Don’t Blame Personalization After Messing It Up

In late 2019, Gartner predicted “80% of marketers who have invested in personalization efforts will abandon them by 2025 because of lack of ROI, the peril of customer data, or both.” But before giving up because the first few rounds didn’t pay off, shouldn’t marketers stop and think about what could have gone wrong?

In late 2019, Gartner predicted “80% of marketers who have invested in personalization efforts will abandon them by 2025 because of lack of ROI, the peril of customer data, or both.” Interesting that I started my last article quoting only about 20% of analytics works are properly applied to businesses. What is this, some 80/20 hell for marketers?

Nonetheless, the stat that I shared here begs for further questioning, especially the ROI part. Why do so many marketers think that ROI isn’t there? Simply, ROI doesn’t look good when:

  1. You invested too much money (the denominator of the ROI equation), and
  2. The investment didn’t pay off (the numerator of the same).

Many companies must have spent large sums of money on teams of specialists and service providers, data platforms featuring customer 360, personalization software (on the delivery side), analytics work for developing segments and personas, third-party data, plus the maintenance cost of it all. To justify the cost, some marginal improvements here and there wouldn’t cut it.

Then, there are attribution challenges even when there are returns. Allocating credit among all the things that marketers do isn’t very simple, especially in multichannel environments. To knock CEOs and CFOs off their chairs – basically the bottom-line people, not math or data geeks – the “credited” results should look pretty darn good. Nothing succeeds like success.

After all, isn’t that why marketers jumped onto this personalization bandwagon in the first place? For some big payoff? Wasn’t it routinely quoted that, when done right, 1:1 personalization efforts could pay off 20 times over the investment?

Alas, the key phrase here was “when done right,” while most were fixated on the dollar signs. Furthermore, personalization is a team sport, and it’s a long-term game.  You will never see that 20x return just because you bought some personalization engine and turned the default setting on.

If history taught us anything, any game that could pay off so well can’t be that simple. There are lots of in-between steps that could go wrong. Too bad that yet another buzzword is about to go down as a failure, when marketers didn’t play the game right and the word was heavily abused.

But before giving it all up just because the first few rounds didn’t pay off so well, shouldn’t marketers stop and think about what could have gone so wrong with their personalization efforts?

Most Personalization Efforts Are Reactive

If you look at so-called “personalized” messages from the customer’s point of view, most of them are just annoying. You’d say, “Are they trying to annoy me personally?”

Unfortunately, successful personalization efforts of the present day is more about pushing products to customers, as in “If you bought this, you must want that too!” When you treat your customers as mere extensions of their last purchase, it doesn’t look very personal, does it?

Ok, I know that I coveted some expensive electric guitars last time I visited a site, but must I get reminded of that visit every little turn I make on the web, even “outside” the site in question?

I am the sum of many other behaviors and interests – and you have all the clues in your database – not a hollow representation of the last click or the last purchase.  In my opinion, such one-dimensional personalization efforts ruined the term.

Personalization must be about the person, not product, brands, or channels.

Personalization Tactics Are Often Done Sporadically, Not Consistently

Reactive personalization can only be done when there is a trigger, such as someone visiting a site, browsing an item for a while, putting it in a basket without checking out, clicking some link, etc. Other than the annoyance factor I’ve already mentioned, such reactive personalization is quite limited in scale. Basically, you can’t do a damn thing if there is no trigger data coming in.

The result? You end up annoying the heck out of the poor souls who left any trail – not the vast majority for sure – and leave the rest outside the personalization universe.

Now, a 1:1 marketing effort is a number’s game. If you don’t have a large base to reach, you cannot make significant differences even with a great response rate.

So, how would you get out of that “known-data-only” trap? Venture into the worlds of “unknowns,” and convert them into “high potential opportunities” using modeling techniques. We may not know for sure if a particular target is interested in purchasing high-end home electronics, but we can certainly calculate the probability of it using all the data that we have on him.

This practice alone will increase the target base from a few percentage points to 100% coverage, as model scores can be put on every record. Now you can consistently personalize messages at a much larger scale. That will certainly help with your bottom-line, as more will see your personalized messages in the first place.

But It’s Too Creepy

Privacy concerns are for real. Many consumers are scared of know-it-all marketers, on top of being annoyed by incessant bombardments of impersonal messages; yet another undesirable side effect of heavy reliance on “known” data. Because to know for sure, you have to monitor every breath they take and every move they make.

Now, there is another added bonus of sharing data in the form of model scores. Even the most aggressive users (i.e., marketers) wouldn’t act like they actually “know” the target when all they have is a probability. When the information is given to them, like “This target is 70% likely to be interested in children’s education products,” no one would come out and say “I know you are interested in children’s education products. So, buy this!”

The key in modern day marketing is a gentle nudge, not a hard sell. Build many personas – because consumers are interested in many different things – and kindly usher them to categories that they are “highly likely” to be interested in.

Too Many Initiatives Are Set on Auto-Pilot

People can smell machines from miles away. I think humans will be able to smell the coldness of a machine even when most AIs will have passed the famous Turing Test (Definition: a test of a machine’s ability to exhibit intelligent behavior equivalent to, or indistinguishable from, that of a human).

In the present day, detecting a machine pushing particular products is even easier than detecting a call-center operator sitting in a foreign country (not that there is anything wrong about that).

On top of that, machines are only as versatile as we set them up to be. So, don’t fall for some sales pitch that a machine can automatically personalize every message utilizing all available data. You may end up with some rudimentary personalization efforts barely superior to basic collaborative filtering, mindlessly listing all related products to what the target just clicked, viewed, or purchased.

