Smart business decisions come from simplifying problems. Bars can help!

Solving complex business problems “ain’t always rocket surgery,” to cite a colloquialism I just discovered. It often boils down to a little bar and a big bar.

As an outsider, I have the luxury of both working with data scientists and in gentle opposition. The first part is easy. Data scientists work with my clients’ data daily, and have produced elaborate models to help make things more understandable.

But I also work in gentle opposition. To quote best-selling author and distinguished professor of economics (and fellow “outsider”) Steven Levitt, it’s all in the incentives. He’s excerpted immediately below from one of his People I (Mostly) Admire podcast episodes. If you’d prefer not to listen, the transcript immediately follows.

Little Bar / Big Bar

When I work with firms that have data scientists, what I find almost uniformly is that they operate in an incredibly complex space. They’re very concerned with technicalities, with techniques, with things being hard. And I think the answers are often very simple. So I try to always do simple things, and try to relate them in very basic ways. Like, my favorite kind of graphs are Big Bar / Little Bars graphs.

They’re graphs that have one really little bar … and one really big bar, and those are the kind of graphs that I show to CEOs if I’m trying to convince them of something. And the CEOs say to me, ‘Wow, that makes sense to me. I don’t understand how you take the same data that my data science team has and I never understand anything they’re saying.’

So, you might say ‘The answer is to do things really simply,’ but I think it’s more complicated when you think of incentives. Because much of the power that comes to data scientists in firms and organizations is because they are completely and totally inscrutable. And the other people have no idea what they’re doing. And by having a set of skills that no one else has, you can wield power because no one understands why you’re doing it. You have a very special talent. And so, I have the luxury of being an outsider.

Steve Levitt, “I’m Not as Childlike as I’d Like to Be” | People I Mostly Admire Bonus Episode

Wow! Can I relate. I hear this frustration when I talk to the leadership within my clients. Which is why I have espoused simplicity in everything I report, and have for years.

Of Oracles and Job Security

At the end of a talk I gave in 2018 about experience optimization, I urged the audience to think of themselves as modern day oracles. But instead of the literal magic tricks that the Greek vestal virgins employed to keep their jobs, we only have our science to keep us regarded as useful and legitimate.

Scrupulous science.

Many data scientists stop there. They generate reports using the latest and greatest methods, but when the output is shared, business leaders too often have no clue what the data is telling them.

That quote I shared with you from Steve Levitt was in response to a question about the future of data science as a profession. If you are a data scientist, consider this advice: Spend more time listening to your employer about the problems dogging them. And if you’re one of those employers, you cannot go wrong by hiring, when appropriate, someone outside of the data scientist industrial complex to whip up the bars!

Why isn’t the Disaffection Index (DI) more popular with campaign marketers?

Many years ago, when self-proclaimed Email Diva and my own personal career doppelganger Melinda Krueger conceived of the Disaffection Index, my only qualms were with its name and acronym. She wrote about this campaign KPI in one of her MediaPost pieces. If I remember correctly, she even sent me a pre-read. I told her I loved the metric but predicted it would become more commonplace with campaign marketers if she gave it a TLA (three letter acronym). I was only half joking.

You be the judge. First, what is it?

Rather than unsubscribe / delivered, the DI is calculated by dividing unsubscribes by the response rate …

Calculated this way, the DI tells you how many people either a.) clicked on your e-mail for the sole purpose of getting off your list or b.) were so dissatisfied … they chose to unsubscribe.

Excerpt from “The New Unsubscribe Rate”

Today every subscriber and brand loyalist is worth too much to squander. So why isn’t this KPI on every campaign manager’s dashboard?

Last Click Index (LCI) to the rescue!

I humbly suggested at the time that LCI was a better term for two reasons. First, it adds drama. When someone unsubscribes, you’d believe you weren’t getting any more clicks from them. B*tch, bye!

And secondly, I had then and still have zero affection for the word disaffection. But it’s her baby, and a rose is a rose by any name. (Clever guy, that Shakespeare).

So if you’re a campaign marketer, start using it, regardless of its inferior name. You’ll thank me. And Melinda.

Melinda is currently an Associate Principal for the Salesforce Marketing Cloud, and I predict she will laugh heartily when she reads this. I do hope so. I miss talking shop with her!

Make your charts more like maps

Pecha Kucha is Japanese for “chit-chat.” I’ve talked about the presentation style on my personal blog.

