Customers are or should be, at the heart of every organization. We segment them to personalize our offering and approach them differently to satisfy their needs most efficiently and effectively. However, in this day and age, are we really doing the best we can? Are traditional segmentation techniques still the way to go, or are we missing something?
We all want to get personal and communicate the right message at the right time at the right place to the right person. This is where traditional segmentation techniques might not be effective enough to enable organizations to offer this kind of personalized experience. We all understand the value of segmentation as a means to better target our effort, but should we continue using segmentation when technology and big data has paved the way for hyper-personalization and one-on-one offerings?
When customers can no longer be pigeonholed
With limited or no computing power, traditional segmentation is often based on four characteristics: geographic, demographic, psychographic and behavioral. The days of making simplistic demographic segments such as "housewife under fifty" or "dynamic young professional" are long gone, although Don Draper from Mad Men makes us all nostalgic for the days where we just needed a good slogan and big billboard to sell a product to the housewife under fifty.
Along came the needs-based segmentation. As marketing moved from product-centric to being much more customer-centric, we started clustering customers based on their needs. Using the traditional matrix, based on two or more dimensions, we like to place every customer in the right box.
So why should we rethink the way we segment today? Static segmentation models don't capture customer’s and prospect’s fast-changing behaviors. Customers have become more demanding, dynamic and eclectic and therefore no longer fit into our rigid pre-defined boxes. They wish to be individually recognized and want the spotlight to be on them. They are characterized by what they do, not just who they are. Hence, it has become more complex to segment them.
Segmentation 2.0: getting small with big data
Today, more data crosses the internet every second than the amount stored over the entire internet just twenty years ago. As marketers, we have the luxury of having an abundance of customer-related data available to play with. Traditional data such as demographic and psychographics are being extended with transactional, interactional, activity-based data and much more. We have access to social profiles of customers, website behavior, purchase histories, mobile data, etc., which enables us to understand a customer on an individual level and draw actionable insights just for him or her.
The rise of big data and the growing number of customer segments has led to the paradox of segmentation. After reaching the highest level of precision, does it still make sense to use this kind of micro-segmentation if we can target individuals directly now that technology has enabled us to do so? Maybe not.
A more advanced option could be hyper-personalization where customer segments are reduced to a single individual. These ‘segments of one’ are targeted with tailor-made solutions in real time, fueled by big data since the volume, variety and velocity of data has to be significant. However, hyper-personalization is not just adapting segmentation to an individual level; it is also about hyper-contextualization. In fact, customers are constantly changing.
Adapting your offerings by analyzing their usage and habits in a specific context in real time is becoming vital. Once you have identified your customer, the real value lies in the activities that leverage the solution, creating greater go-to-market efficiencies and effectiveness. This all sounds very exciting and promising, but are we being realistic?
As we become even more connected, the Internet of Things will further speed up the data collection enabling companies to get to know their customer even better. However, let’s not get ourselves carried away too soon; we still have some challenges to tackle today. To manage and integrate the enormous amount of data coming from internal and external sources, let’s start by having an efficient data management platform. These DMPs allow real time data management and processing, and are becoming more common. In many companies, however, data is still managed in silos. It could be that there are regulatory reasons for this, such as financial institutions that have to isolate their insurance database from their banking database but it is often due to structural issues.
Next to that, there is also an economic consideration: the risk of diluting marketing efforts. While segmentation is intended to facilitate the targeting of priority customer segments and to ensure the most efficient marketing spending, hyper-personalization is meant to enable brands to fulfill the specific needs of every individual within an entire market. Also, from a branding point of view, hyper-personalization makes it difficult for a brand to stay consistent as its value propositions are often different.
On a final note, we can state that hyper-personalization leverages big data to the fullest and is definitely on its way. It has opened the gate to a new era of marketing where customers are taking center stage. However, hyper-personalization is not a complete substitution for traditional segmentation. Based on the challenges we are still facing, it seems more likely that hyper-personalization and segmentation are complementary rather than conflicting alternatives.
How are you leveraging your data today? Complete our quick scan to find out your current level of data maturity.