Such efforts, of course, would be better than nothing.  For some time.  But remember that the goal is to “wow” your target customers and your bosses. Do not settle for some default settings of campaign or analytics toolsets.

Important Factors Are Ignored

When most investments are sunk in platforms, engines, and toolsets, only a little are left for tweaking, maintenance, and expansion. As all businesses are unique (even in similar industries), the last mile effort for custom fitting often makes or breaks the project. At times, unfortunately, even big items such as analytics and content libraries for digital asset management get to be ignored.

Even through a state-of-the-art AI engine, refined data works better than raw data. Your personalization efforts will fail if there aren’t enough digital assets to rotate through, even with a long list of personas and segments for everyone in the database. Basically, can you show different contents for different personas at different occasions through different media?

Data, analytics, contents, and display technologies must work harmoniously for high level personalization to work.

So What Now?

It would be a real shame if marketers hastily move away from personalization efforts when sophistication level is still elementary for the most.

Maybe we need a new word to describe the effort to pamper customers with suitable products, services and offers. Regardless of what we would call it, staying relevant to your customer is not just an option anymore. Because if you don’t, your message will categorically be dismissed as yet another annoying marketing message.


Identifying and Engaging Your Most Valuable Audience Segments

Your audience is made up of a series of segments of audiences, all in a different place in their journey with your brand. From one-time visitors to the most loyal readers, each segment requires a different strategy for engagement and something different from your brand experience.

Your audience is made up of a series of segments of audiences, all in a different place in their journey with your brand. From one-time visitors to the most loyal readers, each segment requires a different strategy for engagement and something different from your brand experience.

To optimize your audience development strategies for 2020, you need to first identify these audience segments and then determine a plan for engaging with and growing each segment.

For the purposes of this blog post, we’ll use Google Analytics to create audience segments, as most readers will be familiar with and have access to the tool. There are, of course, more advanced tools like CDPs that will allow you to act even more strategically.

Note that when creating segments in Google Analytics you can typically look at a segment’s behavior for a maximum of 90 days. So for the purposes of analyzing these groups, we’ll be looking at them in terms of their behavior within a 90-day period.

Segmenting Your Online Audiences

This article overviews one way to segment your online audience into four categories. If this isn’t the right way for your brand, you can segment in a way that works for you. The lesson here is to develop your segments and then create a plan that prioritizes and engages each accordingly.

1. Drive-bys

Drive-bys visit your site once and likely won’t again, at least not for another few months. More than likely they stumbled onto your website through a Google search or a friend sharing a link on social media. They may or may not be in your target audience.

Find these users by creating a segment in Google Analytics and filtering by users who have exactly one session on your website. If your target audience is based on a geographic region, you can create two segments, one within that region and one without.

How to Engage Your Drive-bys

Drive-bys are the audience segment that is likely to be on the bottom of the totem pole in terms of priority, so your strategy for engaging them should be minimal. Monetize their visit with programmatic advertising, but invest little other efforts in engagement.

2. Passive Visitors

Passive visitors to your website are those who return to engage with your brand two or three times in a 90-day period. The fact that they have returned to your website after their initial visit suggests they might be within your target audience and should be engaged accordingly.

One way to identify your passive visitors is to create a segment in Google Analytics that filters users who have exactly or greater than two sessions on your website and more than 30 days since their last session. The second part of this criteria will eliminate your more frequent visitors — we’ll get to those soon.

How to Engage Your Passive Visitors

Passive visitors have much more value to your brand than the drive-by visitors. Based on their return to your website, they are more likely to be in your target audience and should be considered warm leads for your brand.

Keep passive visitors on clean, clutter-free website pages to avoid overwhelming their experience with advertising, if possible. You can do this by exporting audiences from your segments into DFP (Google’s ad platform). The ideal outcome is to secure their email address capture, so present these users with opportunities to sign-up for one of your weekly (or lower frequency) newsletters.

If you’re unable to capture an email address, happily settle for a social follow by targeting these users in DFP with ads driving to your social channels, or retarget these visitors on social media using the same process above.

3. Engaged Visitors

Highly engaged visitors are regularly engaging with your brand and are visiting your website at least once per month. Get excited because this is where your marketing strategies get fun.

When creating the segment in Google Analytics, filter by users who have exactly or greater than five sessions on your website. Based on our work with regional and niche publishers, these visitors are likely to be less than 10% of your total users, yet there’s a good chance they generate more than 40% of your website’s page views, making them an incredibly valuable audience segment.

How to Market to Your Engaged Visitors

These users are already engaged with your brand, so the next step is marketing more of your content to them and deepening their connection with it. If you do not have email addresses in your database for these users, your number-one goal should be to capture them. If you do have an email address, you should be using your email service provider to monitor their behavior with your emails and fine-tuning your content delivery based on their interests. For example, if they are recurringly visiting your food coverage, deliver them news about restaurant openings first or make sure they know about your upcoming food event.

Further engagement includes driving these readers back to your website with advertising on Facebook and Instagram. Consider experimenting with soft gates on popular pieces of content to force an email capture to read the articles. But keep their experience clean and clutter-free. This isn’t the audience you want to hit over the head with invasive or pop-up advertising. This audience is primed for a deeper relationship with your brand and your marketing actions can drive them to your site more frequently — or drive them away.

