

    It was a tough choice to make but I personally wanted the survey hyper focussed and one of the trade offs was this one. There were alternative thoughts on this but I did not want more becuase of a bunch of philosophical questions. But this is a start and not then end. I am very open to learning.

    In your case you can segment by Primary Purpose and that could give you a feel for clients or company reps (assuming they don't use the site for the same purpose).

    As I had mentioned in the post, surveys are quite flexible and a great way to get much deeper into understanding the customer experience. There are great tools in the market that do that if your company has that need.

    Thanks everyone for the kindest of words, your support means the world to me. Thank you!

    -Avinash.

17. Zvika Jerbi Says:
    March 5th, 2008 at 04:29

    I tried it on S.Hamel blog… loved it. very simple. very claver.
    How does this tool work with Google Analytics? would it be considered a bounce and re-entry? wouldnt it harm the data (referrals)? Should we use filters for that? can we have the GA tracking code on the 4Q pages?

    Thanks avinash,
    Zvika

18. Zvika Jerbi Says:
    March 5th, 2008 at 04:33

    Clever. Not claver, of course!

19. angie Says:
    March 5th, 2008 at 05:01

    “Snort, snort, snort, snorty, snort."
    Best. Blog. Quote. Ever!

    Thanks for letting us know about the survey Avinash… This is a very exciting tool to have in our analytics toolbox (and the price couldn't be better). I look forward to trying it out!

20. Tim Leighton-Boyce Says:
    March 5th, 2008 at 06:17

    Thank you so much for this. I'm a huge fan of surveys, like you, and I've been looking for a good , simple and cheap way of doing this. Normally the rule is "pick any two" of those requirements, but in this case I think you meet all three.

    I've got it running on a test site now and I'm keen to put it on something more serious. But I agree with others that the "yes/no" screen needs some tweaking. Maybe making just working on the text would do the trick, or perhaps some way whereby the real site could still be seen behind would be even better. With the full screen version I'm scared that people might bounce.

    It's clear to me that you've created something which really fills a need here. I'm really impressed. Thank you.

21. Monica Valentinelli Says:
    March 5th, 2008 at 09:33

    Awesome stuff, I can't wait to try it out. Thank you for developing this useful survey.

    Are you interested in feedback once we implement it? Or are you set for commentary in that regard?

22. Jon Whitehead Says:
    March 5th, 2008 at 16:20

    Excellent, and how generous to provide for free! will give it a test and let you know how it goes

    cheers

    Jon

23. Manjula Says:
    March 5th, 2008 at 17:16

    Thank you! This is awesome! Appreciate that it is free.

    Manjula

24. MyTechnologyCompany.com » Free Simple Website Survey Tool Says:
    March 5th, 2008 at 18:33

    [...] When dealing with online marketing metrics, I am a big fan of Avinash Kaushik. One of his brilliant simple insights is that most marketers need to supplement their website metrics with a simple survey asked to customer as they depart the website. [...]

25. Steve Jackson Says:
    March 6th, 2008 at 00:43

    Hi Avinash,

    This is a really great service at first glance. Thanks very much for this great tool which I can see great uses for. I will have it tested immediately on a number of our client sites and will pass on any feedback to help you and your partners improve your service.

    Great stuff thanks once again.
    Regards
    Steve.

26. Chris Clapham Says:
    March 6th, 2008 at 02:01

    Hi Avinash,

    A great tool which I am now testing out on one of our websites – Thank you!

    I was just wondering if there was any scope in the future to provide stats in the results interface on the number of times the survey has been served, number of people that click yes vs. no and the number of people who click yes but then opt not to fill in the survey at the end?

    Obviously we know the number of visitors to our website and the percentage being served then we could work out the number of people served the survey but I think it would be nice to have access to this data and more in the 4q results interface? Just an idea :-D. Of course if there are compelling privacy reasons to not doing this then it would be good to hear.

    Thanks again!

    Chris

27. Tim Leighton-Boyce Says:
    March 6th, 2008 at 02:08

    I've been looking at the results of my test now, and one thing which strikes me is that it would be useful to be able to export the free text comments.

    On all the surveys I've done up to this point, I've found the 'voice of the visitor' to be a most illuminating source of insight.

    In my experience it helps to import this data into a database so that I can quickly run through the comments tagging them as being about certain topics. The topics are a self-defined group which are established over the first few weeks.

    One benefit of this is that you can bring some numbers (and trends!) back into the reports. For example "why did comments about log-in issues started to go up on Tuesday and then back down on Friday?" We've discovered real show-stopping intermittent problems that way, which we might not have spotted otherwise.

    Or perhaps you could add some kind of function like that to the system itself?

    I recognise that certainly the second suggestion is probably way beyond the scope of this free product. Maybe features like that could become part of a paid-for version? You deserve to get some revenue out of this, after all.

    Thanks again.

    Tim

28. Andy Breeding Says:
    March 6th, 2008 at 09:29

    I've been waiting for something like this! Is it possible to control the frequency at which a survey is issued (say to every Nth user)?