In a Pecha Kucha, the speaker is restricted to just 20 slides, and only 20 seconds per slide. It’s a 6 minute, 40 second burst of information. As for the one embedded below, here’s some context: Think of the various maps in your life. Mine come to mind easily. I see one every time I go home.

I have a map of the world on the wall of my apartment, with pins showing places I’ve visited and others for where I’d like to visit next, once this Covid nightmare ends. As you can imagine, it drives decisions and actions.

Another map is one we all have. It’s the small map in our phones, powered by GPS. That map drives decisions and actions every day. But if you’re responsible to helping your bosses or clients understand data, what about those data dashboards or charts?

All too often our charts do not drive action

Why is that? In this Pecha Kucha, I explore that question, and show you how you can drive more consensus and action with your data visualizations by choosing and customizing charts to be more map-like. I pull two examples from my decades of experience as a marketing analytics consultant.

The data is all made up, but reflects real life business challenges, and how making your charts more like maps can help to bring your bosses or clients to decisions. And isn’t that what data-driven decision making is all about?


How to lie with data visualizations, Part 2

This is a follow-up to my post from last month. In that Part 1, I wrote about how shifting the ranges for heat maps and starting bar charts at numbers greater than zero can deceive. In this installment, I want to introduce you to the greatest source of lies in data visualization: Cognitive biases of their authors.

I started my earlier post with an example torn from the headlines. I will again below.

If you follow the latest news in the U.S. about Covid, you may have seen this graphic shared over the weekend, included as evidence for emergency approval of convalescent plasma as a treatment for the infection. The evidence was provided by FDA Commissioner Stephen Hahn. His chief graphic is shown in the above image.

The graphic  appears to show a 37% reduction in deaths for Covid patients receiving infused plasma from donors who have survived the infection. Famously, Tom Hanks and his wife donated their plasma to help the research being conducted world-wide.

Rounding down to 35% in a press conference, “A 35 percent improvement in survival is a pretty substantial clinical benefit.”

It is, but that isn’t what the data shows.

After an outcry from what seemed like just about everyone in Covid research whose name was not Stephen Hahn, he apologized. The drop in mortality was from roughly 11% to 7%, which is a drop of somewhere around 4%, not 37%. It was also from research on a small sample size, weirdly without a control group, among many other concerning factors.

In an administration that is arguably more politicized than any in recent history, the impulse is to say this U.S. government official lied. Strictly speaking, he most certainly did — in both his words and his supporting data visualization. But the motivation could be less nefarious — or at least, more common.

Cognitive Biases Can Make Us Lie to Ourselves

Our brains are not the perfect instruments we delude ourselves into believing. Consider the Free Brian Williams episode of Malcolm Gladwell’s podcast Revisionist History about the fallibility of human memory. Or more relevant to this instance of (self-)deception, consider this roughly 10-minute excerpt of a talk I co-presented in 2018 at Adobe Summit in Las Vegas. In that talk I explain why clear, data-focused hypotheses are an important protection a data scientist has against cherry-picking or distorting data to pronounce a test a winner.

In that talk I explained that we as marketers exploit consumer cognitive biases every day, including through the use of A/B testing tools like Adobe Target. That’s a good thing, from a marketing perspective. I also maintain that there is strong evidence all of us, regardless of our training and discipline, probably have biases that literally pre-date us as humans. These biases can cause us to lie to ourselves, and through those self-deceptions, lie to our clients.

Everyone wants to report successes. But science is hard, and replication of test results is a persistent problem. In addition to clear hypotheses, we can also conduct an exercise called preregistration to save us from ourselves.

The Killer of Truth Is Calling From INSIDE THE HOUSE!

In my first installment I talked about how easy it is to distort numbers by using bad or lazy visualizations. In this second and last installment, I want to remind you that a far more pernicious murderer of the truth is your own best intentions. I’m willing to give Mr. Hahn the benefit of the doubt that he wasn’t so much lying as practicing wishful thinking, which blinded him from many clear contraindicators of his conclusion — including the fact that randomized trials of convalescent plasma at scale is dead simple to conduct. They’re indeed being done elsewhere in the U.S., such as at UCLA — where Tom Hanks donated his plasma — and in other countries around the world. Yet no country has rung the bell on such a decisive victory over this horrible virus.

Let his humiliation be a lesson to us all.

The Domestication of Artificial Intelligence

A version of this post originally appeared in my personal blog.