4. Loyalists

Loyalists visit your website at least 15 times over a quarter, which equates to visiting more than once per week. These folks are very likely to be on your email newsletter list or are highly engaged with your brand on social media. Based on what we see from our clients’ brands, loyalists may comprise less than 3% of your visitors but could be making up as much as 25% of your page views.

Create a final segment inside Google Analytics that filters by users who have exactly or greater than 15 sessions on your website. Remember you’ll be looking at these segments over a period of 90 days.

How to Market to Your Loyalists

Like your ‘engaged’ visitors, these individuals are already invested in the content you are sharing, and now your goal is to drive revenue from them.

If you have a metered paywall in place – or are considering putting one on your website – these users are the ones who will hit it. If you have a subscription product to sell, these users will feel the most inclined to support your brand financially.

A great way to extract more value out of these users is to get their feedback. As regular consumers of your content, they are more likely to share their time and their opinions. Whether you’re considering a new product launch or a shift in editorial coverage, this audience’s opinions will be valuable.

Two common ways to solicit feedback are through a traditional survey – typically sent via email (SurveyMonkey is easy to use for something like this) – or through a focus group. The latter especially allows you to gain a deeper understanding of the content they love, what kind of products they want from your brand, and most importantly what they would be willing to pay for.

Not All Audiences Are Created Equal

We know by now that not all audiences are created equal, so your engagement strategies shouldn’t be either. These high-level strategy suggestions are the beginning of engaging your various website audiences differently to make the most out of your time and marketing resources.

How to Employ Segmentation to Improve Your Content Marketing

Evaluating your content marketing specifically for each audience segment will yield insights that a program-wide analysis won’t capture. Audience segmentation isn’t just good for reaching the right people with the right message. Done well, it can help you learn more about your audience.

Audience segmentation isn’t just good for reaching the right people with the right message. Done well, it can help you learn more about your audience. And learn more about how better to meet their needs.Your content marketing should be a part of that process.

  • First, if you are’t creating content specifically for different audience segments, please start doing so now.
  • Second, if you aren’t creating your audience segments based on their attributes and behavior, that’s another change you should make immediately. (“People who buy Product A from us” is not an effective audience segment.)

Assuming you do have useful audience segmentation in place, here’s how you can use it to learn more about your audience.

All Content Is Not Created Equal

Begin by evaluating your content marketing efforts on their own. Identify the 20% of your content that performs best and the 20% that performs worst. (We’ll come back to those bottom-of-the-barrel content elements in a bit.)

Your evaluation can be based on key performance metrics, running the gamut from page views to revenue generated. But you should include a range of process metrics and outcomes metrics.

We define process metrics as those data points, like page views, time on page, CTRs, etc., that can provide valuable insight into your audience’s interests, but don’t measure actual business performance. Outcomes metrics are those that relate to revenue generation, lead quality, lead volume, and so on.

You may be tempted to lean more heavily on outcomes metrics, if you have them. They are clearly more important in the long run. Page likes don’t pay the bills, after all. But our goal is to evaluate the health and effectiveness of our entire content marketing program. So, understanding how well we’re doing with early-stage prospects is important. The data points from early funnel activity are almost always going to be process oriented — content consumption, measures of engagement, micro-conversion numbers.

Cross-Referencing Your Results

With your raw performance information in-hand, look at these numbers again — broken down by audience segment. Here’s where you’ll see real value. If you can identify what content is resonating best with each audience segment, you can tailor programs to those audience segments, based on what they’re most interested in.

Rather than simply trying to double down on your best-performing content, you can provide content that performs best for each audience segment.

What About Your Underperforming Content?

You may also find that, rather than eliminating your less effective content, you can tailor it to a specific audience segment and have it perform much more effectively. This will likely require a deeper dive into whether any of the laggards are weak overall, despite being strong in one particular area.

Some of this work may require updates to your coding or your analytics reporting. Discuss what that investment is going to be with your tech team. Chances are, costs will be recouped quickly.

Once you get the hang of this approach, you’ll see benefits beyond your immediate results. This kind of deeper dive into your analytics data can help you evaluate information from disparate parts of your marketing efforts and yield insights that can impact your broader marketing effectiveness.

How Will Your Audience Receive Your New Product?

Product innovation is necessary for every company to grow and evolve in a competitive market. But if your audience “doesn’t get” your new product, success is much less of a guarantee.

Product innovation is necessary for every company to grow and evolve in a competitive market. But if your audience “doesn’t get” your new product, success is much less of a guarantee. Before you unveil your hard-won innovations, here are some ways to ensure you’re targeting the segments of your audience who will be the most receptive — both to the new product and accompanying marketing efforts.

First, Really Know Who They Are

While basic demographics like age, marital status, geographic location, hobbies and other points help you form a picture of your audience, to really know them means gaining specific, unique insights about them. You want to understand more than just who they are on paper by finding out how they think and feel and what they truly need. To do this, you have to integrate survey data with rich behavioral insights gleaned from big data.

Look at how personality profiles developed through a scan of big data reveal the personality characteristics common to the potential target audience for a new robot vacuum:

Credit: GutCheckIt

This audience ranks high for agreeableness, which points to other traits like altruistic, modest, and empathetic. So when communicating with them about the vacuum, messaging that uses a social responsibility angle will likely attract and feel relevant to them.

How your new product appeals to the individual needs and lifestyles of your audience further deepens your understanding of them. Consider in this summary of needs how the robot vacuum could hit home with the audience’s high ideals, drive toward harmony, and interest in self-expression, as well as how the vacuum could appeal to the audience majority who enjoy keeping their home tidy.