29. Andy Breeding Says:
    March 6th, 2008 at 09:33

    You may disregard the last question I just read the FAQ–and I see that the answer is YES!

30. Luis Says:
    March 6th, 2008 at 10:50

    This is great Avinash! One question: is there a benchmark for how many survey responses it takes to be actionable? In other words, should we try to capture a certain amount (i.e. 1% of unique visits/month) before we have an acceptable sample size?

    Thanks,
    Luis

31. Michael Says:
    March 6th, 2008 at 11:46

    Wowee…already getting super actionable feedback – right from the customers.

    This is kind of value that makes free the best price of all. Thanks for contributing way above the waterline Avinash.

32. Mike Thompson Says:
    March 6th, 2008 at 15:47

    Hi Avinash,

    Thanks for creating this tool along with iPerceptions, we're already getting some interesting results.

    I much prefer the openness and transparent feel of the invitation taking up the whole browser, as compared with the other surveys I see which try and glide into view inconspicuously, but end up being much more invasive!

    It would be great to be able to customise the answers to the "what is the purpose behind your visit?" question, and to see what proportion of people who say they'll complete the survey actually don't.

    Additionally and much more generally, has there been any research done on whether you can expect the sort of people who agree to do a survey to be a representative sample of your user-base?

    Thanks again: it's great to be able to get such immediate, human feedback about a website!

    Mike

33. Avinash Kaushik Says:
    March 6th, 2008 at 22:40

    Thanks for the feedback everyone! One more round of answers. . . .

    Zvika : It should have no impact on Google Analytics, the page still loads and executes the GA code and yes or no on the survey invite has no issue. Remember the survey is just a dhtml layer on the page. So no need to use any filters etc.

    Tim : The feedback is welcome and great, the team is looking very carefully at the invite and expect improvements based on all customer feedback! On your second point, yes the team is working on a report that will show you Response Rate (both how many survey invitations were served and how many completed). Also the ability to export the data and open text will come down the road.

    Both of these were part of the plan but needed to stay back for launch. No plans for a paid version, this is and will remain free. It will be focussed and solve a important problem, more more robust / complicated / expansive / moon shattering versions are actually on the market already.

    Monica : We crave feedback (after all it would be odd to champion Voice of Customer and Customer Satisfaction and not be sold on it ourselves!!!). Please share your feedback, you can also email 4q at iperceptions dot com.

    Chris : On its way, soon. IMP: No personally identifiable information is captured by the survey. It is 100% anonymous. I wanted to stress that, and also of course hence no issues sharing the response rate.

    Typical response rate for a survey on the web is 1%. The response rate trends will allow you to see how you are doing.

    Luis : You have a slightly complex question, but a good one. Here is a table that answers the question, from our good friend Duff from iPerceptions:

    The following table should be referenced in regards to the satisfaction rating out of 100.

    Statistical significance is based on 3 variables, standard deviation, number of data points (respondents in this case), and confidence level.

    The iPerceptions 0-10 point scale provides an average STDDEV of ~2.0, with this assumption that following table shows confidence intervals that are significant 9 times out of 10 (90%) and 19 times out of 20 (95%).

    

    Here's the English:

    The table shows that the margin of error for survey results decreases as the respondent count increases. If you want a reasonable margin of error (say, +- 2%) at an acceptable confidence level (90%), then you need to collect about 300 respondents for a given period. A higher respondent count will yield a lower margin of error, but diminishing returns will set in at some point. You can still obtain good directional data, however, from less than 300 respondents.

    So it is important to know that you don't need to survey everyone to get good data. I also want to add that no matter how many responses you get remember that open text VOC is one of the keys here with 4Q, there every single response is important.

    Michael : I wish I could tell you how happy this makes me! Sweet redemption! :)

    Mike : Thanks for the feedback, much appreciated.

    The more customization the more there is complexity in the survey and backend etc and in turn cost. Hence I was trying to keep things simple. But I want this to be a true customer driven innovation and in as much I am very open, as is the team, to listening. We'll try to see if customization of the question framing can be made to work (I can see how across different cultures etc this could be a issue).

    Having done surveys now across atleast 100 different websites I am astonished at how representative survey respondents are, after all for example if everyone who hated you filled out a survey you would not get a score of 8 out of 10 for customer satisfaction.

    In terms of checking for survey bias:

    My read on any bias is to look at the Primary Purpose data and get a feel for the distribution (you can't be upsetting everyone, and one bucket typically won't dominate).

    I also look at trends over time, point reads are not good. Trends help negate any bias effects (so even if everyone who loves you or hates you takes the survey you want to look at the trend over time, and if it is improving or not).