I’ve thought and read a lot about artificial intelligence (AI). Particularly, its potential threat to us, its human creators. I’m not much for doomsday theories, but I admit I was inclined to fear the worst. To put things at their most melodramatic, I worried we might be unwittingly creating our own eventual slave masters. But after further reading and thinking, I’ve reconsidered. Yes. A.I. will be everywhere in our future. But not as sinister job-killers and overlords. No, they will be extensions of us in a way I can only compare with that most beloved of domesticated creatures: The dog. For you to follow my logic, you’ll need to remember two facts:

  1. Our advancement as a species from hunter-gatherers to complex civilizations would not be possible without domesticated plants and animals
  2. Our collective fear of technology is often wildly unfounded

Bear with me, but you’ll also probably need to recall these definitions:

  • Domestication: Taking existing plants or animals and breeding them to serve us. Two examples are the selection of the most helpful plants and turning them into crops. Michael Pollan’s early book, The Botany of Desire: A Plant’s-Eye View of the World, will bring you a long way to seeing this process in action. As for animals, you may think of dogs as being mere pets, but early in our evolution as humans we bred the wolf to help us hunt for meat, and to protect us from predators. Before domestication, we pre-humans hunted in packs, and so did the wolves … never the twain shall meet. After this domestication, we ensured the more docile canines a better life, under the protection of our species and its burgeoning technologies (see definition below), and they delivered the goods for us by helping us thrive in hostile conditions. It was a symbiosis that turned our two packs into a single unit. No wonder the domesticated dog adores us so, and that we consider them man(kind)’s best friend.
  • Technology: Did you know the pencil was once considered technology? So was the alphabet. You may think of them merely as tools, but technology is any tool that is new. And our attitudes toward anything new always starts with fear. Douglas Adams put it this way: “I’ve come up with a set of rules that describe our reactions to technologies: 1.) Anything that is in the world when you’re born is normal and ordinary and is just a natural part of the way the world works. 2.) Anything that’s invented between when you’re fifteen and thirty-five is new and exciting and revolutionary and you can probably get a career in it. 3.) Anything invented after you’re thirty-five is against the natural order of things.” Fear of technology not surprisingly spawned the first science fiction: Mary Shelley’s Frankenstein; or, The Modern Prometheus, a literal fever dream about a scientist’s hubris and the destruction it wrought upon himself and the world. This fear has a name: Moral panic. And it has created some pretty far-fetched urban myths.

In a Wall Street Journal piece, Women And Children First: Technology And Moral Panic, Genevieve Bell listed a few of these vintage myths. The first is about the advent of the electric light: “If you electrify homes you will make women and children … vulnerable. Predators will be able to tell if they are home because the light will be on, and you will be able to see them. So electricity is going to make women vulnerable … and children will be visible too and it will be predators, who seem to be lurking everywhere, who will attack.” And consider this even bigger hoot: “There was some wonderful stuff about [railway trains] too in the U.S., that women’s bodies were not designed to go at 50 miles an hour. Our uteruses would fly out of our bodies as they were accelerated to that speed.” Sounds messy. I don’t have to tell you about our modern moral panic surrounding A.I. Except there is a bit of reverse sexism going on, because this time it is male workers who are more the victims. Their work — whether purely intellectual or journeyman labor — will be eliminated. We’ll all be out on the street, presumably to be mowed down by self-driving cars and trucks.

The Chicken Littles had me for a while

So what changed? In the same week I read two thought-provoking articles. One was in The New Yorker, The Mind-Expanding Ideas of Andy Clark. Its subtitle says it all: The tools we use to help us think — from language to smartphones — may be part of thought itself. This long piece describes Clark’s attempt to better understand what consciousness is, and what are its boundaries. In other words, where do we as thinking humans end and the world we perceive begin? He comes to recognize that there is a reason we perceive the world based on our five senses. Our brains are built to keep us alive and able to reproduce. Nothing more. All the bonus tracks in our brain’s Greatest Hits playlist … Making art, considering the cosmos, perceiving a future and a past … these are all artifacts of a consciousness that moves our limbs through space.

To some people, perception — the transmitting of all the sensory noise from the world — seemed the natural boundary between world and mind. Clark had already questioned this boundary with his theory of the extended mind. Then, in the early aughts, he heard about a theory of perception that seemed to him to describe how the mind, even as conventionally understood, did not stay passively distant from the world but reached out into it. It was called predictive processing.

