The Rear-View Mirror Problem

Oomiji’s newsletter can also be found on Linkedin and you can join the 1,300 people who have already subscribed  here. We’ll also continue to post articles on the Insights section of our website. As always, we’d love to hear any comments or suggestions that you may have.

Imagine walking into a retail store. A sales clerk approaches and asks you the following questions:

• Did you click on any of our ads? How many times?
• Have you bought from us before? 
• What did you buy?
• How many of our emails have you opened? What were the subjects?
• Did you ever start a purchase and then abandoned the cart? 
• Have you visited our website. How many times? Which sections?
• Do you have loyalty points? Have you used them?

Of course, the scenario is absurd. No salesperson would conduct a conversation with those questions because they don’t provide any understanding of the customer’s interests, motivations or perceptions. In fact, they are nothing but a rear-view mirror into what the customer may have done in the past but tell nothing about the present or future. Yet, this is the kind of information that CRMs and other marketing platforms gather today.

The actual retail scenario is that the customer and sales person have a conversation about the customer’s needs and experience. The customer may discuss their lifestyle, product or service preferences and why they prefer those things. In the end, when the discussion is at its best, they’ve formed a relationship and whether the customer bought or not, they each walk away with an understanding of the other.

In the first example above, there is no engagement or understanding, even though we may think the metrics provide one. In the second scenario, there is the beginning of engagement and a relationship. 

Why then, as soon as we go online and enter the world of digital marketing do we forget everything we’ve learned about how customer relationships are formed? The tools are there to simulate the customer’s store experience but we seem to believe that hard data is the only way to do it. This seems to be a key reason why so many companies have trouble building lasting customer relationships and creating brand loyalty.

We fool ourselves into thinking that past metrics are an indication of future behavior. Every customer leaves behind thousands of digital footprints:

• CRM systems record purchases.
• Websites record clicks.
• Advertising platforms record impressions.
• Social platforms record engagement.
• Retailers record transactions.
• Email platforms record opens.
• Support systems record complaints.
• Loyalty systems record points.
• Surveys collect opinions.
• AI summarizes everything.

Yet executives increasingly admit they don’t understand what customers actually want. That’s not because they lack information. It’s because they have too much of the wrong information. The problem is no longer data collection. It is interpretation.

Why has this never been solved?

For thirty years software companies competed to capture more data. Every new platform promised another dashboard, visualization, score, prediction, automation and integration. The assumption was simple: If we collect enough information, understanding will emerge automatically. But it never did because understanding isn’t additive. It’s interpretive. Humans don’t make decisions like databases. They make them through emotion, identity, fear, regret, hope, memory, and belonging. None of those fit neatly into a CRM field. So technology kept measuring behavior while completely missing motivation.

The analytics are not wrong. They’re simply facing the wrong direction. Just like a rear view mirror, they explain what happened yesterday. Companies spend billions perfecting their rearview mirror but it doesn’t help them see through the windshield. Customers don’t live in historical data. Their decisions are made on what they see in the future.

We’ve thought about all this in creating a system that could do better at understanding tomorrow’s decisions and our solution has been to integrate open-ended questions that gather customer language with email marketing and segmentation down to the individual. Most solutions use AI to try to interpret past data into future projections. But the most productive use of AI is to use it to analyze thousands of customer comments to both provide an aggregate profile but more importantly, to define the parameters by which each person fits into an emotional or perception segment. That way companies can see all of the different “whys” behind what customers feel and want and address them as if they are speaking to a single customer after a similar conversation as the retail store walk-in example described earlier. 

What companies have to decide is which is better: thousands of hard data points that look back or millions of words about the present and future. That’s not a difficult question to answer and it’s time to rethink the way we use data to build relationships.