“Everything comes back to Big Data.”
This was a network security consultant’s conclusion. He was telling me about an international retailer’s data breach. The retailer had warnings that a breach was imminent. However, those warnings were lost among the millions of ultimately benign but undifferentiated notices of possible attack.
What was missing, my friend concluded, was some type of signal that isolated the true threat.
I was reminded of the customer intelligence equivalent: the doorstop binder, or reams of analysis, or zipped data file, all pushed blandly across the table. The work contains anemic results, often with equally weak observations, but no intelligence, no relevance, no significance, no value.
These results are always pushed across as completed work and the presenter smiles triumphantly. However, the work has barely begun because no one has sifted through the great pile of data to extract the signal of greater opportunity.
Deriving customer intelligence from data represents such tremendous potential. That potential is greater understanding of customer needs, greater insights to what drives response and interest, the potential to reduce costs through greater efficiency, and ultimately there is potential to speed and increase revenue.
There is nothing but potential.
And with data sourced from input, from behavior, from things, that potential only grows.
What is also growing is the gap between all that potential and what companies actually realize.
The impact of neglecting to create intelligence is borne out by a distressing statistic: prospects are 75-90 percent through the decision cycle before contacting sales.
This means we have abdicated. Prospects go along, moving from awareness through consideration, tour the company’s website, comparing products against competitors, and only just prior to reaching a decision does the customer make the company aware of the opportunity. This can’t be emphasized enough: The customer makes all the effort, and the company processes the transaction.
This statistic was published in 2013. It is an indictment of what currently passes as customer-centric marketing and sales practice. It underscores the fact we have lost control—of prospects, information, and return. This is a burning platform. And in all the time since, we go on with business as usual.
Let’s look at some evidence of how we fail to realize the potential from data and in turn, fail to derive customer intelligence.
Tyranny of Choice
So many companies now are aggregators: Amazon, Overstock, eBay, etc. They offer a full complement of commodity products. Products available on these sites and others like them can be purchased anywhere. With nothing unique in the product mix, these companies can compete on price and perhaps delivery.
Because their product is undifferentiated, aggregators employ cross-sell and up-sell tactics that essentially present a tyranny of choice.
On one of the websites listed above, I entered very specific search criteria: “Ralph Lauren sheets queen 600 thread count”. The search returned one item that matched, and then presented multiple alternatives—different sizes of sheets. Not helpful.
When I clicked on the one result that matched my search, I was presented with still more alternatives—different sizes, colors, fabrics, thread count. And if those weren’t enough, I was offered alternate websites that might have the product.
Companies are kitchen-sinking it to maximize a transaction.
However, when consumers are offered more than three options, they will sort results based on price. And the race to the bottom is on.
The larger point here is the company had direct input regarding my needs: the brand, size, and quality of sheets. Rather than make the decision (and purchase) easy, they complicated it.
Rather than apply the intelligence derived from the clear data I submitted, they ignored it. At best, they drove a decision based on one attribute (price), or worse, they drove the customer out.
This is a variation of the practice cited above: companies fail to apply the data they have available regarding customers’ interests and needs.
As an example, I received an email from eBay informing me that they had found a cheaper option to a product I had saved in my watch list. I am certain this tactic was intended to send me back to eBay’s website to assess and ultimately purchase the cheaper option.
There was just one problem: I had already purchased the product. On eBay.
In this case, the company took one isolated data point: I was watching a product. They employed a tactic—incenting me with price—attempting to drive a sale. They failed to include other data points (my purchase history, for example) before sending their eager offer.
The tactic had a clearly unintended result—this customer was repelled.
The Slow Beat Down
There is a common digital marketing practice that re-presents the contents of page(s) previously viewed.
We’ve all experienced this: you view a page with a pair of shoes, or jewelry, or a Sub Zero and your fate is sealed. The product is re-presented again and again on different webpages days and often weeks later.
There are so many tragic assumptions in this practice. Among them, a view = purchase intention.
But there is no intelligence present here. The tactic simply applies one single, suspended data point: the customer viewed a page. With that isolated data point, the image is presented repeatedly as a slow beat down to purchase.
This is not marketing. This is regurgitation.
As a friend of mine said, “I don’t buy an $8,000 refrigerator every week”.
No Signs of Intelligence
Still other marketing efforts show no reflection of customer intelligence whatsoever. This is often despite having bountiful data about a spectrum of interests.
Take for example a page suggested by Facebook: that I join the Tennessee Deer Hunters group.
Just a few issues with the suggestion:
- I am not a deer hunter
- And I don’t live in Tennessee
Both are facts that Facebook knew, or should have known. They have my address not in Tennessee. And of all the likes and hearts and laughs or even wow’s, none is related to deer hunting. Or anything like deer hunting.
Here too Facebook took one isolated data element (a friend is a member!) and projected his characteristics onto me—all evidence to the contrary.
This is even more unforgivable given the sheer volume of data Facebook does have about me and my interests. It simply chose to ignore everything I reveal about myself through my direct input and actions and instead took a tactical shortcut. It failed.
For all the data Facebook has, the customer reasonably expects that input to be reflected back through understanding. There is absolutely none here.
Signs of Intelligent Life
Consider the companies used as examples here. They are all large companies, collecting massive data, assumedly with a legion of expert analysts available to drill into that data.
And yet they retreat to tactics. Tactics that ultimately fail.
In all cases, the tactical application of a single data point failed to drive a tactical objective—a sale.
The word “tactic” is intentional. Clear input prompts knee-jerk marketing output intended to prompt a sale. There is nothing elegant or complicated at work here.
And still they all fail.
There is so much potential from big data that will remain the great promise of potential until we roll up our sleeves and dig out the actual meaning. Whether from among heaps of data or in a massive binder of output, we necessarily must isolate those singularly meaningful opportunity signals.
And we are still not done, because any intelligent meaning then must be applied to create true customer strategies. Those strategies will improve customer experience, reflect understanding of interests and needs, make customers’ decisions easier, create internal efficiency, and ultimately improve revenue through mutually-beneficial customer relationships.
Only then will we narrow the gap between potential and realizing the true value in big customer data.
- Big Data, Great Potential: Why We Need to Do More with Big Data - August 3, 2017
- Seven Common Problems with Tech Marketing and What to Do Instead - June 8, 2017
- Are We Big Data Ready? Beyond the Easy Answers of Clicks and Response Rates - October 25, 2016