Personalization is about the person. Some may say that I am stating the obvious. Really? Then explain this.
Some time ago, I bought a new garden hose nozzle for my wife, as she took up a new hobby of cultivating vegetables that we actually consume. The last time I bought such an item was more than 15 years ago, so I did some online research (but don’t automatically label me as an “online” person yet). After a few clicks, I ended up on a good old Amazon site, and bought what I would call the Cadillac of garden hose nozzles. Not only did it come with all kinds of options, intuitive handle and switches, it also was in Ferrari red. All in all, I had a very positive shopping experience.
Then a not-so-great customer experience started to happen. OK, it was still kind of cute that they showed “more of the same” items after I checked out. It is entirely possible that I’d find something even nicer seconds after the transaction, and cancel the original order and buy an alternate item. What was really annoying was that this almighty Amazon started to send a series of emails only featuring, well, guess what, more garden hose nozzles! Ah, maybe they pinned me as a “gardening enthusiast” with some fancy algorithm? Or they thought that I started a wholesale business selling nothing but fancy nozzles? Or I just became a collector of nozzles without realizing it? Then again, was I giving too much credit to their analysts?
The answer revealed itself through my next purchase with them just a few weeks later. Like many modern households, we have lots of computers scattered around our house (although not all of them could be networked together, thanks to not-so-intuitive Microsoft operating systems). One of the older computers was still with a seriously outdated mouse, so I decided to replace it with a newer model that I tried and liked. That means I needed zero research time for that purchase. Just log onto Amazon, type in the brand and model number, and do that famous one-click checkout. Simple, right?
Guess what happened after that purchase? I started to receive a new series of emails from them, this time featuring nothing but more mice! What do they think, I love this mouse so much that I would start a mouse farm? Do they want me to find a better mouse “after” I purchased one already? The “last” thing I would buy for the rest of this fiscal year is another mouse (and another garden hose nozzle, if I might add). I cannot forgive their oversight, because I bought the first mouse with the same merchant, only about 16 months ago. Don’t they have my personal transaction history? For heaven’s sake, I can just log onto Amazon.com and check out what I have been buying from it going back almost 10 years! Why don’t they use such rich data? Isn’t Amazon supposed to be one of the leading database marketers?
How did this all happen? I have two words for you: “Collaborative Filtering,” though I have no idea what they are really collaborating in cases like these. That term has been around for some time now, actually. It basically means, “Oh, you bought this item? You may like these other products, too”-type marketing through some algorithm. Now I know that when we use terms like “algorithmic solution,” we may feel a little smarter about ourselves — as in, “Yeah! I am not afraid of math!” But let’s forget about how the math works, and let’s think about how the consumers feel about it.
If I may share my blunt sentiment about this type of suggestion engine, my language would have to be more colorful than this fine publication allows me to be. Let’s just say that I am pretty far from impressed. And marketers should not even think about calling this barrage of emails “personalized.” They are much closer to spam than personalized emails, because I know this type of personalization is based on products, not people. And this so-called “machine learning” becomes nothing more than a nuisance, if the all-important human touch is missing from the equation.
Clearly, “personalization” is the buzzword du jour, actively pushing “Big Data” out of its short-lived glory. We can see that trend at conferences, marketing meetings, industry papers and blogs like this. And unlike other buzzwords that came and went in the marketing industry, I am boldly predicting that this “personalization” is here to stay for a foreseeable time. Why? Because consumers demand it, they feel that they are entitled to it, and marketers finally have the technology and data at their command to do it. But alas, only if they do it right. Unfortunately, I see a lot of marketers – even so-called leading online marketers – just personally annoying the heck out of their customers. As a result, it is very difficult to find good success stories about personalization. This, even as everyone demands case studies (as if they won’t commit to it, unless others succeed in that endeavor first).
So, I ask marketers this question: Are you really committed to do it, or are you just saying that word simply because it is a new thing to do now? I ask for commitment as an organization, because doing it right requires a lot more than just purchasing a personalized engine off the shelf.
