“I’m just giving you a brain dump.”

Please don’t. Not to me, not to your colleagues, and especially, no matter how dire the circumstances, not to your manager.

Start with the prevalent but inaccurate distinction between data and information. Data are, supposedly, meaningless until processed into meaningful and useful information.

Not to nitpick or nuthin’ but “information” already had a definition before this one came along. It comes, appropriately enough, from information theory, which defines information as the stuff that reduces uncertainty.

As long as we’re being annoyingly pedantic, far from being worthless, data consist of indisputable facts: A datum is a measurement of some attribute of some identifiable thing, taking measurement in its broadest sense — if you observe and record the color of a piece of fruit, “orange” is a measurement.

So a fact can, in fact (sorry) reduce your uncertainty, as in the case where someone has asserted that something is impossible. If you observe and document it happening even once, you’ve reduced everyone’s uncertainty about whether the phenomenon in question is possible or not.

As long as we’re being metaphysical, let’s add one more layer: Meaning isn’t something information confers. Meaning is a property of knowledge — something a person develops, over time, by interpreting their experience, which is a combination of raw data, information, and logic, and, if we’re being honest with ourselves, no shortage of illogic as well.

(If, astonishingly, you’re interested, Scott Lee and I covered this topic in more depth in The Cognitive Enterprise.)

Back to brain dumps. You might think the problem is that the dumper is providing data, not information. Au contraire, mes amis. In my experience, brain dumps contain precious little data. They are, instead, a disorganized jumble that does include some information, interspersed with anecdotes, opinions of varying degrees of reliability (the brain-dumper would consider these to be knowledge), and ideas, which, as we’re being definitional, we might think of as hypotheses only without the supporting logic that makes good hypotheses worth testing.

And so, now that I’ve thoroughly buried the lede, the reason brain dumping is generally worse than useless is that it’s an exercise in reverse delegation.

Brain dumps happen when one person asks another person to figure something out and then explain it so they’ll both be smarter about the subject at hand.

But instead of making the delegator smarter, the brain-dumper has instead de-delegated the hard work of organizing these bits and pieces into a clear and coherent narrative.

It’s as if I were to assign you responsibility for baking a cake, and to satisfy the assignment, instead of returning with my just desserts, you were to dump a bunch of raw foodstuffs on my desk, some of which might be useful as cake ingredients and others not, along with 23 recipes for pies and cakes, plus commentary about how eating too much sugar causes cavities and adult-onset diabetes.

When receiving end a brain dump I often conclude the dumper has lost track of the explanation’s purpose. Instead of trying to make me smarter about a subject, the presenter is, instead, trying to show me how smart he or she is.

But it’s more likely I’ll reach the opposite conclusion, due to one of Einstein’s dicta: “If you can’t explain it simply, you don’t understand it well enough.”

Bad meta-message.

How can someone keep themselves from becoming a brain-dumper? Here’s one approach: Start by carefully choosing an entry point.

Imagine I’m supposed to explain something to you. Presumably I know quite a lot about the subject at hand or you wouldn’t ask. I know so much, in fact (this is, you understand, hypothetical) that I can’t explain anything I know about it until you understand everything I know about it.

And as you won’t be able to understand anything I have to say about it until you’ve heard everything I have to say about it, my only choice is to dump the contents of my brain onto your desk.

But if I choose a good entry point I’ll be starting my explanation with something about the subject you can understand immediately, like, “We have a problem. Here’s what it is, and why you should be concerned about it.”

Then comes the second-hardest part: Leaving out everything you know about the subject, except what helps explain what the problem is and why your listener should be concerned about it.

Leaving out any of my precious knowledge out hurts.

But that’s better than the pain I’d inflict by leaving it in.

Rank the most-reported aspects of COVID-19, in descending order of worst-explained-ness. Modeling is, if not at the top, close to it.

Which is a shame, because beyond improving our public policy discussions, better coverage would also help all of us who think in terms of business strategy and tactics think more deeply and, perhaps, more usefully about the role modeling might play in business planning.

For those interested in the public-health dimension, “COVID-19 Models: Can They Tell Us What We Want to Know?” Josh Michaud, Jennifer Kates, and Larry Levitt, KFF (Kaiser Family Foundation), Apr 16, 2020 provides a useful summary. It discusses three types of model that, translated to business planning terms, we might call actuarial, simulation, and multivariate-statistical.

Actuarial models divide a population into groups (cohorts) and move numbers of members of each cohort to other cohorts based on a defined set of rules. If you run an insurance company that needs to price risk (there’s no other kind), actuarial models are a useful alternative to throwing darts.

Imagine that instead you’re responsible for managing a business process of some kind. A common mistake process designers make is describing processes as collections of interconnected boxes.

It’s a mistake because most business processes consist of queues, not boxes. Take a six-step process, where each step takes an hour to execute. Add the steps and the cycle time should be six hours.

Measure cycle time and it’s more likely to be six days. That’s because each item tossed into a queue has to wait its turn before anyone starts tow work on it.

Think of these queues as actuarial cohorts and you stand a much better chance of accurately forecasting process cycle time and throughput — an outcome process managers presumably might find useful.

Truth in advertising: I don’t know if anyone has ever tried applying actuarial techniques to process analysis. But queue-to-queue vs box-to-box process analysis? It’s one of Lean’s most important contributions.

Simulation models are as the name implies. They define a collection of “agents” that behave like entities in the situation being simulated. The more accurately they describe agent behaviors, estimate the numbers of each type of agent, the probability distributions of different behaviors for each type, and the outcomes of these behaviors … including the outcomes of encounters among agents … the more accurate the model’s predictions.

For years, business strategists have talked about a company’s “business model.” These have mostly been narratives rather than true models. That is, they’ve been qualitative accounts of the buttons and levers business managers can push and pull to get the outcomes they want.

There’s no reason to think sophisticated modelers couldn’t develop equivalent simulation models to forecast the impact of different business strategies and tactics on, say, customer retention, mindshare, and walletshare.

If one of your modeling goals is understanding how something works, simulation is just the ticket.

The third type of model, multivariate-statistical, applies such techniques as multiple regression analysis, analysis of variance, and multidimensional scaling to large datasets to determine how strongly different hypothesized input factors correlate with the outputs that matter. For COVID-19, input factors are such well-known variables as adherence to social distancing, use of masks and gloves, and not pressuring a cohabiter to join you in your kale and beet salad diet. Outputs are correlations to rates of infection and strangulation.

In business, multivariate-statistical modeling is how most analytics gets done. It’s also more or less how neural-network-based machine learning works. It works better for interpolation than extrapolation, and depends on figuring out which way the arrow of causality points when an analysis discovers a correlation.

As with all programming, model value depends on testing, although model testing is more about consistency and calibration than defect detection. And COVID-19 models have brought the impact of data limitations on model outputs into sharp focus.

For clarity’s sake: Models are consistent when output metrics improve and get worse in step with reality. They’re calibrated when the output metrics match real-world measurements.

With COVID-19 testers have to balance clinical and statistical needs. Clinically, testing is how physicians determine which disease they’re treating, leading to the exact opposite of random sampling. With non-random samples, testing for consistency is possible, but calibration testing is, at best, contorted.

Lacking enough testing capacity to satisfy clinical demands, which for most of us must come first as an ethical necessity. Modelers are left to de-bias their non-random datasets — an inexact practice at best that limits their ability to calibrate models. That they yield different forecasts is unsurprising.

And guess what: Your own data scientists face a similar challenge: Their datasets are piles of business transactions that are, by their very nature, far from random.

Exercise suitable caution.