Skip to content

How to Solve 5 Common Consumer Analysis Mistakes

2010 May 5

Conversation is in vogue this year, which begs the question:  Who are you talking to?

Analyzing consumers is one of the most important things marketers do and therefore it’s important to get it right.  Unfortunately, there are many pitfalls and it’s easy to get lost.

Here’s a quick guide to common mistakes and how to avoid them.

False Assumptions

Consumers are much more complicated than we give them credit for.  A doting mother of three might also love freestyle snowboarding and Metallica.  Everybody has a variety of interests and takes on various roles that often don’t fit into the neat categories we marketers like to use.

One common assumption is that rich people buy expensive things and people of more moderate means buy cheap things. In fact, we all sacrifice in some categories in order to splurge in others  (like a receptionist who scrimps and saves to buy Prada shoes or a millionaire who likes to go bargain hunting).

This is a well documented phenomenon.  For those who are interested  in exploring it further, I recommend the book Trading Up.

Most of the time, people’s purchase decisions aren’t driven so much by demographics, but by psychographics.  How people respond to questions like “I like to take risks” or “I prefer to spend time with my family” can tell you a lot more  than knowing their age and sex.

However, as the Ad Contrarian points out, in using psychographics we often fall into the trap of amateur psychology.,  So it’s important to ensure that psychographic targets emerge from clear segmentation derived from research, rather than from office navel gazing.

Failure to Segment

Often, a market can look like one big consumer group, but it’s actually a collection of very different groups of people.  If you only look at the aggregated group, you will often be analyzing a non-existent consumer.

One good example of this was when my team was analyzing the Ukrainian internet market a few years ago.  The typical user was considered to be male and 27 years old.  However, we found 4 segments: 2 male and 2 female.  The younger segments averaged around  20 and the older ones closer to 40.  Once we realized that, consumption patterns started to make sense.

A classic case of uncovering a segment is Sony PlayStation, which directly targeted adults who were purchasing the console for themselves, not for their children.  It seems quaint now, but in the mid-90’s it was a radical idea.  It also resulted in one of the most award winning ad campaigns ever.

We should always be on the lookout for new segments.  When your data isn’t making sense, it should be the first thing you check.

Sample Size

Sample size is probably the most overblown issue in research, mostly because it is the one source of error which cannot be the research agency’s fault.  It’s easy for them to claim that if you were willing to put up the money for a larger sample, there wouldn’t be any problems.

Another common mistake is that it doesn’t really matter how big the sample is for the entire survey or even for the group you’re looking at.  The only thing that is important is the size of the sample for the question being asked (i.e. a target group with a sample size of 100 will fall to around 50 when you segment by sex).

Sample size issues can seem a bit confusing, but there is a really simple solution. Just divide one by the square root of the sample size (1/√sample size) and that will get you the sample error with about 95% confidence.  So if you have a sample size of 100, your total sample error will be 10%, or +/- 5%.

This is simple way but fairly accurate way to always stay on top of sample size.

Looking for Something That Isn’t Really There

Analyzing consumers entails going through lots of data and using some math.  At times, it can become complicated.  The results, however, shouldn’t be.  If they are not crystal clear, something is wrong.  Most probably you’ve been looking for something that isn’t really there.

This is a common pitfall that we’ve all fallen into.  A notion just seems to make so much sense that you feel that there has to be data to back it up.  If you look long enough, you will always be able to find something that seemingly backs up your case, but you’ll still be wrong.

You should be able to arrive at the same conclusions using a variety of methods.  To use an overly simple example, if a consumer group is young, you would expect a high proportion of students.  Just like in your high school math class, you need to be able to check your work.

While being unable to find the answers you’re looking for can be frustrating, it is also often a very good thing.  It alerts you to the possibility that the world isn’t quite how you expected it to be, but different somehow.  Uncovering why is how true insights are born and the reason we do marketing analysis in the first place.

Over the years, I’ve developed a very simple way to avoid this problem.  Simply lean back from your computer screen.  If you can’t find what you’re looking for from that vantage point, it probably isn’t there.

Have a Point

Having a point not only makes your analysis more interesting for those who you present it to, it also makes it much more likely that you will achieve some actionable insights that will result in concrete actions.

Years ago I was working with a marketer who proudly announced to me that he had just prepared 160 slides on internet users.  He had broken them down in every conceivable way and produced a large and utterly useless document.  It was an incredibly foolish waste of time and effort.  He had no point.

The best way to avoid this is to split your work into two stages.  An investigative, data driven stage where you delve into the research and an analytical stage in which you build a clear story that lead to specif marketing activity.

If your data doesn’t tell a story, then you haven’t found anything of value.

- Greg

4 Responses leave one →
  1. Carole Lotito permalink
    May 5, 2010

    Your last line says it all.

    The story is the prize at the bottom of the box.

    Telling it effectively to an audience with a short attention span is the real accomplishment.

    [Reply]

    Greg Reply:

    Carole,

    Thanks. Obviously, I couldn’t agree more!

    - Greg

    [Reply]

  2. Robert Neuschul permalink
    July 7, 2011

    Counterpoint from the empiricist’s perspective: very old warning line for doctoral sceince candidates.

    “Just because one can see a pattern in data doesn’t mean that pattern has relevance significance or value.”

    :-)

    Good piece Greg (again).

    [Reply]

    Greg Reply:

    I like it!

    - Greg

    [Reply]

Leave a Reply

Note: You can use basic XHTML in your comments. Your email address will never be published.

Subscribe to this comment feed via RSS

CommentLuv badge