    Finally I also correlate the survey data, atleast at simple high levels with my web analytics data (so are people really looking at more product pages or support pages etc etc, do they really hate the site but look the bounce rate is only 6%, or hmm these are the keywords driving people to my site and the intent behind those keywords ("avinash kaushik rocks") is reflected on the survey (high number of people choose primary purpose "download avinash's latest video because he rocks"). :)

    Oh and remember that perhaps the most core value of 4Q is your ability to tap into the VOC and there bias or no bias those are your customer's own words and every single one of those is of value! :)

    -Avinash.

34. Kyle M. Brown Says:
    March 6th, 2008 at 22:59

    This is outstanding! Ive signed up and plan to place this on one of our clients ecommerce properties to kick the tires.Thanks.

    Regards,
    Kyle

35. Small Business E-commerce Link Digest – March 7, 2008 Says:
    March 7th, 2008 at 09:22

    [...] I would recommend that you take a good look at Google’s policy and decide if it’s right for you. I would also suggest relying on a second source for analysis, if you can manage it, just in case Michael’s “Say it ain’t so, Joe,” scenario plays out. For instance, though it’s not the same type of tool as GA, you should take a look at Q4, a new qualitative survey tool from Avinash Kaushik and iPerceptions. Knowing why your customers choose to do what they do is often better than knowing what they do. [...]

36. PPC Advertising Roundup for Monday through Thursday 3/3/08 – 3/6/08 | semvironment Says:
    March 7th, 2008 at 12:25

    [...] 4Q – The Best Online Survey For A Website, Yours Free! [...]

37. BryGuy Says:
    March 7th, 2008 at 16:11

    Outstanding job, Avinash. After digging through several survey responses a few thoughts occurred.
    It would be extremely helpful to segment responses by keyword search. Example: we set the survey to ONLY be shown when someone clicks on a specific Adword ad. The add itself is designed to draw in visitors who are late in the buying stage. Example, the ad shows "Buy XXXXX Now!" as the title followed by additional persuasive lines of text. The results were not as expected. We found that 65% of those who clicked on the ad were "looking for product information". Only 17% were actually visiting the site to buy.

    If this feature could be added I think it would add a tremedous amount of value to the service.

    Thanks again for sharing with the community. Greatly appreciated.

38. Richard Hayes Says:
    March 8th, 2008 at 00:21

    Hi, the 'survey code' page does not display any code.

    Thanks, Richard

39. Mike Says:
    March 8th, 2008 at 08:40

    Same problem – I got the email that said we were all set up and to sign back in to get the code, but the "survey code" page is not displaying anything at the bottom.

40. Jose Davila Says:
    March 8th, 2008 at 15:38

    Hi Avinash,

    Thank you for such a useful tool. I will definitely be using it at several projects I am working on right now. However, I noticed that the survey is for now only available in English. That's why I am offering to you my help to translate the whole survey to Spanish. It happens I am a native Spanish speaker, with strong Spanish writing skills and experience in translation and copy writing. Also as my career development is now closely related to the web analytics and web intelligence field, writing surveys is something I definitely enjoy . I will be glad to contribute to this great initiative of yours. Please let me know if you are interested

    Thank you again and all the best.

41. Jonathan Levitt Says:
    March 8th, 2008 at 22:36

    Richard & Mike: We found a corner case where some people were not able to see the get survey code, we have fixed that now so all should be well. Could you please try again, it should be there now. Apologies for the inconvenience.

42. Web Analytics Bloglinks der Woche (KW 10 / 2008) | Web Analytics & SEO – News Says:
    March 8th, 2008 at 23:04

    [...] 4Q – The Best Online Survey For A Website, Yours Free! (Avinash Kaushik) [...]

43. Richard Hayes Says:
    March 9th, 2008 at 11:08

    Hi, thanks for this great bit of survey software. Would it be possible to have a popup on exit instead of on entrance? And therefore adding survey code to every page…

    Many thanks,

44. Shelby Says:
    March 10th, 2008 at 06:18

    Thanks Avanish! This is fantastic. I have a quick question about survey applications in general.

    What do you see are the advantages and disadvantages of "survey" type applications (like iPerceptions, ForeSee, etc.) compared to "feedback link" type applications (OpinionLab, etc.)? I've heard opinions on either side.

    … although I suppose it's better just to have something, anything qualitative, whether it's a survey or a feedback link. : )

    Thanks!

45. Michael Whitehouse Says:
    March 10th, 2008 at 07:25

    Shelby: Just wanted to jump in on this one. From my talks with clients who've used both Opinion Lab-type studies and iPerceptions solutions, they've outlined three main differences.

    1) Response rate: The more passive approach employed by the "feedback link" studies tend to yield lower (sometimes, sub 1%) response rates, because the call-to-action is often below the fold, sometimes even way down in the footer, where only the bold venture. Whereas, with our up-front solicitation, the mean response rate is the area of 6%-8%.

    2) Representative-ness of feedback: Because the "feedback link" studies are opt-in, they tend to attract a very polarized response set. People will seek out this link if they are very upset or it they are overflowing with goodwill and want to share it. A randomized invitation will capture a better cross-section of the online population.

