Owl speaks lion

Communicating analytical results to senior decision-makers.

By Jeff Kline

Jeff Kline

Jeff Kline

The admiral nodded in agreement! Since he had no background in operations analysis, I was pleasantly surprised he was following my explanation of integer linear programming used in the newly developed theater ballistic missile defense decision aid I was presenting. Maybe I will explain the history of the simplex algorithm next!

No, wait. Was that a nodding of understanding or was he falling asleep? Rats, his eyes were drooping.

Then he perked up enough to say, “Excuse me, but let me ask this. Is the math right?”

“Well, yes sir” I responded, “but, like all models, its efficacy is dependent on the quality of the input data.”

“Sure, sure,” he said. “Will it give our planning staff a quick ‘what-if’ tool to rapidly explore different courses of action and their risks?”

“Yes sir,” I responded.

“Fine,” he answered. “Let’s test it out in the next major theater ballistic war game.” As he was dismissing me he turned to his assistant and said, “Now, what is my next meeting?” I guess he did not want to learn about the simplex algorithm.

Almost all senior analysts have similar stories on how they learned not to brief a senior executive. They will tell you organizational lions respond to overly detailed briefings in various ways including confusion, boredom, anxiety, anger, impatience or any combination thereof. Not only is the analytical work’s value diminished, but the analyst’s future contributions may be questioned in the eyes of the executive. The art of communicating analytical work to senior executives is an integral part of any analytical endeavor, and it is best learned through practice. In the next few paragraphs we discuss observations and lessons learned from various senior analysts, with an eye toward accelerating a junior analyst’s maturity in the “translation art” of owl to lion.

The first step for “owls” is to know your “lion.” Image of Owl © Vaclav Volrab | 123rf.com.   Image of Lion © Michal Bednarek | 123rf.com

The first step for “owls” is to know your “lion.” Image of Owl © Vaclav Volrab | 123rf.com. Image of Lion © Michal Bednarek | 123rf.com

Know Your Lion

It is the analyst’s or owl’s responsibility to ensure he or she communicates the relevance of analytical work as concisely as possible in the executive’s, or lion’s, view. A demonstration of analytical brilliance is not appreciated, but a work’s usefulness to reduce uncertainty in decisions or streamlining processes certainly is.

The first step is to know your lion. Almost all are very busy and on the go. Brevity is essential. Some executives already value analytical input and are comfortable with data-driven decision-making. Some are owls themselves, which means that transparency in the analytical work’s limitations are expected and valued. The executive’s time and experience will influence how best to send your message for actionable receipt.

Next, understand the executive’s view of the problem or issue the analysis addresses. Is it a production line decision, a marketing opportunity, an acquisition question or a political decision? The category of problem influences the contribution analysis makes to that decision. For example, optimization can directly influence delivery truck schedules to minimize delivery time. A truck scheduler will use it to save driver time or make more deliveries. However, in the political realm there may be conflicting interests, opposing objectives and hence different rationalities, and multiple organizations involved. In politics, sometimes the best course of action may be what can be implemented, instead what is optimum according to a set of metrics. In this domain, analysis may be used to assess opportunity costs for alternatives that show promise using technical rationales but are not politically implementable due to conflicting interests.

How best to present your analysis is influenced by the executive’s view of which domain the problem occupies and the freedom to select a course to follow. That said, a global constant to show relevance to a profit-oriented lion is demonstrating whether a decision will make money, or save it. For policy- and service-oriented lions, showing efficiencies or money savings will capture interest.

Be Concise, Clear and to the Point

Then there is the speaking part. Albert Einstein allegedly said, “If you can’t explain it simply, you don’t understand it well enough.” Using the same vocabulary as the executive is essential. I have learned that reading business books written by analysts like Douglas Hubbard and Sam Savage provides a way to communicate analytical concepts to a larger audience. Savage’s example of using the term “uncertain number” instead of “random variable” is illustrative.

Counsel written specifically for giving analytical presentations is also useful. Jerry Brown in “How to Write About Operations Research” (PHALANX, 2004, Vol. 37, No. 3) poses five questions to answer when communicating analytical results:

  1. What is the problem?
  2. Why is the problem important?
  3. How will the problem be solved without your help?
  4. What are you doing to solve the problem?
  5. How will we know when you have succeeded?

E.S. Levine in his August 2013 article, “Good Guidance for Presentation of Data,” in Analytics magazine provides solid advice. He says strive for clarity (simple, direct, efficient and effective), transparency (clearly defined method and assumptions), integrity (defensible and sound conclusions) and humility (removing the analyst from the message; another way of saying don’t show how smart you are, but how relevant your message is).