Credit: GutCheckIt
Credit: GutCheckIt

Then, Determine How Best to Reach Them

Once you’ve formed a full understanding of your audience’s personality, needs, and lifestyle, combine your learning with a study of the type of media consumed and during which times of day. For example, the vacuum audience learns about new products mainly through social media rather than television or promotional emails. They spend 7-plus hours per week on the web and using apps, mostly in the early evening hours between 5-8 pm.

Credit: GutCheckIt

To reach this audience effectively, online or mobile campaigns work best, with ads that could be shown on traditional TV in the later evening hours between 8-11 pm.

To learn what type of unique insights you could uncover about your brand’s audience before you launch a new product, visit the GutCheck website to learn more.

Almost the Ultimate in ‘Not Interested’ Segmentation

If data is the fuel that is powering today’s marketing engine, Google has discovered a real gusher. Google, or any other AI-aided advertising sales effort, can add these datasets to already copious databases and use them as the almost ultimate tool to segment advertising messages to people most likely to be interested in them and to avoid sending ads in specific categories to those people who have signaled they are “not interested.”

I couldn’t quite believe my eyes.

Peter J. Rosenwald illustration one
Credit: Peter J. Rosenwald

That “i” with a circle around it and the accompanying “X” may have been there before, but I must never have noticed. They certainly didn’t leap off the page at me (that’s why I added the arrow) and would have had a hard time competing for attention with those cool 3D T-shirts.

Credit: Peter J. Rosenwald

But there they were, hiding in the upper right hand corner of lots of ads, placed every five paragraphs or so by those lovable folks at Google, certainly intended to interrupt my reading of every salacious breaking news story about the White House and porn star Stormy Daniels (upright in real life, Ms. Stephanie Clifford).

Seduced away from learning the latest secrets of how to earn $130,000 for allegedly having a soft porn one-nighter with a presidential candidate who, between rallies imploring the U.S. electorate to make America great again, found time for a little R & R, I became intrigued by those mysterious letters and moved my cursor to discover what they were telling me.

Credit: Peter J. Rosenwald

OK as far as it went, but by now in a state of aroused curiosity about the encircled “i” and accompanying “X,” I wanted to know more. Click on the encircled “i” and it takes you to AdSense, a website providing everything you ever wanted to know but perhaps never thought to ask, about Google’s targeted advertising and data protection policies.

Click on the “X” and here is what you get.

Credit: Peter J. Rosenwald

Google has now neatly positioned me to click on “Stop seeing this ad” or “Ads by Google,” un-highlighted and again accompanied by the encircled “i.” Addicted as I am to Sherlock Holmes, I felt compelled to move ahead and to click the “Stop” button.

But before I did, I began wondering; what’s in this for Google other than having delivered a possibly unwanted ad, then creating a nice warm “feel good” atmosphere. It’s rather like the guest who tracks mud into your living room, and then apologizes and promises to try not to do it again.

Credit: Peter J. Rosenwald

Now I get it.

It’s a brilliantly laid back survey generating invaluable data about;

  • Those “Not interested in this ad,” those on whom promotion money for this category should not be wasted.

The first time I clicked on “Not interested,” I was taken to a section of the site which showed me a wide range of interest categories and asked me to eliminate those for which I had no interest. As I’d indicate one, it would go away and another one would pop up in what became an endless five-minute project. Wanting to show it here, I tried again to find it, but Google had obviously gotten what it wanted from me and it mysteriously disappeared.

  • Those who didn’t have anything against the specific category but had seen it “multiple” times (too many), that number informing Google to limit the ad frequency for this person;
  • Those who consider the ad “inappropriate,” a clear signal to Google to send only “Jello’” ads here; no “Tamale flavored hot, spicy yogurt.”
  • When I clicked on “Ad-covered content,” I received a message that the ad had been ‘closed’ by Google, nothing else that might have explained what “Ad-covered content” might have meant.

If data is the fuel that is powering today’s marketing engine, Google has discovered a real gusher. Google, or any other AI-aided advertising sales effort, can add these datasets to already copious databases and use them as the almost ultimate tool to segment advertising messages to people most likely to be interested in them and to avoid sending ads in specific categories to those people who have signaled they are “not interested.”

It’s a win, win. For Google, because it should be a very sexy addition to its advertising sales platform. For advertisers, who must applaud a new ability not to spend the marketing budget talking to people who do not wish to hear their message.

Why Modeling Beats Rule-Based Segmentation

I have been talking about “employing all available data” for targeting and customer insights for some time now. So allow me to pick a different bone today. Let’s forget the data part, and talk about the methodology. When machines can build models super-fast, aversion to modeling only limits the users. After all, I am not asking any marketers to get a degree in statistics. I am just asking them to consider modeling techniques.

I cringe when I hear “rule-based” segments are sitting on top of so-called state-of-the-art campaign engines. This is year 2018 A.D. It’s the age of abundant data with an ample number of tools and options to harness their true powers. And marketers are still making up rules now? It’s time for marketers to embrace modeling.

I wonder what most of the rules marketers are using are made of. Recency? Certainly, but how recent is recent enough?

Frequency? Sure, why not? The more the merrier, right? But in what timeframe? Are you counting transactions, orders or items? Or just some “events”?

Monetary? Hmm, that’s tricky. Are we using an individual-level lifetime total amount, value of the last transaction, average spending per transaction, average spending amount per year, or what? Don’t tell me you don’t even have individual-level summary data. No customer is just a reflection of her last transaction.