    3) Page-level vs. site-level: Feedback link studies tend to provide very solid and actionable feedback on the content/functionality/usability of specific pages. This granularity is their big advantage. Conversely, survey-type studies capture the big picture: how is the site working as a cohesive unit to help visitors complete their tasks, to build brand affinity, and to drive successful outcomes.

    Sorry for the long-ish response. Hope this helps!

    Best,

    Michael
    iPerceptions

46. Shelby Says:
    March 10th, 2008 at 09:17

    Thanks for the response, Michael. That definitely helps.

    Is there a general threshold that you try to stay behind as far as survey length before you start finding data integrity issues? For instance, after *n* screens of questions, users tend to get bored with the questions and just put in any answer to get through it?

    Is there a general rule of thumb besides just *shorter is better?*

    I know this is a little off the topic of 4q, but I thought I'd ask.

    Thanks!

47. Richard Hayes Says:
    March 10th, 2008 at 11:24

    It would me good to have your company logo maintained in the uploaded format. It appears .png and .gif are converted to .jpeg which causes distortion – images with only a few colours are not suitable for .jpeg format.

    Many thanks, Richard

48. Mike Sukmanowsky Says:
    March 10th, 2008 at 12:01

    Very very cool Avinash! The instant you put this up I switched over from SurveyMonkey on my blog :).

    The dashboard is great to read too – only request is a custom calendar control please! I'd love to be able to take the satisfaction score and dump it in here

    All the best,
    Mike

49. Travis Cline Says:
    March 10th, 2008 at 12:33

    It'd be nice if they provided callbacks for

    - if they're redirected to the survey page
    - if they hit yet
    - if they hit no

    You could then do ga tie-in.

50. Travis Cline Says:
    March 10th, 2008 at 12:34

    yes*

51. Avinash Kaushik Says:
    March 10th, 2008 at 20:39

    Everyone: The forum is up! Please visit: http://iperceptions4q.com/forum/

    It is the optimal way for you to get support, ask other questions, make suggestions and generally have a ball with the community. Please check it out.

    Even though it is just a week since launch we have already made some improvements!

    + The survey invite is slightly more eye pleasing.

    + We have added the ability to edit your account (so you can change most of the stuff now).

    + We are going to allow you to update/change your logo.

    + There is better reporting of open text VOC (the report shows you the segment you have selected, doh! :).

    More ability to edits are on their way, as are more languages (there has been a huge demand for this!), and we are looking to do more on the invite (this is my personal passion).

    All your suggestions are excellent, the team and I truly want this to be customer driven innovation so please keep it coming.

    This is a 100% free no strings attached, no bugging involved, faith based initiative. You can expect us to keep innovating at a nice clip, but we do request your patience. We'll balance our costs while delivering high value to you.

    BryGuy :I am glad you found the survey to be valuable. This is a great example of how your conversion might not be high and without the survey you might only be able to guess why people were not buying!

    Segmenting by keyword is more complex than might be apparent, it would need merging with your web analytics tool (itself a non trivial task). But it is something that is on my mind and perhaps in the future there might be a easy way to do that.

    Thanks so much for sharing the story with all of us!

    Shelby:I touched on Page and Site level surveys here:

    Eight Tips For Choosing An Online Survey Provider

    Part One, point number 3.

    Page level surveys solve for something Site level can't, so each has it own unique use. But one thing is for sure that they are not interchangeable.

    So call 'em, stress test them with questions I provide in the above post, then make up your mind. :)

    On your second point, survey length it really depends on the survey experience (survey presentation, actual look and feel, number of questions).

    If you do a pop up in the middle of the customer session with a long list of 30 questions you might as well kiss that customer good bye. But if you show the survey no exit, like 4Q, and have a more ajax driven experience, like 4q, (for example no need to click next, once you answer a question the next question automatically shows up) then you can actually show as many as 25 questions and get a very high response rate.

    Because the customer experience is good (both the invite and taking the survey itself).

    There is some drop off observed in studies after 25, though if you target the right audience (people love to talk!) then you can do up to 40, though between 25 and 40 we observe drop off.

    Is it not astonishing how patient people are when presented with the right experience? I was personally shocked! But after hundereds of these studies my initial doubts have been proven wrong (though I stay under 25 (!) and make it a pleasant experience, i.e. no questions puking).

    Mike: Duly noted.

    PS: I absolutely adore your dashboard! Excellent effort, very impressive!!

    Travis: By callback you mean how many people saw the survey and choose yes or no?

    If that is the case…

    We are planning to add a response rate report that will show you some of this data. It is already on the to do list so please hang in tight.

    Thanks everyone, and enjoy the ride!

    -Avinash.

52. John Hossack Says:
    March 11th, 2008 at 15:25

    Great tool. We had been looking for such a tool for a little while now and this was just what we need at this point. It was easy and quick to set up and data that can be reviewed within hours. My question is (and perhaps I missed it somewhere) does the reporting provide data on the ratio of visitors who were offered to complete the survey vs. the number that actually completed the survey. I'm trying to gage how the request to complete the survey is being received. I'm a little worried about asking more people to fill it out if it appears that many of the visitors are not offering to complete the survey (perhaps because they are annoyed by it). I would like to ask more people to complete the survey, but not at the expense of hurting the user experience.