Senior Analysts (Wise Old Owls)

Have a five-minute version and
a two-minute version
Clearly answer: What? So What? What now?
Limit your presentation slides:
Save brilliance for back-up slides
Admit ignorance when you don’t know
Be prepared to talk without slides
Send your presentation ahead
Practice and murder board before briefing
(with a parliament of owls)

Senior Executives (Old Lions)

If I have only five minutes, so do you
Don’t put the executive back in math class
It is not necessary to share with me everything you have learned in reaching
this point in your life
Do not raise an issue unless you also provide
Give me the main points early
If you can’t answer the question, say so and get back with me. Anything else is a waste of time.
More pictures, fewer words

One straightforward way to create your presentation is to use the order of the analytical process you used in your effort. For example, make the first draft cover problem formulation, objectives, key assumptions, values and criteria, boundaries and constraints, generation of alternatives, analytical method or model, analytical results comparing alternatives, sensitivity analysis, and recommendations or conclusions. If and only if your lion was once an owl, this 50-slide presentation may be sufficient. Otherwise, move the recommendations or conclusions slide up front and pare the presentation down to 10 or fewer slides that give direct evidence supporting your recommendation or conclusion. Save all the rest as back-up to respond to potential questions.

That makes the wording of your first slide or leading point critical. It must present the problem and recommendation as concisely as possible. Think of this as a long bumper sticker. For example, to an Australian fleet commander assigning forces across the northern maritime border, the first slide showing simulation results reads: “With three to five illegal immigrant boats attempting passage across the Timor Sea every day, two patrol aircraft and five ships are needed on station to maintain a 95 percent chance of intercepting all of them. The issue driving required ship numbers is the time to intercept and inspect the boats before others make landfall.”

In two sentences we cover the problem and threat level, the simulation results for a certain risk level and the most sensitive parameters driving the results. The remaining 10 or fewer slides cover assumptions, concept of operations, model, sensitivity and areas of uncertainty that support this conclusion. Note I’ve suggested 10 or fewer slides twice now – well, now, three times! A general rule is the more senior the executive, the fewer the slides. Recall the importance of brevity.

The use of the “headline” or concise summary up front can carry danger, too. Our colleague, Jerry Brown, once briefed a U.S. admiral with these words: “Following this optimized schedule is like creating an extra aircraft carrier.” It was Jerry’s misfortune that this admiral was responsible for justifying how many aircraft carriers we should have in the fleet. Result: Lion eats owl. Lesson: Know your lion.

Perhaps Jerry should have focused on the ability to complete more missions, faster.

Having back-up material is important to keep the owl off the lion’s menu. If senior executives are uncomfortable with the study’s results, they will question the assumptions, scenario or situation, and will want to know the quality of the data. Ready-to-go back-up slides and material help you answer these concerns. Executives rarely question the mathematics or model – they trust you on that score (unless your lion was once an owl). They may ask the question, “Has the model been verified?” But, if the lion is policy-oriented, what he or she really wants to know is, “Has it been used before by another decision-maker?”

Reassurance is sought that there is not additional risk in the decision by relying on an untested model. Particularly valued is if the past use had a positive result for some other decision-maker. The other two ways a model is affirmed in an executive’s mind are: 1) if its results are within their experience (seen before in the real world), and/or 2) if the results are in line with their intuitive reasoning (makes sense). If the answer to any of these questions is no, you will need to be able to explain in detail why not.

Use Headlines and Graphs

Now for some advice on presentation slides: Use as few as words as possible, and feature as many graphs and charts as needed. Title each chart like a newspaper headline that explains the chart’s or graph’s message. For example, a chart comparing a delivery company’s current truck loading method with a new one should read, “New packing method saves drivers 30 minutes daily,” not “Results.” Be prepared to present a crystal-clear legend voice-over for each graph or chart. For illustrations of estimated returns for a range of courses of action, you cannot know in advance where a lion will decide to choose, so be prepared for any alternative. If we get this right, our executive presentation may consist of only one or two slides – the conclusions and the analytical graph that shows how we came to those conclusions. POTUS’ (president of the United States) analytical presentations are frequently just that.

Wise Owls and Old Lions

To reap additional lessons in communicating analytical results, I conducted an informal poll of senior analysts and senior executives on their thoughts. I asked them to write down what single phrase they would put in a fortune cookie to provide advice to junior analysts on communicating analytical results to senior executives. They provided the following wisdom to share with you. Note some recurring themes.

In industries where technical rationale prevails, such as engineering, technology, manufacturing, communications and data services, analytics’ value is naturally understood. In these areas we need only ensure we deliver our message effectively. In areas where other rationales dominate, such as public policy, national security, politics and international relations analytics, value must be learned by senior executives, which makes demonstrating the relevance of our work that much more important. The better we communicate the analytical results to those who use it, the more analytics and its practitioners will be appreciated. The more we are appreciated, the more influence we can have on making better decisions for industry and society.

Jeff Kline (captain, U.S. Navy, ret.) is a professor of practice in the Operations Research Department at the Naval Postgraduate School in Monterey, Calif., where he teaches joint campaign analysis and executive risk assessment and coordinates maritime security education programs. Kline supports applied analytical research in maritime operations and security, theater ballistic missile defense and future force composition studies. He received the 2011 Prize for the Teaching of OR/MS Practice from INFORMS.