Actually, if a company is using some RFM (Recency, Frequency, Monetary Value) data for targeting, that is not so bad. At least it’s taking a look at what actually happened in terms of monetary transactions, not just clicks and page views, along with basic demographic data.

I have been talking about “employing all available data” for targeting and customer insights for some time now. So allow me to pick a different bone today. Let’s forget the data part, and talk about the methodology. When machines can build models super-fast, aversion to modeling only limits the users. After all, I am not asking any marketers to get a degree in statistics. I am just asking them to consider modeling techniques, as this data industry has moved forward from the days when some basic RFM rule sets used to get a passing grade.

Let’s look at the specific reasons why marketers should consider modeling techniques more seriously and ditch rule-based segmentation.

Reason No. 1: Variable Selection

We are surrounded by data, as every move that anyone makes is digitized now. When you describe a buyer, you may need to evaluate hundreds, if not thousands, of data points. Even if you are just using simple set of demographic data without any behavioral data, we are talking about over 100 variables to consider out of the gate.

Let’s say you want to build a rule to find a good segment for the sale of luxury cruises. How would you pick the most predictable variable for that one purpose? Income and age? That is not a bad start, but that is like using just two colors out of a crayon box containing 80 colors.

Case in point: Do you really believe that the main difference between luxury cruisers and luxury car buyers is “income”? Guess what, those buyers are all rich. You must dig much deeper than that.

Marketers often choose variables that they can easily understand and visualize. Unfortunately, the goal of the targeting exercise should be effectiveness of targeting, not easy comprehension by the marketer.

We often find obscure variables in models. They may “seem” obscure, as a human being would never have instinctively picked them. But mathematics doesn’t care for our opinions. In modeling, variables are picked for their predictive power, nothing else. The bonus is that this is exactly how new patterns are discovered.

We hear tidbits such as “People who tend to watch more romantic comedies are more likely to rent cars over the weekend,” “Aggressive investors are less likely to visit family restaurants” or “High-value customers for a certain teenage apparel company are more likely to be seasonal buyers with high item counts per customer, but relatively lower transaction counts.”

These are the contributing factors found through vigorous mathematical exercises, not someone’s imagination or intuition. But they should always make sense in the end (unless of course, there were errors). Picking the right predictor is indeed the most important step in modeling.

Reason No. 2: Weight Factor

Let’s say that by chance, a user stumbled upon a set of useful predictors of certain customer behavior. Let’s go back to the last example of the teenage apparel company’s high-value customer model. In that one sentence, I listed: seasonality (expressed in number of transactions by month, regardless of year), number of item counts per customer (with time limits, such as past 36 months), and number of transactions per customer.

In real life, there would be a far greater number of variables that would pass the initial variable selection process. But for simplicity’s sake, let’s just review these three variables.

Now tell me, which one of these three variables is the most important predictor of this high-value customer model? (Please don’t say they are all equally important.) Model scores are made of selected variables multiplied by the weight of each, as not all predictors carry the same level of predictability. Some may even be “negatively” correlated to the ideal behavior that we are going after. In this example alone, we saw that the number of items was positively related to the high value, while the number of transactions are negatively related. When investigating further about this “strange” correlation, we found out that most of the high-value customers are trained by the marketer to wait for a big sale, and then buy lots of items in one transaction.

The main trouble with the “rule-based” segmentation or targeting exercise is that human beings put arbitrary weight (or importance) on each variable, even if “the right” variables were picked — mostly by chance — in the first place.

The modeling process reveals the actual balance among all important predictors, with the sole purpose of maximizing predictability. Conversely, I have never met a person who can “imagine” the dynamics of two or three variables, let alone 10 to 20 (the typical number of variables in models).

Forget about the recent emergence of machine learning; with or without human statisticians, modeling techniques have been beating rudimentary rules by end-users for decades. If solely left to humans, the No. 1 predictor of any human behavior would be the income of the target. But that is just a reflection of human perception and a one-dimensional way of looking at a complex composition of human behavior. You don’t believe you can explain the difference between a Lexus buyer and a Mercedes buyer with just income, do you?

Reason No. 3: Banding

Much of data are composed of numbers and figures. The rest of them are called categorical variables (i.e., data that cannot be added or subtracted, such as product category or channel description).

Let’s assume that income — not my first pick, as you can see — is found to be predictable for mid- to high-scale female accessory buyers. Surely, different ranges of income would behave differently in such models. If the income is too low, they won’t be able to afford such items. Too high, then the buyer may have moved on to even more expensive handbags. So, the middle ground may seem to be the ideal target. The trouble is that now you have to describe that middle group in terms of actual dollars. Exactly where does that ideal range begin and end? To make it even more complicated, what about regional biases in buying power? Can one set of banding explain the whole thing? We’ve gone way past any intuitive grouping.

Moving onto categorical variables, one of the most predictable variables in any B2B modeling is the SIC code. There are thousands of variations in any one field, and they are definitely not numbers (although they look like them). How would one go about putting them into ideal groups to describe the target (such as “loyal customers”)?

If you are selling expensive computer servers, one may put “Agricultural, Fishing and Mining” as a low priority group. Then, how about all those variations in huge groups, such as “Retail,” “Business Service” or “Manufacturing,” with hundreds of sub-categories? Let’s just say that I’ve never met a human being who went beyond the initial two-digit SIC code in their heads. Good luck creating an effective group with that one variable with rudimentary methods.

Grouping “values” that move together in terms of predictability is not simple. In fact, that is exactly why computers were invented. Don’t struggle with such jobs.