53. Measuring customer satisfaction: the Avinash Kaushik interview Says:
    March 12th, 2008 at 01:09

    [...] I sat down with Avinash (well, emailed, but you get the point) to talk about qualtitative analysis and how site owners can make the most of 4Q. We also talked a bit about the future of the tool and how it came to be. [...]

54. Kelly Kubrick Says:
    March 12th, 2008 at 07:58

    Avinash – this is great – thank you! I have several clients in need of this kind of VOC application – can't wait to try it on my own site to see if it might work for them.

    best,
    Kelly

55. AdWords-Articles.com Says:
    March 12th, 2008 at 19:33

    Short and free online website survey by Googles analytics evangelist…

    Avinash and Iperceptions have launched a new type of survey aimed at answering important questions efficiently. Did I mention that its free.
    A true permission based on-exit survey that provides an easy to deploy, easy to use and easier still to analyze…

56. Judd Exley Says:
    March 12th, 2008 at 23:38

    Hey buddy,

    You continue to make this here e-world a better place. Good on ya.

    I'm all inspired to come up with something on my own now. Something original… or at least really easy to steal. Hahhaha…

    Good luck man.

57. Jorge Cunha Says:
    March 15th, 2008 at 16:02

    Very good post and very useful about what really matter's and again simple.

    Best Regards From Portugal

    Jorge Cunha

58. Ron Spinner Says:
    March 19th, 2008 at 01:26

    Hi Avinash,
    Was wondering if you know what the impact of this survey is on organic ratings in the search engines. Since it is a DHTML it alters the text of the page (or is it a graphic?) which could impact this if the spider came when the survey goes on.

59. Avinash Kaushik Says:
    March 24th, 2008 at 16:58

    Ron: The survey uses JavaScript to trigger the invitation and subsequent activity. Search bots / spiders don't execute JavaScript and hence there should be absolutely no impact of any sort on robots / organic rankings / that type of stuff.

    Judd: I think "web 2.0" for stealing is called borrowing! :) Borrow away!

    -Avinash.

60. Thomas Faivre-Duboz Says:
    March 26th, 2008 at 03:01

    Hello Avinash,

    Too bad it is not currently available in french. I would be very interested in being able to translate it for you… Can you e-mail me if you are interested ?

    Thanks a lot Avinash

    Thomas

61. 4Q – Free online survey « University Usability Says:
    April 2nd, 2008 at 10:30

    [...]
    Avinash Kaushik blogs about 4Q, a new free online survey tool he has put together with iperceptions. The ‘best survey questions ever’, which 4Q uses, were the inspiration for the static (i.e. non popup) survey we have on the english pages of http://www.lu.se/english.

    The video at his post gives a compelling argument of why the data this kind of survey produces – the voice of the customer- is so valuable for showing what people are actually trying to do on your website.
    [...]

62. 22 Great Web Analytics Videos Says:
    April 9th, 2008 at 18:32

    [...] 4Q Survey tool Video all about a great new free survey tool http://www.kaushik.net/avinash/2008/03/4q-the-best-online-survey-for-a-website-yours-free.html [...]

63. In search of the perfect question | Tracking Tourism: The Tourism Research Blog Says:
    April 20th, 2008 at 06:37

    [...] – you might also be interested in this post by Avinash Kaushik, about 4Q, a new permission based on-exit survey that provides an easy to deploy framework to answer 4 questions that no website owner can live without. Readers of this article may find those questions somewhat familiar! [...]

64. Ross Dunn Says:
    May 15th, 2008 at 10:50

    Thank you so much Avinash and the iPerceptions team! I am pumped and I am enjoying 4Q's valuable feedback immensely. Here is an article I wrote for Search Engine Guide passing on my praise of 4Q to all who care read it: Free Survey Tool Fuels Marketing Insight – 4Q

65. Jonas Says:
    June 5th, 2008 at 11:40

    Very cool. Maybe I missed it, but, what triggers the survey on any particular page that it is installed on? Is there a website I could visit to see it in action? THanks.

66. links for 2008-06-06 « The Adventures of Geekgirl Says:
    June 5th, 2008 at 18:49

    [...] 4Q – The Best Online Survey For A Website, Yours Free! | Occam's Razor by Avinash Kaushik (tags: analytics internet online research survey statistics webanalytics) [...]

67. Mario Bilotas Says:
    June 18th, 2008 at 11:49

    Hi Avinash,

    Always enjoy reading your posts, and learning new technologies, latest trends and more. Being that I work with OpinionLab, and have gone through some of the comments, I recognize the many positive attributes 4Q presents, in the VoC category that our organizations share, and yet I find it that it should not be a matter of choosing page specific versus site wide, as it would be equivalent to choosing a car radio over having speakers. Each suits its own purpose, reveals different metrics, and although both ask a user to provide insight, both are approached differently by the users.

