The opportunities and challenges line up pretty well here: how can we minimize the amount of behavioral data we need in order to develop accurate descriptions (and forecasts) for each individual? Most of the data we have is basically useless–particularly the non-behavioral measures. Too many retailers and analysts are obsessed with the torrent of new measures that have become available recently–but few have shown meaningful incremental value from them.

The “many sides of me” story points out the inherent limitations we face–no matter how much data we have, we’ll never come close to capturing all these behavioral facets for any individual. And bringing more data into the mix isn’t necessarily going to help. We have to learn to live with these limitations, and scale down our data wishlist in accordance with them.

Segmenting your consumer or shopper base is a great way to develop tailor-made strategies for different groups of consumers. It can be done with survey and/or shopper data using a variety of statistical tools. The key to a good segmentation–whether it’s based on demographics, psychographics, degree of customer loyalty, or shopper data–is making it actionable. Too often, companies spend lots of money to conduct segmentation studies only to have them sit on the desk of market researchers. Once you’ve segmented your customer base in a way that makes sense for your company, it’s imperative to identify a marketing approach that will have the most impact on each segment. Key questions that every marketer should ask: How does each segment relate to your brand? What is the best way to reach them? What are their sweet spots? How should the segments be priotized? Ultimately, it’s not how sophisticated your segmentation is that matters, but what you do with it. I’ve seen some ingenious segmentation schemes go to waste because companies didn’t do their research and the… Read more »

This is a great article that highlights a critically important issue–the same shopper can be of very different importance to different retailers (and different brands) based on what they buy from each. But the article should go further–the same shopper can be of very different importance to the *same* retailer (or brand) on a goal-by-goal basis. Let’s look at both retailer and brand examples. For a single retailer, say a full-line grocery chain, a frequent shopping family of four might be a critically important target for a basket-lift program where high-frequency shoppers are encouraged to spend more per trip. The same family might be a non-target for the big meat department promo, because they only buy non-meat products throughout the store (e.g., they are assumed vegetarians). For yet a third program, say an effort to get non-buyers of toilet paper to shop the paper goods aisle, the family might once again be an important target. Simply segmenting shoppers into tiers like “Platinum,” “Gold,” “Silver,” etc. would miss the nuance of all these programs. For a… Read more »

Mark registers strong points in this commentary. By listening to, and observing consumers throughout their purchase cycle (prior, during, and after), not only can big retailers remain as flexible as their smaller cousins, they MUST. Mark accurately points out that retailers must collect, collect, collect data/information. I’d be an advocate that surveys ARE the answer (and, yes, BIGresearch conducts syndicated and custom surveys for the world’s largest retailers in the U.S. and China). Just make certain that those surveys have a large enough sample size–statistics work. With a large sample, such as that of BIGresearch’s, of over 8,500 adults each month, big retailers achieve a margin of error of +/- 1%. It’s at that point that retailers need to zero in on: A. MINDSET – what is the consumers’ “Attitude” about the economy, confidence levels, life changes, personal finance, and YES, “Attitude” about the retailer’s goods and service B. NOW – what is the consumers’ “Behavior” in terms of shopping, saving, how they use credit, what motivates them to buy, use/influence of media, and “where”,… Read more »

My take is only one facet of this complex issue, but I still firmly believe: ASK your customer what they want from you. Not through surveys, but through personalization of experiences. Would the pizza place capture more of Mark’s business if they had online ordering and order history so that he could easily order the same thing over and over? If he could get a text alert based on his purchase history, which has figured out that he only contemplates ordering pizzas on Wednesdays? If that text alert went out at 4pm, promising delivery at 6pm? The best customer service, whether delivered in a high-end retailer or a low-end cost leader, helps customers achieve their objectives. In aspirational purchases, that might be as complex as REI’s travel division. In everyday purchases, that might be as simple as a weekly grocery list built off of past purchase frequency. However, one thing sticks with me when it comes to contemplating behavior vs. demographics. As the mother of a 5 year old, who still receives diaper and formula… Read more »

The article and accompanying comments focus on gathering data and pushing information to consumers. Companies should gather data from consumers, as much as possible. But first and foremost, they should engage consumers in a dialogue. Consumers control which message they will listen to and when they choose to listen. If they really care about a brand or experience, either positively or negatively, they will permit a dialogue to begin.

A positive dialogue can drive more sales through recommendations and word of mouth than a highly targeted advertising campaign. A negative dialogue can hurt sales.

Focus on engaging in a dialogue. Listen to what the consumer has to say and where she/he wants the dialogue to go. Then respond and keep the dialogue going.

“The Many Sides of Me” should also take into account that you are different people at different times. We saw data at a recent ARF Shopper Council that 50% of your “hard loyal” customers, who would not consider any other brand, are NOT hard loyal one year later. For that matter, since half of all supermarket shopping trips buy 5 or fewer items, I have often pointed out that these small basket shoppers are your stock-up shoppers, they just don’t happen to be stocking up on this particular trip!

The fact is, that if there are copious amounts of a given data available, people will try to make it tell them marvelous things. Sometimes the data speaks, but often you only hear your own voice echoing from the barrel.

