ADOTAS – At the eMetrics Summit in Washington, DC, today, iPerceptions is unveiling the upgraded freemium versions of its 4Q exit survey software-as-a-service, which was initially developed with analytics blogger and author Avinash Kaushik. Though the basic version remains free, there are now three subscription models that feature enhancements such as integration with Google Analytics, individual behavioral analysis, industry benchmarks and social media integration.
iPerceptions President and CEO Claude Guay took a minute to talk to us about exit surveys versus quantitative analytics as well as describe the new 4Q Basic, Plus and Premium models.
ADOTAS: How did iPerceptions initially come to work with Avinash Kaushik?
Guay: After a conference where Avinash proclaimed that the three most important survey questions were “Purpose of Visit,” “Task Completion” and “Why or Why Not,” our VP of Marketing sought him out to create a product. They added “Overall Experience” as the fourth question.
Can you explain the difference between “Reporting Squirrels” and “Analysis Ninjas”?
Reporting Squirrels: People who collect data for the sake of collecting data and reporting without practical analysis or insight. Analysis Ninja: People who find practical insight and execute changes based on the data to better the experience and/or the results.
Have you ever considered “Be a Ninja, Not a Squirrel!” as a marketing slogan?
Yes, but there is only one Avinash and we are not going to steal his brand. ;-)
Do you agree with Kaushik that exit surveys are more valuable than quantitative analytics?
Yes, if you could only get one or two metrics and you have no time, experience or resources, you should focus on getting the Purpose and Task Completion levels, it is the basis of what you visitors are trying to accomplish and the result of the effort on your website.
Or should exit surveys and quantitative analytics never be parted?
Behavioral is really valuable because it tell you what they did in addition to why they did it. Both sets of data give context to each other and are both a good place to start analysis but for different cases. We are absolutely convinced that putting the two together is like 1 + 1 = 3.
In general, what percentage of departing visitors complete permission-based on-exit surveys?
Two to six percent of people accept to answer a survey when politely asked on arrival. More than 90% of those who said they would actually do it on their exit.
Does that vary greatly from industry to industry?
A little bit by industry but more important is the emotional attachment to the brand. If the link is strong with a brand — i.e., Mercedes or BMW — they will tend to have a higher acceptance from people willing to take the survey.
You’ve just begun offering three new turbo-charged pay models of the 4Q SaaS — how can publishers determine which is the best fit?
Number of visitors and collection rates are important factors. The more data (respondents) you have the more you will want iPerceptions to process the data into insight for you because it becomes time consuming. If your number of respondents is low due to lower traffic, you can actually process the data and see variation just but looking at it.
Could you explain how the social network integration works?
On the “Thank You” page of the survey, website owners will be able to offer visitors to link to a Facebook account/page or Twitter, etc. More interesting, you will be able to filter data by referral (Facebook, Twitter, etc.) and see why they came — that is, if they self-declared they accomplish their task and verify with the actual conversion. For example, whether the visitors coming from the social marketing campaign or the ad campaign differ in Purpose of Visit and Task Completion.
Of the many new features offered in the new 4Q, which one was most clamored for by users?
The Google Analytics Integration, which makes it really easy to combine both sets of data.
Which feature do you think will provide the greatest long-term benefit?
Detecting the most significant trends/changes in attitude and behavior with a high level of confidence. We pre-process the data to tell the user all the interesting things they should be looking at.