    When I speak with prospective and current clients I recognize the placement and value thereof, having a page-specific, user-invoked (non-intrusive) capability to bring the "why" of analytics on to a page level site quality monitoring and feedback by a user to hone in on real-time issues pertaining to that specific page. (a frustrated user will lose respect for a credit card company when the link to receive a 0% credit card is not working, and unless there is a quick way to rush a red flag to that location, the next 1000 people may also walk away)
    Survey tools (such as 4Q) benefit analysts and marketers in capturing overall feeling of their experience, help determine WOM, or NPS and gain that high level overview of what their experience was upon finalizing the tasks on a site-wide level.

    The core differences between the two are:
    • Location of feedback source
    • Approach to requesting feedback
    • Metrics provided by the feedback given
    • Purpose and Action of feedback gathered
    • Project (Survey) versus Ongoing (OL)

    You and I both know, as users go through web pages in lighting speeds, capturing them on a problem page, or site wide market research is priceless but each (iP & OL) offers core benefits the other do not. I see marketers gather data via specific survey platforms and make business decisions based on the end results. I also see usability and customer insight making site management decisions in a quicker motion to ensure revenue is not lost from page specific feedback. Each alone is great, but having both is awesome.

    As much as I enjoy competition, I have to say that comparing the two for the same project means either we as organizations have failed to voice our core capabilities, or the marketers are unsure of what results they are exploring. In the end, I believe both applications can continue to grow some of the most customer-centric platforms in the world, and get customers from talking around you, to directly with you.

    …sorry for the long explanation, and hope you are doing well. Almost done reading your book…good stuff. Let me know if you have Q: mbilotas@opinionlab.com

68. OX2 Web Analytics blog » Blog Archive » Yes, but how does it make you feel? Adding qualitative insight to your web analytics approach Says:
    September 5th, 2008 at 05:19

    [...]
    The most recent entry into the field is by our friends at iPerceptions and Avinash Kaushik, widely known as “4Q”. The simple and free service works as a permission-based pop-up on your web site, prompting visitors to think about whether they are satisfied, their purpose for visiting, if their task was accomplished, and if they have other feedback to offer you. These really are the fundamentals, and it’s good to see such a clear distillation of what makes a web site visit successful, without oversimplification or sacrificing open-ended feedback.

    One goal of 4Q is to show the tremendous value potential in survey analysis of your web site visitors, and guide you into unlocking it further with the iPSI, the iPerceptions Satisfaction Index, where you can first ask more detailed custom questions, and also compare results against benchmarks for your industry.
    [...]

69. laurent burman Says:
    September 22nd, 2008 at 12:45

    has anyone compared iperceptions to using a tool like poll daddy (also free) to try to capture this information on exit? also, are there any free tools that enable the on-page type survey that opinion labs offers? thx

70. Avinash Kaushik Says:
    September 22nd, 2008 at 12:57

    Laurent : I always encourage folks not to mix "page level" surveys with "site level" surveys. The vendors try to say they are interchangeable but in reality they are not – each serves a very different purpose.

    In this post I had covered the differences:

    Eight Tips For Choosing An Online Survey Provider

    Part One, point number 3.

    In terms of free alternatives to paid page level surveys, please check out Kampyle:

    http://www.kampyle.com/

    My friend Stephane seems to really like them!

    -Avinash.

71. Mike C. Says:
    November 18th, 2008 at 15:45

    Thank you for 4Q. This is a real time-saver, and a great way to open up the conversation with the people who use your website!

    Question:
    Is it possible to preview my survey before launching it? I was able to see it once, but there were a few display issues and I need to review/preview the survey at least another two times before it goes to launch.

    Thank you!

72. Tara Says:
    November 26th, 2008 at 06:45

    We were going to implement this on both our English and Japanese websites, but unfortunately the Japanese text which appears is actually good enough for us to use. This is a real shame since I would like to be able to gather and compare data from both sites.

73. Sunny Says:
    December 7th, 2008 at 10:19

    Avinash,

    The tool is wonderful! Simple, sharp and to the point. On another note, I realized that there was no way to implement it for my iWeb business site. Do you happen to know any web developers/programmers who have written a plugin to do so (perhaps using Automator)?

    Keep up the fine work,
    Sunny Lam

    http://www.sunnylam.ca || http://www.linkedin.com/in/sunnylam || http://twitter.com/sunnylam

74. 3 Easy & Affordable Ways to Put Your Website to Work | catchthis.ca Says:
    December 10th, 2008 at 19:08

    [...]
    There are many, many survey tools available. My current favorite is a nifty and free little survey tool by 4Q.

    It is super easy to sign up for, simple to setup, and once you’ve personalized it all that is needed to implement it on your site is a small piece of code. I could go into greater detail, but why would I when there is an excellent (and entertaining) video about the power of using survey tools on your website here: http://www.kaushik.net/avinash/2008/03/4q-the-best-online-survey-for-a-website-yours-free.html
    [...]

