I have made this longer than usual because I have not had time to make it shorter.
—Blaise Pascal
I was listening to a podcast the other day. The speaker had developed a theory that had really pushed his thinking. It was right on the edge of his ability to comprehend. I gave myself a self-satisfied pat on the back; I could understand it no problem.
But then I thought about it some more. I could understand the idea fully formed, but only because someone else had done the difficult formulating work.
Could I have come up with that idea on my own? No way.
The Paradox of Brilliance
It seems that the amount of intelligence needed to understand an idea is inversely proportional to its brilliance. My first conjecture: great ideas are grokkable, marginal ideas are not.
Let’s take an example. For millennia, bankers and merchants and mathematicians tried to solve simple equations without the number zero (think roman numerals). Progress in math was stymied, and commerce and finance were a tad more difficult than they needed to be. This condition continued for centuries. People simply could not conceive of counting nothing.
Eventually, some bright minds in India popularized the idea of zero and the diminutive digit made its way through the Middle East and finally the rest of the world.
We now teach the concept to three-year-olds. And they’re unimpressed.
We take great ideas for granted because they seem so obvious. The opposite is also true. The more half-baked the idea, the more likely we are to think that it’s us who are stupid - especially when it’s outside of our domain of expertise.
This has huge implications for data visualization. A clear visual is often maddeningly hard to create. Consider the two images below, each based on the same dataset.
The complexity of the first makes it look impressive. The second graph, not so much. But when you try to interpret the results, the second one wins hands down.
The same thing applies in data science and analytics. Suppose I can use a gradient boost model to predict the likelihood of an undrafted teenager becoming an NHL superstar. Or I can use a pivot table to get the same answer, but with less precision. The first seems impressive; yet the second is almost certainly more effective.
Pruning your Products
There’s an excellent quote from Albert Einstein: “Everything should be made as simple as possible, but not simpler.”
Notice that he doesn’t say: “Everything should be made as complex as necessary, but no more complex.” He recognizes that the creative process almost always overshoots. We arrive at a place of complexity and then work back to simplify. This is how ideas form - the tree grows, and then it is pruned. And then it bears fruit.
With the focus on prototypes and minimum viable products it’s easy to imagine you can start small and then stop at just the right place. But really, how often do we get it right? I suspect we overshoot far more than we know. The temptation to add features and increase precision is powerful. Yet we never have a project phase associated with simplifying. It never even crosses our minds.
My second conjecture: great ideas need pruning
If you reverse your perspective, you’ll find this process comes more naturally. Stop thinking like a modeler or designer and start thinking like a decision-maker. It’s not about maximizing simplicity, it’s about understanding the amount of evidence needed to decide with confidence.
What’s the threshold of information that flips the decision? Or put more simply, what’s the minimum necessary evidence?
This should be your guiding star. Minimum Necessary Evidence. MNE. Em, En, Ee. And to help us remember, we’ll use the sea aneMoNE as a MNEmonic device.
Take a close look at this creature. Notice that it has just enough tentacles to catch the fish it needs.
How do I know? Easy. If it had too many, then it would be obese.
(In fact, part of the anemone life cycle includes a phase where some of its weaker or poorly-positioned tentacles fall off*. It’s this self-pruning process that makes it even more effective.)
This is all well and good, you say, but how do we put it into practice?
Talk to actual users early and often. Show them mockups, prototypes, and toy models. But don’t just ask them if they like it. Have them explain their decision process. Strive to understand how they process information and what elements are the most critical. Stay focused on these and be ruthless with everything else. If they don’t need an answer to three decimal places, don’t you dare present it that way. Partner with them and co-create.
Set aside time for pruning. When you think you’re done, take a hard look at all the elements of your design or your model or your slide deck or your blog post and begin to chop away. Let the different features compete for your attention, and only the strong survive. What remains is the beautiful essence.
At the very least, keep the anemone in mind. Over time, you will begin to gain a reputation for clear thinking. Your ideas will seem to carry more weight and your products will get used more often.
So remember the anemone. Hear its tentacles whispering in the current: “less is more…”
*This has yet to be observed, but I believe in my heart that it must be true.