None of this is intended to detract from the fantastic achievements of well executed loyalty programs.

It seems intuitive to me that our individual behaviors vary by circumstance and state of need. Historical efforts to segment shoppers into enduring groups based on observable demographic traits were admirable in their time, but are proving inadequate today. As an improvement, we now try to group shoppers by their past purchases and present attitudes, and tailor our messaging and merchandising accordingly. This behavioral segmentation opportunity quickly gets murky in multi-category retail environments, since each shopper or shopper group may interact differently with each category. The data requirements get really big really fast, and as a practical matter any retail can only pursue a subset of purchase behaviors that seem likely to have the best return on effort. In other words, we “roll up” behavioral data into manageable segments and still struggle to choose the right shopper marketing actions to pursue. So the core marketing question has expanded, from “Who is my customer and what does she want?” to “Who are my customers now; in what times, places and categories; and what are their needs… Read more »

The opportunities in this area are great. The challenge is in collecting and storing behavioral data in a manner that can be analyzed. The volumes of data can be unlimited. Identifying the best data to analyze is key. However, I do believe that smaller retailers have an advantage in this area. They have the opportunity to get to know their customers on a personal basis and even get to know them by name without the need for large, expensive databases. Even without any software applications, large retailers need to emphasize with their employees that their customers attitude and therefore sales can be influenced by them. For example, the other day, I went through a drive through where my order was taken. The girl on the speaker kept interrupting with “will that be all?” In addition, she never said thank you and never told me the amount of the order. When I asked the amount she responded “pull around.” When I got to the window she stuck her hand out for money while talking to another… Read more »

I like the ideas of segmentation but, as usual, I have something to say. Could this be over complicating the retail process? I recently had a great meeting with The Retail Prophet and we talked a bit about how overly complicated the analysis side of our business is. Customer segmentation is something that smaller merchants can really capitalize on. Mass merchandisers rely on product turns and traffic to succeed, so segmenting may not provide any real benefits to larger retailers. Loyalty programs are a different story though, and benefit from segmentation programs and matrices.

Is this complicated? Yes it is. If it was easy, we’d all be doing it. Bottom line, there are simple ways to view the customer. At the highest level, a segment of customers–rather than focusing on individuals–can often reap most of the insights that return the benefits. It’s an 80/20 thing. However, to get truly insightful results, a retailer or CPG company needs to drive consumer value with a “Single View of the Customer.” This follows the customer through multiple sales channels, multiple shopping missions and can leverage predictive analytics to create a far more proactive marketing and sales approach. This kind of information gathering and management can be a challenge for retailers that sell through multiple channels, to say the least. Retail organizations are overwhelmed with the amount of customer data flooding their companies as a result of the Web, catalog and loyalty programs. Often, this information is kept separately within each channel, with no consistency or effort to normalize and share across the enterprise. There are multiple versions of the truth, with data… Read more »

This is one of the best articles I have seen on RetailWire. Mark is on target. The conclusions he suggests ultimately will lead to one-on-one marketing.

Segmentation is the result of what has been measured not what information is needed. To paraphrase the old Wanamaker quote, “Half the people who see my demographically target ad, never come in my store. I just don’t know which half.”

Small retailers have been using one-on-one marketing for years. It includes everything from knowing my name to having my drink waiting for me on the bar at the local pub to knowing how I like my hair cut.

Now there are a few more retailers out there who know the precise me. Netflix is very successful in suggesting movies I would like. Amazon knows an entire spectrum of items that would interest me. Drugstore.com knows when it is time for me to reorder. A retailer can’t get much more one-on-one than that. Unfortunately, most are not and cannot get that connected.

The corollary to “The Many Sides of Me” is “Who Are You to Me”. That’s the key question for retailers. Who is each customer, based on their behavior, and how can we increase our importance to them? It goes beyond simple segmentation and drills down to the one-on-one relationship.

This is something that the very best independent retailers have long understood and practiced. It takes the customer beyond being a mere data point, to a unique individual, and allows relationships to be built and nurtured over the long haul.

Applying these ideas and approaches to the mass-market world of chain retailing is a daunting challenge. Their business models are built around scale rather than individualized relationships. Technology can help, but only people can truly build relationships. That’s why my instinct tells me Mark’s ideas will have far greater traction in high-end and independent retailing than in mass-market chain retailing, where price is still the unchallenged driver.

When retailers take on the task of surveying customers, some lose sight of the end-goal: putting that knowledge into action to develop relationships rather than just transactions (as Mark Price knowingly puts it). The skeptics don’t understand the realities of how many individuals will actually take a survey if prompted and provided with an incentive. It can be a challenge to embark on the task of understanding behavior, as it’s a full-circle initiative that involves asking the right questions; collecting the data; analyzing the data; discovering areas for improvement; identifying what prompts customer behavior; making appropriate changes that help to influence that behavior in a positive way, and finally asking more questions to see if those changes did influence the customer. Companies big or small can certainly do this, but they must be committed to developing a strategy, not just doing a quick customer survey. Surveys alone, though while proven highly effective in learning customer sentiments, are not the only way to segment individual behaviors. There are in-store texting tools that allow retailers to get… Read more »