Recently, I attended an eTailer conference and happened to sit next to a digital marketer at a networking lunch (i.e., a free piece of meat with a salad). When I tried to explain what I do for personalization efforts, she stopped me and said, “Oh, personalization engines do that for us.” Really? To me, that sounds a lot like saying that coffee comes out of an espresso machine. I didn’t say anything at the time (I was busy chewing the meat), but I believe that such a myopic view is the main cause for all of those rudimentary and ineffective personalizations. For you to enjoy that cup of coffee ever so conveniently, someone had to cultivate coffee plants (I bet without any fancy hose nozzle), harvest the beans, process them, transport them, do all the paperwork to go through customs, domestically distribute them, roast them to various degrees and package them for espresso machines. Likewise, for personalization engines to function properly, incoming data must go through some serious refinement process.
Without a doubt, proper personalization starts with a personalized data view, which is skipped over all too often. Some may use terms like “360-degree customer view,” “single customer view” or my favorite, “customer-centric portrait.” No matter. All the transactional, behavioral, demographic and environmental data must be realigned around “each” customer or prospect. Some may say that they already have some fancy ID system that connects all those data points (many don’t). Great, but that is just a good beginning. We still need to convert such “event”-level data into “descriptors” of individuals. Transaction-level data may tell you what happened on a certain date, for how much money and for what product. Descriptors of individuals display their personal spending patterns, such as personal compositions of categorical purchases and browsing history, frequency of purchases and spending levels in each category or channel, and sets of times series variables nicely lined up around the person (refer to “Beyond RFM Data”). This is quite different from stacks of transaction or event-level data sitting in data platforms designed for mass storage and rapid retrieval.
When we line up information around people, we often find out that we really do not know much about our customers. All those fancy variables created around the target individuals have many holes in them, for various reasons. Maybe they are new customers, or they just browsed a few items but never bought anything yet. Some customers may have shopped only in certain categories, but stayed away from others. Some customers may have been very diligent in deleting their online trails. To do the personalization properly and consistently, we need to fill in such gaps.
Most of personalization engines, unfortunately, are designed to act only on available (largely, “known”) data. When marketers go too far with what’s known to them, the customers who casually let some parts of their lives known to marketers get bombarded with the same messages until they get completely sick of them. That is a sad situation as, categorically speaking, people with known behaviors often account for less than – at times far less than – 5 percent of the approachable universe. So, in that scenario, 5 percent get to be stalked, while 95 percent are ignored. Not ideal at all.
Enter statistical modeling. I have been emphasizing the importance of statistical modeling even in the data-rich environment, because we will never know everything about everyone, and statistical modeling systematically converts “unknowns” to “potentials.” No, we may not know for sure that a particular target is indeed a “gardening enthusiast” (and no, buying just “one” garden hose nozzle may not be enough). But yes, we can say that she is “very likely to be” a gardening enthusiast, with statistical techniques effectively mining available data — such as what other products she purchased and browsed with varying frequencies and intervals. The results of the models are “scores” by which you can measure the degree of confidence, as in a nine out of a 10 scale. This is much simpler than having to worry about hundreds of variables with more holes Swiss cheese.
Building a customer-centric view and filling in the gaps with modeling techniques is far more superior to a personalization engine that would just ingest unrefined SKU-level data. For one, we don’t get to bother people just because we had a glimpse of certain behavior, as statistical modeling considers hundreds, if not thousands, of variables around the person. Secondly, having “potential” values for certain behavior enables marketers to act on most of the targets, not just fractions of them. Going further, we can even estimate channel and timing preference in addition to what we often call “personas” or propensity scores.
The result of modeling work will make the personalization engines run better, too. After all, those software solutions are designed to ingest any type of variable. And the model scores – which are summaries of hundreds of data points – look just like another set of variables, anyway. Consider such scores without any missing values as really tasty coffee beans that you can put into your shiny espresso machine.
Country store owners in the old days were known to have personal touches, because they treated their customers as people. (Well, of course without being creepy. Refer to “Don’t Do It Just Because You Can”). They would not have offered more hammers to you just because you just purchased a hammer. They would have put it in context (i.e., human touch), and then they would have suggested products that you may benefit from. (As in “Hey, don’t you need protective gloves, too? I know you’re a klutz!”)
Now we have access to enough technology, mathematical skills and data to do such personalized marketing to millions of people at a time. But it will work only if marketers do not lose sight of what matters, and commit to the proper way, preparing the data specifically for personalization efforts and programming personal touches into algorithms. Technology made things easy for us, but it is equally easy to abuse it. And let’s not forget that we are just personally abusing other human beings when we abuse technology and data. We’d better not call that personalization.