These are just a few reasons why we must rely on advanced modeling techniques to navigate through complex data. The benefits of modeling are plenty (refer to “Why Model?”). Compared to our gut feelings, statistical models are much more accurate and consistent. They also reveal previously unseen patterns in data. Because they are summarized answers to specific questions, users do not have to consider hundreds of factors, but just one model score at a time. In the current marketing environment, when things move at a light speed, who has time to consider hundreds of data points in real-time? Machine learning — leading to full automation — is just a natural extension of modeling.

Each model score is a summary of hundreds of contributing factors. “Responsiveness to email campaigns for a European cruise vacation” is a complex question to answer, especially when we all go through daily data overload. But if the answer is in the form of a simple score (say, one through 10), any user who understands “high is good, low is bad” can make a sound decision at the time of campaign execution.

Marketers already have ample amounts of data and advanced campaign tools. Running such machines with some man-made segmentation rules from the last century is a real shame. No one is asking marketers to become seasoned data scientists; they just need to be more open to advanced techniques. With firm commitments, we can always hire experts, or in the near future, machines that will do the mathematical jobs for us. But marketers must move out of old fashioned rule-based marketing first.

The Secret Sauce for B2B Loyalty Marketing

Who’s likely to be your valuable customer? What will their value be in next few years? How long will they continue to do business with you? Which ones are in vulnerable positions, and who’s likely to churn in next three months? Wouldn’t it be great if you could identify who’s vulnerable among your valuable customers “before” they actually stop doing business with you?

B2B loyalty
“business-agreement,” Creative Commons license. | Credit: Flickr by Kevin Johnston

Properly measuring customer loyalty is often a difficult task in multichannel B2B marketing environment. The first question is often, “Where should we start digging when there are many data silos?” Before embarking on a massive data consolidation project throughout the organization, we suggest defining the problem statements by breaking down what customer loyalty means to you first, as that exercise will narrow down the list of data assets to be dealt with.

Who’s likely to be your valuable customer? What will their value be in next few years? How long will they continue to do business with you? Which ones are in vulnerable positions, and who’s likely to churn in next three months? Wouldn’t it be great if you could identify who’s vulnerable among your valuable customers “before” they actually stop doing business with you?

Marketers often rely on surveys to measure loyalty. Net Promoter Score, for example, is a good way to measure customer loyalty for the brand. But if you want to be proactive about each customer, you will need to know the loyalty score for everyone in your base. And asking “everyone” is too cost-prohibitive and impractical. On top of that, the respondents may not be completely honest about their intentions; especially when it comes to monetary transactions.

That’s where modeling techniques come in. Without asking direct questions, what are the leading indicators of loyalty or churn? What specific behaviors lead to longevity of the relationship or complete attrition? In answering those questions, past behavior is often proven to be a better predictor of future behavior than survey data, as what people say they would do and what they actually do are indeed different.

Modeling is also beneficial, as it fills inevitable data gaps, as well. No matter how much data you may have collected, you will never know everything about everyone in your base. Models are tools that make the most of available data assets, summarizing complex datasets into forms of answers to questions. How loyal is the Company XYZ? The loyalty model score will express that in a numeric form, such as a score between one and 10 for every entity in question. That would be a lot simpler than setting up rules by digging through a long data dictionary.

Our team recently developed a loyalty model for a leading computing service company in the U.S. The purposes of the modeling exercise were two-fold:

  1. Find a group of customers who are likely to be loyal customers, and
  2. Find the “vulnerable” segment in the base.

This way, the client can treat “potentially” loyal customers even before they show all of the signs of loyalty. At the opposite end of the spectrum, the client can proactively contact vulnerable customers, if their present or future value (need a customer value model for that) is high. We would call that the “valuable-vulnerable” segment.

We could have built a separate churn model more properly, but that would have required long historical data in forms of time-series variables (processes for those can be time-consuming and costly). To get to the answer fast with minimal data that we had access to, we chose to build one loyalty model, making sure that the bottom scores could be used to measure vulnerability, while the top scores indicate loyalty.

What did we need to build this model? Again, to provide a “usable” answer in the shortest time, we only used the past three years of transaction history, along with some third-party firmographic data. We considered promotion and response-history data, technical support data, non-transactional engagement data and client-initiated activity data, but we pushed them out for future enhancement due to difficulties in data procurement.

To define what “loyal” means in a mathematical term for modeling, we considered multiple options, as that word can mean lots of different things. Depending on the purpose, it could mean high value, frequent buyer, tenured customers, or other measurements of loyalty and levels of engagement. Because we are starting with the basic transaction data, we examined many possible combinations of RFM data.

In doing so, we observed that many indicators of loyalty behave radically differently among different segments, defined by spending level in this instance, which is a clear sign that separate models are required. For other cases, such overarching segments, they can be defined based on region, product line or target groups, too.

So we divided the base into small, medium and large segments, based on annual spending level, then started examining other types of indicators of loyalty for target definition. If we had some survey data, we could have used them to define what “loyal” means. In this case, we mixed the combinations of recency and frequency factors, where each segment ended up with different target definitions. For the first round, we defined the loyal customers with the last transaction date within the past 12 months and total transaction counts within the top 10 to 15 percent range, where the governing idea was to have the target universes that are “not too big” or “not too small.” During this exercise, we concluded that the small segment of big spenders was deemed to be loyal, and we didn’t need a model to further discriminate.

