you just Google it? Perhaps throw together a quick Survey Monkey? Buy a survey research report on that topic? Use an automotive market research team?
It really depends on three major factors - your question, the amount of time
you have to spend, and whether you need to be able to build actionable strategies
from that data.
It’s never been easier to run surveys among target consumer
groups. With the proliferation of survey tools, and the rise and rise of Survey Monkey, how do you make sure you are getting the right
information to help your business move forward?
A single factor has remained
constant over time, regardless of new techniques - the demand for quality, robust insights. The consumer survey is much
maligned, but remains a vital tool in the generation of validated, robust
insights that can drive business decision making.
The market research industry is very good at overwhelming
its clients in data and analysis, because that’s the space they live in. It’s easy to assume that everyone loves detailed analysis when, in fact, everyone actually loves clear answers. While the two are similar and linked,
they are definitely not the same.
Segmentations, or cluster analyses are good examples.
There are many different tools for segmenting the market, choosing the right one for the right job is vital, and having this knowledge upfront is key to a successful outcome.
Two common methods of clustering (or segmenting) data are hierarchical
cluster analysis, and k-means cluster analysis. Hierarchical cluster analysis
is very good at grouping what feature sets are the most useful or appealing to
a customer group. This is because it works on the biggest/best/most appealing
feature and works down from there (hence “hierarchical”). This can be less
useful for grouping your customer base, unless you like grouping on the most
obvious features, like 'male or female'….
k-Means cluster analysis is very good at trying to identify
the best groups of people from a set of questions, for example, a set of
behavioural, or attitudinal questions. This is because it creates a range of
different clusters and groups people to the nearest mean-score, hence k-“means” (don’t ask about the k, that’s just a stats thing).
"Depending on the outcome you’re looking for, it’s important
to know not just how you’re going to use the information, but the specific way you
are going to extract the insights from the data" says Stephen Scales, Head of Consumer Research at SBD Automotive. "Knowing this from the outset
will determine the way you go about designing the entire survey, and whether
you should use something like Survey Monkey yourself or get an Automotive Consumer
Research team (like the one at SBD) to develop a bespoke survey from the ground