Market segmentation is a powerful tool for identifying, acquiring and retaining customers in the most efficient and effective manner possible. However, too many segmentation implementations don't work well despite large investments in time and money.
A successful segmentation effort needs to have what those in the entertainment business call "legs", that is do well not just at the opening weekend, but in subsequent weeks and release windows, Pay TV, DVD, Cable etc. Likewise, for a segmentation project to be successful, it needs to survive and thrive in operational "windows", being implemented in product development, marketing campaigns and sales efforts.
How do you ensure your segmentation thrives in these operational environments and does not simply gather dust on your desk?
A lot can be done in the planning and design stages to ensure success. Here is a partial list of some of the important considerations.
1. Define the operational environments first. Who will use the resultant segment map? How will they use it? What will they need to successfully implement it? A recent example is a client whose sales process and CRM tools were not aligned to accommodate the segmentation work. The segmentation therefore never got used for one its originally intended purposes-helping close more sales.
2. To make the segmentation as useful as possible, let the data drive the segmentation map. Tell your research supplier that you are interested in conducting a post-hoc (versus a-priori) segmentation. The basis for segmentation will come from the data itself by identifying naturally forming clusters through multivariate analysis. A classic example is Crest toothpaste. Going into the segmentation effort, the benefit sought of "no cavity checkups" was not known to Procter and Gamble and the cluster emerged from an analysis of the data. The result is a highly successful premium brand build around the benefit of no cavity checkups.
3. In order to conduct a successful post-hoc segmentation, enough questions need to be asked to enable natural clusters to emerge. To this end, we highly recommend a qualitative in-depth exploratory phase upfront to help define the questions to be asked in the segmentation survey. This phase will need to be very open ended and exploratory. Resist the temptation to be too structured in this phase. Make sure you allow yourself to wander far afield in this phase.
4. Ensure you focus on characteristics beyond segment size. The largest customer segments may not be the most interesting. In the toothpaste example above, the no cavities checkup benefit segment was not the largest but offered the best potential for both category creation and premium product placement.
5. Last but not least, keep it simple. Segment maps that are complex and long typing tools may be impressive and sophisticated looking at first but loose luster and "legs" if they are too complicated to implement.
Al Nazarelli is President & CEO of Silicon Valley Research Group