Stephen Yu's B2B loyalty marketing chart
Credit: Stephen H. Yu

As expected, models built for small- and medium-level spenders were quite different, in terms of usage of data and weight assigned to each variable. For example, even for the same product category purchases, a recency variable (weeks since the last transaction within the category) showed up as a leading indicator for one model, while various bands of categorical spending levels were important factors for the other. Common variables, such as industry classification code (SIC code) also behaved very differently, validating our decision to build separate models for each spending level segment.

The Future of Retail Is in a Data Stream Near You

For all of the digital disruption, store closures and bankruptcies, the retail industry still has not had its transformative “moment.” While some might see a slow death through a thousand cuts, there’s really no need for such doom and gloom.

Magnificent Mile, Chicago, Nov. 13 | Credit: Chet Dalzell

For all of the digital disruption, store closures and bankruptcies, the retail industry still has not had its transformative “moment.” While some might see a slow death through a thousand cuts, there’s really no need for such doom and gloom. Unless you’re a retailer that can’t handle change.

For every piece of negative news out there, there’s a slew of innovation happening now. Even at the macro level, retailing is thriving!

First off, consumers are still shopping. Despite $1.2 trillion in student debt and $800 billion in credit card debt, the American consumer has resilience and fortitude. The National Retail Federation forecasts 3 percent-plus growth in retail spending this year.

But don’t dare measure retail health any more in year-over-year “comps” (comparable sales by store).

The metrics of success are undergoing a (sea) change — much because the customer journey is changing … rapidly. Smartphones, tablets, kiosks, websites, search… and stores — it’s all in the mix. And it’s not just Millennials who are all over the customer journey map.

To capitalize on today’s highly personalized paths to purchase, data is the currency that’s making a difference. In its most recent U.S. Retail Industry StatPack 2017 (download access), eMarketer reported:

“If I can start to take browsing history, social media history and tie that to your transaction history, I can start to do very specific segmentation … If you can master the data, you can really target customers with what they want and optimize your marketing,” said Michael Relich, COO, Crate & Barrel.

“There has been device proliferation in natural consumer shopping behavior,” said David Doctorow, head of global growth, eBay. “To serve customers well, we have to identify them no matter the device they’re interacting with us on.”

So much for CPMs and sales per square foot.

Look at these profound shifts in retail ad spending. Retail dominates U.S. digital ad spend among vertical categories, primarily because of e-commerce competition. Retail accounts for 21.9 percent of total U.S. digital ad spend, and will grow 15.8 percent this year to more than $18 billion. (Automotive and Financial Services — second- and third-ranked industries, respectively — each net around 12 percent of total U.S. digital ad spend, for comparison). Kantar reports that TV retail ad spend sits at $7 billion — digital’s nearest above-the-line neighbor.

Within digital advertising categories, tried-and-true display ads have nearly closed the gap with search — the former helped by mobile display and digital video display ads targeted to Millennials, eMarketer reports. Video is now 14.2 percent of total digital ad spend among retailers, and more than half of this spend is purchased directly with premium video sites. Meanwhile, programmatic ad spend is driven by retailer spending on social media: Facebook, Instagram and elsewhere, though brands reportedly are cautiously guarding where their ads appear — turning to “private, one-to-one setups to buy high-quality inventory.”

Search ads, as one might guess, prevail in retail’s mobile ad spend, as consumers conduct price comparisons and look for product reviews and recommendations while in-store.

And what of Amazon, the so-called retail killer? On the contrary, Amazon (and other marketplace platforms, such as Walmart) “storefronts” may prove a survival mechanism for many stores — though don’t expect that highly sought path-to-purchase data to find its way back to the retail brand. Data usually is not shared outside the “walled garden.” According to retail consultant Ryan Craver, speaking recently at Marketing Idea eXchange, half of Amazon’s U.S. product sales are now sold through third-party vendors (storefronts).

Yes, there’s more to today’s and tomorrow’s retail story than data-driven targeting and marketing … yes, physical stores need to be destinations, malls need to be destinations, retail may need to entertain, etc. But mass marketing is scaling to 1:1 personalization, anonymized or other, through data. Recognize that customer well, she will reward you. Sounds a lot like the 19th Century shopkeeper, virtually yours.

How to Reach Your Customers at Home or at Work

Have you ever wished you knew more about your customers’ consumption habits? I have just learned about a new service from the San Antonio-based database marketing company, Stirista, that offers a way to link an individual’s consumer record with your corresponding business record.

unity-1763669_640Have you ever wished you knew more about your customers’ consumption habits? I have just learned about a new service from the San Antonio-based database marketing company, Stirista, that offers a way to link an individual’s consumer record with your corresponding business record. With StiristaLINK, you can enhance your business contacts with a personal email address, social media handles, home address, phone number, demographics and personal interests, vastly enhancing your understanding of your business contacts.

On the flip side, consumer marketers can use this capability to broaden the profile of their targets by providing additional insight — where they work, their titles, schools attended, past employers, their LinkedIn URL and perhaps even their work email address.

StiristaLINK’s files are pretty sizable: 20 million B-to-B profiles are enhanced with consumer information, while 55 million consumer emails are linked to a B-to-B profile.

I’ve been thinking about some of the things a marketer could do with this new capability. The use cases seem endless:

  • Enrich business and consumer profiles for better targeting and segmentation.
  • More touchpoints: Target consumers during business hours. Reach business people at home.
  • Access hard-to-reach segments (e.g.: Say you’re selling graduate business education and want to find 20-somethings who still don’t have an MBA).
  • Expand your universe of display and social media advertising targets.
  • Offer business people consumer products based on their work status. For example: Insurance companies can offer health insurance to workers in companies that are known to have decided to issue vouchers to their employees.

