In the never-ending struggle to gain a competitive advantage and achieve market dominance, the go-to tool for almost every digital marketer is market segmentation. Unfortunately, apart from a few exceptions, segmentation has failed to produce the big uplift in sales and revenue it promises. This is because, even in the age of digital, marketers are using conventional segmentation when today’s prospects and customers expect brands to speak to them literally one on one.
What’s the Issue?
The issue with conventional segmentation is more strategic than tactical. This is because it does not provide us with the information we need to build concrete strategies to improve returns or even tell us what products and services need to be developed for each customer to improve the outcomes that matter (sales, retention, profit)
To understand this better, let’s dive deeper to see where conventional segmentation is starts to fail. The prevailing logic is that companies need to build products that meet a need. Prospects who have this need become customers and continue to be customers. Conventional segmentation was developed around this logic because it targeted a customer need and then met that need with an appropriate product or service.
This means that conventional segmentation is seen as a one-off activity because it’s something you do at the beginning of building a business, product or service. It is rarely revisited thereafter. Also, segmentation traditionally starts with the creative or intuitive naming of segments. These names can feel very descriptive and comprehensive but often the segments don’t act as expected in the real world. This is because the segments themselves tend to be generic and there is no validated way to link them to an individual customer.
Despite its unpredictable record, marketers remain loyal to conventional segmentation. They keep rolling out segmentation plan after segmentation plan hoping the next one works better than the last. But in the digital age, where customer attitudes and change at the speed of an individual’s internet connection, conventional segmentation is just not appropriate. For segmentation to succeed, it has to be about individuals not large homogenic groups. It has to be about needs not products. It has to be about tactics not strategy.
Here’s a famous quote from IBM CEO Ginni Rometty that validates this claim:
“Data will spell the death of customer segmentation and force the marketer to understand each customer as an individual within eighteen months, or risk being left in the dust”
The conventional way of looking at segmentation is inadequate and therefor ineffective. Prospects today are demanding that brands speak to them ‘one on one’. The marketing triple play of segmentation, targeting and positioning (STP) needs to be reimagined and augmented with greater and better data so businesses can find unique, profitable segments.
The good news is that today, literally every customer touch point is captured with vast amounts of data are now stored in data swamps and lakes. Fortunately, retrieving it and making informed decision has become much easier thanks to API’s.
Segmentation in the Digital Age
The biggest advantage of the digital age is the opportunity to go beyond census, demographics and socio-economic classification data sets (SEC). Businesses today need to treat segmentation like a living document – one that never finishes since customer behaviors are constantly changing.
Here’s how segmentation today should look. On top of the standard segmentation, variables such as search intent, webographic data and behavioral data from your analytics should be included. Some of the webographic factors are your site visitor’s device type, operating system, browser version, connection speed. Behavioral data factors could include time on site, number of visits, traffic source, user affinities, gender and many more.
These multiple variables create millions of permutations with high intent. It’s humanly impossible to compute or make sense of all of them. This is where artificial intelligence (AI) and machine learning (ML) come into play. The objective is to identify profitable audience segments. Clustering the data using ML allows you to break a dataset into groups based on observations and then them further into homogenous groups for targeting.
With all this information residing on databases in the cloud or on the ground, the data points can be linked back towards unique segments resulting in information that is highly tactical and actionable. ML algorithms can further identify clusters based on KPI’s such as cost per acquisition (CPA), return on ad spend (RoAS) and lifetime value (LTV). This gives businesses the opportunity to identify smaller groups with higher intent and a better product fit. Additionally, it enables marketers to establish a personal connection with prospects and customers to develop focused content and creative assets to target these groups to for even better results.
Performing this kind of segmentation is becoming increasingly easier to do as most of the core data points needed for it reside within online ad platforms (Facebook, AdWords, Bing, LinkedIn etc.), your web analytics and your own databases. AI segmentation and targeting applications like Alavi.ai (www.alavi.ai) can then quickly analyze this data to identify audiences with the best potential to maximize returns. They can make segmenting work for you.