75. Tara Says:
    December 11th, 2008 at 02:41

    As a follow up to my previous post, the Japanese translations have been improved and we will be implementing it soon! Impressed with the speed of response!

76. Tara Says:
    December 12th, 2008 at 03:04

    Whilst we love to implement useful and informative features to our website, we are mindful about not disrupting our existing set-up.

    We started with this questionnaire by setting a 25% exposure rate for a set number of hours to be able to monitor any possible changes to our stats. This has been done 2 times now with great feedback. However, we have noticed a decrease in our bounce rate during these periods by approximately 15%. It seems that when the questionnaire runs, the desired page is loaded in the background and then when the visitor clicks on the thickbox window to say yes or no to the survey, the page reloads.

    When I go to 4Q.iperceptions.com, this does not appear to have the same behaviour? Any ideas? (I have also sent this to iperceptions)

77. Michael Whitehouse Says:
    December 12th, 2008 at 08:05

    Hi Tara,

    We've seen several instances of this before. One issue that can arise with the 4Q layered invitation pertains to websites running Flash content. Flash objects and DHTML elements (i.e. the survey invitation) are classified as windowed and windowless elements, respectively, and follow their own z-index ordering, with the windowed elements always on top. As such, if Flash content is loaded concurrently with our DHTML invitation, some wonky things start happening, as Flash objects get written on top of the DHTML elements.

    The simple workaround we came up with for 4Q is to temporarily hide the Flash objects on the page and then perform a page re-load after the invitation closes.

    I'm not sure if this is in fact your issue, but it is a plausible explanation for the deflation of your site bounce rate.

    Please don't hesitate to contact me directly if you would like to chat further. I'm at mwhitehouse @ iperceptions.com

    Thanks!

78. Ginger Says:
    January 20th, 2009 at 16:25

    This is a great video!

    I was wondering if your survey every visitor to a site or what are the best practices?

79. Help Your Small Business Customers Feel Smart: Search Engine Strategies, NY | Web 2.0 Internet Marketing Says:
    March 29th, 2009 at 08:50

    [...] A free tool to help you understand why visitors do what they do on your web site is 4Qsurveys. With this application you can study tasks. Did they complete them? Use this to get feedback. [...]

80. jim Says:
    March 29th, 2009 at 17:35

    I would love to try this immediately but have been unable to get a response in order to activate..

81. Sarah-Jane Morris Says:
    March 30th, 2009 at 07:55

    Hi Jim,

    This is Sarah-Jane, Community Manager at 4Q. Would you be able to send me the email address you used to sign up to 4Q and I'll check on the status of your account? Send it to 4Q at iperceptions dot com.

    Thanks Jim!

    Sarah-Jane Morris
    4Q Community Manager
    http://community.4Qsurvey.com

82. eTc :: El blog de Marketing en Español » Blog Archive » Escuchar la voz del usuario, ventaja segura Says:
    April 29th, 2009 at 07:14

    [...]
    Finalmente, las encuestas online. En mi opinión, funcionan. Puede darnos miedo no llegar a tener una muestra representativa, pero a mi entender los comentarios de los clientes son impagables y y pueden ponernos sobre la pista de lo que necesitamos mejorar. Yo recomiendo una gratuita que desarrollaron IPerceptions y Avinash Kaushik llamada 4Q. Me gusta porque son pocas preguntas pero fundamentales que van más allá de que nos digan lo que les gusta y lo que no, ya que podremos recoger información sobre lo que les ha frustrado, lo que no han encontrado o sobre lo que les ha encantado y nos diferencia del resto.
    [...]

83. Nicole Williams Says:
    April 30th, 2009 at 21:31

    I have a new website and it is important to make it interactive with members and guests that visit. It sounds like you might be able to help me better understand people who stop by. An online community is an on going art. I look forward to learning more…

84. Panu Kuuluvainen Says:
    May 6th, 2009 at 05:17

    Hey,

    thanks for a great app! I'm going to try it on my nutrition site, I've always wanted to know if people are there to buy or just to find information on sports nutrition :).

    Any chance of getting this in Finnish at some point? I mean come on, Swedish, Dutch, Norwegian, Danish… could we add Finnish to the list :)? I can help translate!

85. Paras internet-kysely koskaan, julkaisijat kuulolle! « Kuuluvainen Says:
    May 6th, 2009 at 05:40

    [...]
    Tässä herra Kaushikin esitys asiasta – 4Q – The Best Online Survey For A Website, Yours Free!

    Mielestäni työkalu kuullostaa todella kätevältä, mutta saa nähdä, mitä kävijäni ovat mieltä. Laitan kyselyn näyttöfrekvenssin 90%-100% välille ja tarkkailen Analyticsin avulla miten kävijät reagoivat.
    [...]

86. Sky Says:
    June 7th, 2009 at 22:52

    I think if the content you offered is important or interested enough, people will be willing to spend time on the surveys that they have to answer ahead.

