How to Use Numbers Properly

Podcast: Play in new window | Download

Can’t we all just get along? I’m not talking about ethnic groups or political parties—I’m talking about an even greater divide: the numbers people and the words people. The word tribe says that people decide on stories and emotions, so it doesn’t make sense to bore your listeners with numbers that they won’t understand anyway. The numbers tribe will tell you that the plural of anecdote is not data and if you can’t measure it, it doesn’t exist.

I say that you need to become comfortable in both languages. You need to be bilingual in your presentations, and make adjustments up or down on the balance between numbers and words depending on the topic or the audience or the purpose of your presentation.

The case for more numbers:

When you can measure what you are speaking about, and express it in numbers, you know something about it; but when you cannot express it in numbers, your knowledge is of a meager and unsatisfactory kind; it may be the beginning of knowledge, but you have scarcely in your thoughts advanced to the state of science.  Lord Kelvin

How numbers make you more persuasive:

1. There are two forms of evidence that carry greater weight than any others: eyewitness testimony and empirical data. But when you think about it, empirical data actually is a form of eyewitness testimony, because it could only have been gathered by someone or something being there to measure it. Using data in your argument proves that you or someone has actually witnessed a phenomenon, an furthermore has observed it closely enough to precisely describe it.
2. Numbers can improve your thinking. Philip Tetlock, who studies the track record of “experts” forecasters, has found that asking people to quantify their forecasts—by switching from words like probably to 75% chance, for example, forces them to think more carefully about the factors that may influence the actual event. It forces you to think in finer-grained detail about what you know and don’t know. By the same token, who is more believable?
3. Numbers can clarify. In other words, putting a number on something reduces the chances for error, uncertainty or miscommunication.
4. Numbers can provide cover to decision makers who plan to do something anyway, and they can make it easier for them to sell the idea to others.

The case for fewer numbers:

“Not everything that counts can be counted, and not everything that can be counted counts.” William Cameron

1. Using too many numbers can either bore or overwhelm your audience. They can’t be convinced if they’re not listening.
2. People don’t charge up a hill waving a spreadsheet. Numbers by themselves don’t carry meaning or value
3. Sometimes it’s too costly or time-consuming to get exactly the information you need, so you can run into analysis paralysis and lose opportunities
4. Numbers can give you a false sense of certainty. There’s something convincing about a precise number, but you might forget that it’s derived from a model, and that model is based on assumptions that are only educated guesses.

Find the right balance

As Colin Powell said:

“My own experience is that you get as much information as you can and then you pay attention to your intuition, to your informed instinct. Sometimes what my analytical mind says to me is not what I’ll do.

“Generally you should act somewhere between P40 and P70, as I call it. Some time after you have obtained 40% of all the information you are liable to get, start thinking in terms of making a decision. When you have about 70% of all the information, you probably ought to decide, because you may lose an opportunity in losing time.”

Know your audience. Know who decides, how many in each audience, and what the corporate culture expects.

The key is to make the numbers meaningful to your audience

Meaning before detail. Give them the bottom line and the conclusion of the numbers, and then use the numbers to support. Use only relevant numbers…every number should answer the So What test.

Provide the context first, and then the numbers. Believe it or not, an annual report provides an excellent example of this. You don’t get to the numbers section until way back in the annual report. The first sections are devoted to management’s explanations in the form of the Chairman’s Letter and Management’s Discussion of results. You need to find the meaning and the conclusions in the numbers and then back it up with the data.

Make the number something they can relate to. People have difficulty understanding large numbers; it’s best if they are on a human scale. For example, a $14 trillion national debt is incomprehensible—but not when you know it’s $140k for every non-government employee. (I hope that didn’t ruin your day.)

Percentages vs. real numbers. Another choice is to decide whether to express your numbers as real numbers or as percentages. In general, real numbers have greater impact, as demonstrated by research as well as common sense. Research shows that people generally find percentages easier to understand, but numbers have more impact. For example, in a study that sent postcards out asking for support for cancer research, a statement saying that cancer could strike thirty million Americans got more contributions than one that said it could strike 10% of the population. 10% is a concept; 30 million Americans is a lot of people.

Speak the language of their numbers. Senior level decision makers have a scorecard they follow very closely. At top levels, the scorecard may measure financial results such as revenues and profitability; at operational levels they are going to pay attention to process metrics such as throughput, cost per ton, etc. Know your audience, find out how they are measured, and use those measurements  in your presentation for credibility and impact.

Choose the reference point. If your reliability is 99.7% and your next best competitor’s is 99.4%, that doesn’t sound like a large enough difference to be worth touting. That’s because you’re starting from zero. But if you start from 100% and work backwards, the picture totally changes, as you can see in this explanation:

“Last year, Delta canceled just 0.3% of its flights, according to flight-tracking service FlightStats.com. That was twice as good as the next-best airlines, Southwest and Alaska, and five times better than the industry average of 1.7%.”

Seen in this light, Delta is 5 times better than the industry average. It is totally accurate and there is nothing underhanded about saying it this way. If you had to choose between two airlines for an important trip, would you pay a little extra for the one that has half the chance of being cancelled? Maybe the practical side of you says the chances of cancellation are so small in either case that you wouldn’t spend the premium, but what if the choice was between two surgical procedures, would that change your decision?