When Google announced it was entering the world of survey research earlier this year, my initial response was fear and trembling.  After all, Google’s self-service platform and search business has hastened the end for many an industry.  Anyone remember the last time they used the Yellow Pages as more than a doorstop?  Could market research be next?

In spite of these concerns, as a research practitioner not beholden to any one product or service, I was intrigued what the search giant would do for our business. It turns out, at least at the moment, not much!

Google Taps Double Click Publisher Network to Serve Questions
Google’s Consumer Surveys leverage Google’s content network to serve survey questions across publisher sites.  Sites in Google’s Double Click network provide content in exchange for readers answering a single question.  Readers get free content.  Publishers earn additional revenue from visitors and Google makes money from each click. Marketers are provided a quick, easy and inexpensive source of consumer insights.  It’s a win, win, win!  Right?

Well, sort of.

Google has provided a broad-based national platform for marketers to cost effectively get consumer feedback using a limited number of questions.  Because of Google’s scope, most surveys can be fielded quickly and the self-service nature of Google allows for easy user analysis.  In addition, Google offers a limited number of commonly used age and gender breakouts.

Testing Approach Against U.S. Census
To prove its methodological approach, Google commissioned a test comparing results from its consumer surveys with those conducted using two well-known online survey research providers against selected U.S. Census data.  The test demonstrated lower deviations from the actual Census data for Google than for other leading online survey providers.

In addition, Google’s initial participation rates are in the high teens, significantly higher than most other online survey firms. While online surveys do have limitations, Google has done its homework to demonstrate its Consumer Survey offering can stand its ground in the online survey space.

Limitations to Consider
Unfortunately, there are a number of significant limitations that limit the offerings value to researchers.

  • National sample only – Google does not yet offer any local geographical breakouts, nor are there opportunities to conduct research outside of the US. This limits its value to most local or regional marketers or those with significant international operations.
  • Limited to no more than two questions per respondent – if marketers have additional questions, Google offers ‘like’ types of respondents based on their patented algorithm (think black box).  The words ‘patented algorithm’ should make researchers nervous.  It’s essentially saying “trust us, we know what we’re doing but we aren’t going to tell you”.
  • Demographic data is inferred by IP Address of browser behavior.  Since Google has taken the ‘less is more’ approach to the number of questions each respondent answers, no standard demographic questions are provided on each respondent.  Again, Google is asking the marketing community to trust them that my 10-year old son isn’t on my laptop answering a question intended for me.  That may be acceptable to a digital advertiser but survey research needs a higher standard of certainty.
  • Google’s pricing model quickly increases when age and gender breakouts are used.  To Google’s credit it offers a very inexpensive solution when asking a few questions of a broad general population sample.  However, the price quickly escalates when marketers want to limit questions by age and/or gender.  This large price increase means Google Consumer Surveys quickly lose their cost advantage when a more exact sample is required.
  • Google only provides aggregated reporting.  Users do not have access to raw data.  As expected Google’s self-service analysis tool makes for easy analysis, but its limited reporting features make more detailed analysis challenging if not downright impossible.
  • There are no opportunities to ask open-ended questions.  Again, Google’s effort to keep the process simple and straight-forward is limiting when additional qualitative feedback is needed in the form of open-ended responses.

Future Opportunities for Google’s Consumer Surveys
While Google’s Consumer Survey tool does provide a low-cost alternative for national general population surveys, its value is so limited as to be primarily an alternative to already inexpensive omnibus studies.

This cannot be what Google intended!  Nor is it where I expect Consumer Surveys will ultimately land if they are to have a winning survey research product.  To truly provide a winning solution for marketers, Google must address these limitations starting with providing marketers greater flexibility in selecting sample and creating multiple question packages that make surveys beyond a few questions more affordable.

:: By Kirby Thornton, Director of Consumer Insights, Decision Sciences

One Response to Google’s Consumer Survey Offering Shows Promise, Despite Limitations

  1. I like what you’ve done here; I am a little late to the game but I am looking at the Google survey machine and figuring out a way to talk to my clients about this beast.
    I will say that you can in fact run an open end – what you can’t do is an other specify.
    There’s zero opportunity for any analytics of any sort from this tool.
    Now, I used to work for DCLK’s survey group (10 years ago). I had two theses. 1) We need to get response rates up somehow 2) We need to randomly intercept across the cookie network for things other than just ad effectiveness, which was my product. Because we were locked only into ad effectiveness, the company’s research endeavors failed.
    No one listened to me on either point. It happens when you’ve only been at a company for four months and they shut down your department.

    Anyway, would be a pleasure to chat about this if your head’s still in here. I am wondering if you’ve found any published research on the utility of Google’s product. I think their whitepaper, while nice, has a lot of holes and is not solid research work.

    John Mitchell

Leave a Reply

Your email address will not be published. Required fields are marked *

*

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>

Before you post, please prove you are sentient.

what is 8 + 6?