    For example, if you are offering online video clips people are keen to watch online or download, it will make better motivation.

87. Nevin Says:
    June 7th, 2009 at 22:54

    The post is very informative. Thanks a lot for sharing it online.
    Keep working!

88. Help Your Small Business Customers Feel Smart: Search Engine Strategies, NY Says:
    June 14th, 2009 at 23:10

    [...]
    A free tool to help you understand why visitors do what they do on your web site is 4Qsurveys. With this application you can study tasks. Did they complete them? Use this to get feedback.
    [...]

89. Matt Says:
    July 7th, 2009 at 17:31

    Great article. Thanks, Avinash.

    I would like to provide my customers with an incentive to take the survey. Is there a way to include text describing the incentive? Or maybe send out an email after the survey is complete?

    Thanks.

90. Irfan Malik Says:
    August 28th, 2009 at 18:54

    Its really really very cool stuff. I would try it.

91. panu Kuuluvainen Says:
    October 15th, 2009 at 04:45

    Hey, I was wondering if it is possible to translate this wonderfull survey :).

    Please check my post on your forums:
    http://community.4qsurvey.com/forum/topics/4q-in-finnish

92. Scott Lum Says:
    November 9th, 2009 at 13:29

    This looks very useful. Is there a way to display how many users chose to take the exit survey and how many did not?

93. Rene Says:
    December 29th, 2009 at 04:40

    Seems like a powerful little tool to me. I'm in the midst of developing a website, and as a researcher, the whole idea of testing and analyzing your website using surveys and click analytics etcetera really appeals to me. However, what I have not seen so far is how to link the survey results to click rates. For instance, how many persons landing on your page from the keyword 'dog' are actually finding what they were looking for? Are there any tools available to link the survey outcomes to say, the google analytics results? I haven't seen this so far, but I'm really new to the field…

94. Franklin Says:
    January 13th, 2010 at 04:20

    Excellent read, I just passed this onto a colleague who was doing a little research on that. And he actually bought me lunch because I found it for him smile So let me rephrase that.

95. Nestor Says:
    January 28th, 2010 at 11:13

    Hi,
    We are trying to implement it our site (it's a really big one) but the architect says it's too dangerous to have a third party script downloaded in home page. What should I be able to say to our architect?

    Thanks,
    Néstor.

96. Virtual Merchant Today » Blog Archive » Why Oh Why? Says:
    May 4th, 2010 at 16:34

    [...] by Web analytics consultant Avinash Kaushik and implemented by iPerceptions, 4Q is a quick and simple survey, offered to visitors, with their [...]

97. How to use KPI’s | Sangat Pedas Says:
    November 13th, 2010 at 06:18

    [...]
    If you were not able to complete your task today, why not?
    Primary Purpose. Task Completion Rate. Segments of Discontent .

    Some good stuff about doing an exit survey with for instance 4q can be found on kaushik.net
    [...]

98. Loggenbloggen » Blog Archive » Webbenkäter är inte svaret på allt Says:
    November 16th, 2010 at 08:27

    [...]
    Kort går metoden ut på att lägga ut tre frågor som besökarna får svara på:

    What is the purpose of your visit to our website today?
    Were you able to complete your task today?
    If you were not able to complete your task today, why not?
    Läs mer på Kaushik egna webbplats: http://www.kaushik.net/avinash/2008/03/4q-the-best-online-survey-for-a-website-yours-free.html

    Jag har inte testat metoden själv men tror att den fungerar utmärkt för löpande utvärdering och analys av en webbplats. För det är ju så med webbanalys. Det är något jag tycker man ska göra löpande som en del i förvaltningen, inte som enstaka nedslag då och då.
    [...]

99. Rohan Bhargava Says:
    May 29th, 2011 at 03:39

    Hello, This goes out to Avinash and others on this forum. I think implementing a survey is a great idea, but how do you get it from not being a little annoying. Ie, I exit from the site and now I have another window pop-up asking me for a survey.

    I am assuming thats what happens with an on-exit survey right?

    Would love to hear from anyone who has implemented this and would love to see this work in practice if there are any suggestions.

100. Avinash Kaushik Says:
     May 31st, 2011 at 11:46

     Rohan: A pop-up window in the middle of someone's session with 37 questions is a horrible experience. If the choice is between presenting that horrendous survey and not presenting one… choose not doing it.

     Surveys have evolved magnificently over the last few years. They are unobtrusive (show up on the bottom right, in a small window), they show up when people leave the site, they show up with only two or three, and I can't stress this enough, relevant non-pimpy questions. All these things ensure that the customer has a good experience, and they are more likely to complete the survey and give us valuable feedback.

     I have outlined a set of survey tools that are free in the VOC part of this post: http://zqi.me/watool , I encourage you to review the, experiment with real world implementation and let that influence your decision (rather than an opinion that you, or I, might have).

     Avinash.

Join the conversation. . . .