I asked Stirista’s CEO Ajay Gupta about the technology behind the links. He explained that Stirista already had a massive database of B-to-B and consumer records. To create the linkage, they took two innovative approaches. First, they matched the Twitter handles in the business record to those in the consumer database. To validate the match, they conducted research into a sample of the linked records, and found a 97 percent accuracy rate.

Next, Stirista engineers developed a creative approach to inferring current employment among consumer records that happen to have unusual names, geo-coding them by home address and matching that to the same unusual name in a nearby company.

As an example, let’s look at the case of Ajay Gupta himself. While there are scores of Ajay Guptas in the New York region, there happens to be only one in San Antonio. So the engineers could reasonably conclude that the Ajay Gupta working for Stirista is the same as the consumer Ajay Gupta who lives in San Antonio.

Of course, the technique does not work for the John Smiths, but it did add another layer of names to the file.

An early adopter of StiristaLINK was Weight Watchers, which was looking to improve its marketing to HR directors. Despite high brand awareness, selling Weight Watchers group packages had become increasingly difficult. The service is offered as a free benefit to companies, and employees receive a discounted rate when they join through their firms. But with more competition for their attention, HR directors were less and less motivated to pick up the deal and offer it to their employees.

The breakthrough came with the application of StiristaLINK to identify HR professionals and senior managers in the target companies whose consumer profiles indicated a personal interest in fitness and health. Stirista used a menu of about 20 keywords — cycling, exercise and softball, among others — to identify likely prospects and email them with a message about the importance of weight as a part of employee health and productivity. And the response rate boomed.

For B-to-B marketers, the most immediate benefit of this capability is reaching larger custom audiences. Most Twitter, Facebook and Google AdWords custom ad selection is based on the personal email address that was collected on sign up. By adding consumer data to your audience build, you’ll improve your reach dramatically.

There’s seemingly no end to the new data-driven marketing innovations these days.

A version of this article appeared in Biznology, the digital marketing blog.

Personalization Framework

In the age of constant bombardment with marketing messages, staying relevant to prospects and customers is not just good practice in the manual; it is a matter of survival.

personalizationIn the age of constant bombardment with marketing messages, staying relevant to prospects and customers is not just good practice in the manual; it is a matter of survival.

Recipients of marketing messages are more immune to generic offers than ever, and a relentless series of emails and we-will-follow-you-to-the-end-of-your-journey attitude literally trained them to ignore anything that even resembles commercial messages.

You want to stand out in this world of omnichannel marketing? Try to stand out by making it about “them,” not about “you.”


Personalization is not just another buzzword that came after the Big Data hype. It actually is something that marketers must care about.

According to Gartner Research, “By 2018, organizations that have fully invested in all types of online personalization will outsell companies that have not by more than 30 percent.”

I am not sure how they boldly put such a numeric prediction out. But in this case, I honestly think that the gap could end up being even larger, because the winners in this zero-sum game are moving at light-speed, while others still stubbornly carry that “If you keep reaching out to them, they will respond” attitude.

Being Clueless

I’ve actually met marketers who asked me how many more emails they should send out each week to compensate for an increasing number of non-responders.

They actually asked me if they can poke their customer base even more frequently. (They were sending uniform messages to everyone more than six times a week.) That means they had been diligently training the customers to ignore their emails.

I bluntly told them they just can’t mail their way out of that trouble. They should think about contacting their targets less frequently, and staying relevant as much as possible.

Do Unto Others

It is not difficult to sell the concept of personalization to marketers. They, too, are recipients of irrelevant marketing messages, and I bet that they mercilessly purge them out of their personal inboxes on a daily basis.

Surely, there are enough conference tracks, webinars, whitepapers and articles about this subject. But how are they supposed go about it? Do we even agree what that word means? (Refer to “What Does Personalization Mean to You?”)

Based on all of the client meetings that I’ve been to, the answer unfortunately is a hard “no.” And that conclusion was not solely drawn from some rudimentary practices being conducted by many marketers in the name of personalization, either. Because of available data and in different stages of customer relationship development, we do need to differentiate various types of activities under that all-inclusive personalization banner.

We Can Get There From Here

There are many personalization frameworks out there, listing various endeavors, such as collaborative filtering (as in “if you bought that item, you must be interested in these products as well”). Then there’s customer segmentation, and personas development based on predictive modeling techniques, usually in that sequence. If you add technical elements in terms of ability to show different things to different people, multiplied by content generation and content management pieces, things get complicated quite fast.

In any case, I do not agree with such sequential framework, as that is like saying the patient cannot be admitted to the operating room unless the doctor’s exhausted all of the simpler forms of treatments. Needless to say, some patients need surgery right away.

Likewise, when it comes to maximizing the value of data assets for personalization, marketers should not avoid predictive modeling by habit, just because it sounds complicated. That shouldn’t be the way in this age. If you want to be sophisticated about personalization, you’ve got to get serious about analytics without resorting back to simper, often ready-made, options. Unless of course, you as a consumer think that seeing offers for similar (or the same) products that you’ve just purchased for next couple of months is an acceptable form of personalization. (I don’t.)

Nuts and Bolts

Then, what should be the not-so-sequential data framework for personalization? Allow me to introduce one based on activity type and data availability, as no marketer can be free from data scarcity issues at different stages of customer relationship development.