Last month, Mike explored the different approaches to segmentation and how these could be successfully utilised within targeted communication. Once key target segments have been successfully identified the question remains how these segments should be validated in order to determine where targeting activity should be focussed. In the second part of the article, Mike identifies key markers that can be used within this final process of segmentation.
Mike Owen 
21 Jul, 2009
Effective Segmentation for Successful Communication – Part 2
Read Effective Segmentation for Successful Communication (Part 1)
Validating the segmentation
Having defined our segments in qualitative terms we are then faced with two further issues. Firstly we need to be sure that our chosen segment or segments are sufficiently large to warrant the effort that will be focused on them. The size of a particular segment may determine whether or not they are even included in our target. In some cases we may also want to target segments that have different needs where we feel our brand can deliver in a number of different ways. Secondly we need to find some means of identifying individuals who fall into our segments or to decide that that this is not appropriate or necessary.
To deal with the latter point first. Where our segment forms a very large proportion of a clearly identifiable audience, for example if our key segment was ‘most psychiatrists’, or where it would be impractical to try and segment a target audience say through lack of resources, we might actually skip this stage and decide to orient our communication to the needs of our key segment but to communicate with the broader audience and accept that for a proportion of that audience our message would be wasted.
However, assuming that we do want to validate our segmentation and create
the means to identify individuals who fall into our segment(s) the approach
is typically quite straightforward. Here the decision to use an internet based
approach, telephone or a face to face approach will depend on logistics and
practical considerations as all three have the potential depending on exact
circumstances. Personally I like the flexibility and interactive potential
of internet.
The most important element is to identify clearly what has to be measured. Given that respondents often cannot articulate their latent needs and find it difficult to recognise that their needs may change in a future world we have to use ‘markers’ that are indicative of their future response to brands that address particular needs sets. The nature of these ‘markers’ will vary by therapy area and according to brand but are likely to be:
- Attitudinal: the presence of certain beliefs may suggest a predisposition to engage with particular needs sets when shown appropriate evidence or argument.
- Behavioural: certain behaviours may indicate that individuals will respond in particular ways to future needs sets even if they do not recognise this themselves
- Contextual: response to particular ‘offers’ may indicate how individuals will respond to future situations when presented with particular brand propositions
To return to our earlier bottled lager example, those respondents most likely
to adopt bottled beers would have been innovative and articulate, users of
‘youth’ brands in other markets, users of more upmarket drinking establishments
and responsive to the concept of ‘badges’ for lager brands. But if asked directly
do you want a bottled lager at twice the price of your draft brand they would
have largely rejected the proposition.
The other important part of this stage is to create some sort of ‘typing tool’ that allows the client and their representatives to put individuals into a particular segment. The way this is done will depend very much on individual client needs and resources and can vary from a simple few questions a representative needs to answer about a particular doctor to something of much greater complexity.
Some suppliers have offered clients targeting approaches by which a small number of Doctors whose segment is known are then used to extrapolate a target list based on common characteristics between these Doctors and the broader sample. Whilst not entirely without merit this approach assumes that attitudes and behaviour relating to one aspect of medicine will correlate in another aspect. For example that a ‘forward thinker’ in say asthma will also be a ‘forward thinker’ in say Women’s health; in our view this is a dangerous and not yet validated assumption and such approaches should be treated with caution.
Identifying and targeting
Once the means to identify individuals has been created the job of the researcher is largely done. It is then up to the client to put theory into practice. However, a key decision still needs to be made if more than one segment is targeted and that is whether or not each segment should have the same or a different communications strategy and materials. This depends on both practical considerations, especially is it cost effective to produce two or more different sets of materials, and also on the nature of the difference between the segments. Crucially will they respond to similar core messages or not?
In our experience it is more often the case that one approach will work for
several segments even though each segment may focus on the communication in
a slightly different way. In this way a unified approach may be viewed by different
people in different ways, but the ultimate impact on their behaviour is the
same.
Final thoughts
A lot of segmentation work contributes relatively little to the final sales and marketing strategy because brand teams do not have the resources to identify which individuals fall into which segment or to create different stories and materials for each segment. In this scenario they would be better off using their segmentation research to identify a common approach that will have broad appeal because it addresses fundamental needs rather than concerning themselves about differences they cannot realistically address.
Your response:
Login to respond to this article: