Your data is lying to you.

Your data is lying to you, and it’s costing you your brand. High fidelity predictions seem to be where everyone is headed as AI rears its head and machine learning presents its opportunities. It’s where everyone is focused, and what everyone wants. It’s a ridiculously complex quest. A bitter pill many of us have had to swallow as we attempt to eke out a better ROAS, return on investment, ecom performance, profitability, etc. At the heart of it is getting to a meaningful view of customer behavior, which requires mining myriad points of data and then piecing them back together again to understand how the product lines we’ve sold in the past relate to what we sell today and will be selling in the future, and how all of that relates to customer acquisition and retention, and ultimately to the success or failure of our businesses. 

Ideally, it’s a view of the physical and digital enterprise, experience by experience, getting to know the customer beyond just a single digital browse or in-store transaction. It requires knowing who the customer is, how they chose A over B or vice versa. Chasing down all of the little moments leading to milestone transactions and the myriad moments that lead to the next transaction and the next. It all has to be accounted for before any algorithm attempts to squeeze out its very first prediction.

This is where the industry is headed, supercharging the Customer Data Platform in such a way that it’s not left to an analyst to build query after query in Power BI, Looker, Peel, or whatever user-defined dashboard is employed in an attempt to satisfy executives as they assess how to allocate resources to scale and net a profit. 

The mission is straightforward, empower the entire organization with massive scale learning, supporting all of the common business goals and outcomes with predictive future insights, while moving ever closer and closer to the customer. Then, connect with them as individuals, harmonizing the customer experience to the value they require from your brand, increasing the quality of the customer’s experience throughout their journey with you from their first purchase and beyond. The looming question is how? How do you move beyond disparate data inputs and perspectives to a holistic understanding of how well your brand is imparting value across this journey? Recency, frequency, monetary metrics? Nope. Cohort analysis? Nope. Persona segmentation? Nope.

The most important part of the entire exercise is understanding relationships; i.e. understanding how relationships form between a brand and its constituency, its customers. This requires moving beyond singular data events, applying knowledge about customers, products, and experiences in order to arrive at business and customer outcomes that portend a healthy relationship; i.e. a regular pattern of repeat purchases sustained over time. This, of course, is where both profit and scale reside, and hence the goal of the holistic data endeavor. 

For example, programs are being run for churn, shopping cart abandonment, browsing triggers, look-alike seed audiences, and 30, 60, 90 day post purchase email offers, among myriad others. The industry zeitgeist (brands and big data providers) are trying to identify shopper health, recover abandoned shoppers, and in parallel they’re also trying to find their next generation of customers. The connective tissue sitting between the intention of the marketer and the outcome of the shopper is the brand’s CRM Platform, or the more comprehensive Customer Data Platform (CDP). The insights engine is hyper sensitive to myriad, yes, myriad seeming moments of decisions the data analytics world identifies as triggers. These are not emotional triggers, these are statistical likelihoods that set off alarms for the marketer to intervene in the moment with a message or promotion. The idea that has taken hold is that deploying (pardon the misnomer here) ‘insights-driven’ marketing programs, will make more money. The holy grail currently sought is to scale efficiencies associated with these programs under the guise of providing an “Enterprise Shopper View” that ultimately generates increasing revenue returns.

The “opportunity” falls into the category of what the author of Thinking Fast and Slow, Daniel Kahneman, describes as a “Linda Problem.” Meaning, a misunderstanding of events based on a subjective emotional interpretation. Not rational thinking. As example, let’s focus on churn risk, defined as a percentile likelihood of disengagement from a purchase habit. The Enterprise Shopper View defines the “where” of the opportunity. In the example of churn, marketer A wants to know if the churn event is in-store, on the website, or at the brand level. They may also want to know if that churn event is specific to a product, a seasonal launch, a service experience, or possibly a browsing experience. This is where scale comes into play. Let’s suppose you have 25 stores (digital/physical), selling 150 products, recognizing 3 discrete seasons a year across 10 actionable customer ‘types or cohorts,’ aka personas. In this example there are potentially millions of permutations, or opportunities, to create unique, high-value marketing programs, which is way more than anyone has the time, dollars, or staff to assess, let alone act on.

The drive in the CRM and CDP data world, digital marketing industry, and now brand management culture is to pinpoint opportunities at scale and then automate specific marketing programs to affect overall “customer performance,” which can also be described as trying to cajole the customer into a purchase.

Returning to the Churn example. Most CRM and CDP platforms, and specifically machine learning platforms, can provide a churn risk estimation, and they can pinpoint location and circumstance, but still missing is a reliable, well-informed understanding of predicted customer value, which quantifies the risk. Also, and more dramatically missing, is clarity across these estimated millions of opportunities is the true cost of the demise of the brand/customer relationship and the path to optimizing that relationship. In the architecture of CRM and now machine learning are the human emotional motivators that are the true drivers of shopping behavior, moreover the true drivers of why people love or hate a brand.   

The hard truth is that as OKR’s and KPI’s continue to drive decision making around event ROI, replicability, and accountability, the value of a single customer’s commitment to a regular buying pattern, the value of that commitment, and the emotional motivators surrounding that commitment are entirely lost. This is where the Brand Equity Index™ is uniquely focused, surfacing predictions and diagnostics, as an algorithm that defines how well a Brand engages its constituency. Understanding what compels the customer to commit to and exhibit a regular buying pattern. It doesn’t take place in a single instance as the above machine learning technology hopes. It doesn’t even take place in a literal fashion at all. Until now, brand managers have been the connective tissue between data and outcome; i.e. they have been the observers providing an empirical guess of how efficient a brand introduces itself to a consumer, engages that consumer, and nurtures that consumer to the extent that same consumer initiates a relationship with said brand, and consummates that relationship with a repeating purchase habit. Until now, all of that has been an estimation by a human. No longer.

“Extraordinary claims require extraordinary evidence.”

This phrase was made popular by Carl Sagan who reworded Laplace's principle, which says that “the weight of evidence for an extraordinary claim must be proportioned to its strangeness (Gillispie et al., 1999).” This idea is at the heart of the scientific method, and a model for critical thinking, rational thought and skepticism everywhere.

Here then is the evidence. Even with all of the critical thinking happening in the customer management realm and the onset of machine learning and AI, the typical “engagement” or “relationship” development that we see in analyzing brands across categories, product types, and the spectrum of success, scale, and profitability, looks like what is depicted here.

If we follow a set of Prospects from their first experience with a brand along their journey to what is nirvana in terms of loyal customer behavior, aka Cheerleaders (multiple buyers in a repeating purchase habit), what we witness, typically, is dramatic failure, and, let’s emphasize ‘typically.’ Some brands do perform better. But those are the exceptions to the rule, not the rule. The rule tells us that of 1M Prospects, only 0.01% ever go on to form meaningful, long-term, buying relationships with the brands that service them. And (not depicted here, above) only 0.006% become Brand Advocates (who we affectionately call Cheerleaders). Brand Advocates are the most valuable customers. They sustain their purchase habit over time. They recruit from their circle of influence on behalf of the brand. And, they pay a premium to stay in relationship with the brand they love. 

If these results weren’t sobering enough, in terms of profitability—we will lean on the data that we have witnessed in the aggregate over the past 20 years—Casuals are by far the largest segment of buyers, but the least profitable. By definition they are first time customers. Only a second purchase ‘Migrates’ them to the status of “Loyalist.” When one subtracts fully allocated marketing expenses and COGS, most often Casuals are barely above breakeven. On the other hand, Cheerleaders are extremely profitable. They make many purchases on a frequent basis and maintain that habit over years and years. This is where profit resides. 

And now for the extraordinary. In terms of scaling businesses, there is one driver that is affordable, efficient, repeatable, and does not increase in cost over time. It’s the rate of recommendation from Cheerleaders. Looking at our data of over 100+ brands across categories, on average, Cheerleaders recruit 37x (37 times) more new customers in brands with strong Cheerleader metrics versus brands with weak Cheerleader metrics. If we then think of the holistic point of view, rather than the machine learning driven “Enterprise Shopper View,” the measurement of the efficiency of creating a robust Cheerleader segment is where health and wellness (i.e. scale and profit) reside. Measuring that efficiency is possible. We call this efficiency “Brand Equity,” and the measurement is an index, The Brand Equity Index™.  

With an extraordinary claim comes extraordinary evidence. We can prove the above phenomenon with your own data. Give us your ecom platform’s historical transactional data and we will show you how this perspective changes everything. It’s the construct. It’s the data. It’s the context. It’s the analysis. It’s the predictions. It’s the recommendations—all presented back to you in a one hour session. Start here.