Compelling, creepy, annoying or just bad? Retail’s personalization opportunity.

It’s hard to believe it’s been over 20 years since Martha Rodgers and Don Peppers’ seminal book The One to One Future. At the time, Dr. Rodgers and Mr. Peppers (not to be confused with Mr. Rodgers and Dr. Pepper!) offered up the radical notion that mass, one-size-fits-all marketing would begin yielding to a brave new one-to-one world. Followed just three years later by Seth Godin’s classic Permission Marketing the more intrepid among us started to make “treat different customers differently” our mantra and advocate for a shift to more targeted and personalized campaigns. Alas, we were a bit ahead of our time.

Despite years of missteps and hype, some two decades later the business case for greater marketing and experiential personalization remains strong. Fortunately, lower cost data storage and more effective technology solutions, along with general advances in know-how and the ability to reach customers through digital channels, now make it possible for most retail brands to realistically differentiate themselves on the basis of deep customer insight, data science and advanced targeting strategies. From where I sit, it won’t be long before advanced personalization skills become table-stakes in the battle for customer share of attention. To remain relevant — to become the signal amidst all the noise — retail marketers will have to get good at one-to-one marketing and in delivering more personalized experiences both in the store and on the web.

Yet, despite the strong business case, advancing capabilities and many years of experimenting, personalization’s potential remains largely untapped. For every success story, it seems as if there are dozens of weak efforts or outright debacles. In fact, a recent study by Accenture estimates that personalization failures cost US firms $756 billion and a total of $2.5 trillion globally. While I have a hard time getting my head around the accuracy and magnitude of those numbers, there is no question poor data management and far from stellar personalization can chase away business as well as leave a lot of money on the table.

As we start to understand how to both avoid problems and seize on opportunities, I find it’s worth asking a few basic questions.

Is it compelling?

The essence of good personalization is two-fold: is it relevant and is it remarkable? Delivering intensely relevant one-to-one (or mass customized) experiences is predicated on deep customer insight and the ability to target the right interaction (or offer) to the right customer at–or as close as possible–to the right moment. Retailers that are getting it right use data science to ascertain customer needs and wants and to better predict the next most effective marketing action. Stitch Fix is a great example of a company that has built predictive analytics and targeted marketing into the fabric (heh, heh) of their enterprise. The other key element is “remarkability.” Even if an offer is relevant, simply serving up the same old tired promotional tricks is unlikely to get a good response and help enhance the brand’s image. According to the Accenture study, 44% of all customers feel that brands fail to deliver relevant personalized experiences. Plenty of untapped opportunities here.

Is it creepy?

In my experience, the vast majority of customers have no idea how easy it is for marketers to purchase potentially useful pieces of data to better inform their targeted marketing strategies. Moreover, many customers fail to grasp how their lack of attention to privacy settings on places like Facebook allows marketers to glean all sorts of insights from the data breadcrumbs left behind by our traffic, likes and so on. Advances in statistical techniques and artificial intelligence allow for powerful inferences to be made by analyzing behaviors, transactions and demographic information. Walking the thin line between delivering surprisingly useful recommendations and something that smacks of Big Brother –or that raises unnecessary privacy concerns–is challenging. In the bloodthirsty quest for incremental revenue, it is all too easy for undisciplined marketers to step over the line. Resist the temptation. Strong brands are based on trust. Tread lightly.

Is it annoying?

I’ve met few marketers that believe less is more. For most, more is more, often to the point of going well beyond diminishing returns. Since email (and certain other digital messages) are often quite cheap at the margin, retail marketers often take the bludgeon approach to their campaign messaging. They dial up frequency until we yell “Uncle.” They chase us all over the internet with retargeting ads. They offer us products we just bought (oh yeah, sure I often buy a second dishwasher or espresso machine the week after I bought my last one). The holiday shopping season is a particularly bad time of the year where frequency goes to 11 and many promotional strategies look like they were created by Jackson Pollock. Just because you can, doesn’t mean you should.

Is it just bad?

In 2011 I started pointing out when bad personalization happens to good people and it’s become a bit of a hobby for me (apparently I have that kind of time). A certain airline (I won’t tell you which one, but their initials are “AA”) regularly sent my teenage daughters offers “specially selected” for them which included deals for mortgage financing. We were nicely generous with their allowances, but not enough for any real estate speculation. Neiman Marcus (where I once, ironically, oversaw our customer insight and personalization efforts) often encouraged me to redeem my InCircle Rewards points. Which would be great if I actually had any. Citibank still pitches me a credit card I already have, while AT&T, um, well where to start?

The first rule of personalization club is to not ask a customer to provide information that you already have (unless it’s to verify identity). The second rule is to demonstrate that you know the customer and understand their relationship with your brand. Any offer that belies that is likely to make a brand look dumb. The third rule is to show the customer that you value them: value their time, their spending, their loyalty, the exchange of information they may have provided you. Don’t waste a customer’s time by misusing their data, failing to protect their privacy, trying to sell them stuff they already own and not making a real effort to treat different customers differently. Don’t mistake simple or cheap for useful or effective.

Personalization is not easy. But the revolution sweeping retail demands that brands get more relevant, more differentiated and more remarkable. And fast. For many, delivering more personalized experiences and marketing may be the difference between success and being roadkill in the age of Amazon and digital disruption.

The changes that many brands need to make are not insignificant. They typically require new technology, new people, new processes, new metrics, material incremental investment and a willingness to aggressively experiment. But to paraphrase Eric Shinseki, “If you don’t like change, you’re going to like irrelevance even less.”

A version of this story appeared at Forbes, where I am a retail contributor. You can check out more of my posts and follow me here.  For information on keynote speaking and workshops please go here.