Predictive processing starts with our bodies. For instance, we don’t move our arm when it’s at rest. We imagine it moving — predict its movement — and when our arm gets the memo it responds. Or not. If we are paralyzed, or that arm is currently in the jaws of a bear, it sends the bad news back to our brains. And so it goes. In a similar way we project this feedback loop out into the world. But we are limited by our own sense of it. Domestication of canines was such a game-changer because we suddenly had assistants with different senses and perceptions. Together humans and dogs became a Dynamic Duo … A prehistoric Batman and Robin. But Robin always knew who was the alpha in this relationship. Right now there is another domestication taking place. It’s not of a plant or an animal, but of a complicated digital application. If that seems a stretch … If grouping these three together — plants, animals and applications — keep in mind that domesticating all of them means altering digital information.

All Life Is Digital

Plants and animals have DNA, or deoxyribonucleic acid. They are alive because they have genetic material. And guess what? It’s all digital. DNA encoding uses 4 bases: G,C,T, and A. These are four concrete values that are expressed in the complex combinations that make us both living, and able to pass along our “usness” to new generations. We’re definitely more complicated than the “currently” binary underpinnings of A.I. But as we’ve seen, A.I. is really showing us humans up in some important ways. They’re killing us humans at chess. And Jeopardy. So: Will A.I. become conscious and take us over? Clark would say consciousness is beyond A.I.’s reach, because as impressive as its abilities to move through the world and perceive it are, even dogs have more of an advantage in the consciousness department. He would be backed up by none less than Nobel Prize in Economics winner Daniel Kohneman, of Thinking, Fast and Slow fame. I got to hear him speak on this subject live, at a New Yorker TechFest, and I was impressed and relieved by how sanguine he was about the future of A.I. Here’s where I need to bring in the other article, a much briefer one, from The Economist. Robots Can Assemble IKEA Furniture sounds pretty ominous. It’s a modern trope that assembling IKEA furniture is an unmanning intellectual test. But the article spoke more about A.I.’s limitations than its looming existential threats. First, it took the robots comparatively long time to achieve the task at hand. In the companion piece to that article we read that …

Machines excel at the sorts of abstract, cognitive tasks that, to people, signify intelligence—complex board games, say, or differential calculus. But they struggle with physical jobs, such as navigating a cluttered room, which are so simple that they hardly seem to count as intelligence at all. The IKEAbots are a case in point. It took a pair of them, pre-programmed by humans, more than 20 minutes to assemble a chair that a person could knock together in a fraction of the time.

Their struggles brought me back to how our consciousness gradually materialized to our prehistoric ancestors. It arrived not in spite of our sensory experience of the world, but specifically because of it. If you doubt that just consider my natural and clear way just now of describing the arrival of consciousness: I said it materialized. You understood this as a metaphor associated with our perception of the material world. This word and others to describe concepts play on our ability to feel things. Need another example: This is called a goddamn web page. What’s a page? What’s a web? They’re both things we can touch and experience with our carefully evolved senses. And without these metaphors these paragraphs would not make sense. Yes, our ancestors needed the necessary but not sufficient help of things like cooking, which enabled us to take in enough calories to grow and maintain our complex neural network, and the domestication of animals and plants that led us to agriculture and an escape from the limitations of nomadic hunter-gatherer tribes (I strongly recommend Guns, Germs and Steel: The Fates of Human Societies for more on this), but … To gain consciousness, we also needed to feel things. And what do we call people who don’t feel feelings? Robots. “Soulless machines.” Without evolving to feel, should A.I. nonetheless take over the world, it’s unlikely they will be assembling their own IKEA chairs with alacrity. They’ll make us do it for them. Because our predictive processing makes this type of task annoying but manageable. We can even do it faster over time.

It’s All About The Feels

But worry not. Our enslavement won’t happen because — and I’m feeling pretty hubristic myself as I write this — we’re the feelers, the dreamers, the artists. Not A.I. Before we domesticated dogs, we were limited in where in the world we could roam, and the game we could hunt. After dogs, we progressed. We prospered. Dogs didn’t put us out of jobs, if you will, they took the jobs they were better at in our service. Inevitably, we found other ways to use our time, including becoming creatures who are closer to the humans we would recognize on the street today, or staring back in the mirror. We are domesticating A.I. Never forget that. And repeat after me: We have nothing to fear but moral panic itself.