How Numbers Drive Behavior, for Better and Worse

The joy of numbers

I was one of those kids obsessed with counting stuff.

I collected business cards, and kept a running daily tally, stopping only when I reached 10,000, and after a friend of the family asked me why I was wasting my time on that when I could be learning a new language. I timed how long it took me to walk home from the subway stop in Toronto where we lived, logging my times in a notebook and getting a strange satisfaction out of beating my previous record by 15 seconds. I counted all books in the Hardy Boys series I avidly consumed. In fact, in high school, I began keeping a list of every book I read, a habit that continues 35 years later.

Starting in eighth grade, I began tracking my own modest earnings and expenses. Perhaps an early sign of fastidiousness that I periodically worked to overcome via reckless teenage stunts.  (but our family still keeps a budget). In my twenties, my focus shifted to beating my personal best marathon time. I mostly didn’t set specific targets for myself. It was more about watching the numbers grow, or shrink, as the case may be. The intrinsic awards were enough. They pushed me to do better.

All that is fine and good. It is a truism that numbers affect our priorities and our actions. The standard way to judge success is by setting targets and measuring achievements.

Targets can be internal – a New Year’s resolution to lose 20 pounds next year, or do 40 squats per day. The proliferation of fitness apps that track how many steps you take every day, how many hours you sleep at night, and other metabolic processes – although of dubious reliability – are all the rage. All of these things drive behavior, in relatively benign ways.

Targets can also be external, of course, such as those set by a business, as in monthly sales targets or daily active users. In the development field, projects use results frameworks with a detailed breakdown of quantitative targets against which project goals are measured. These are all instances of using numerical targets to drive decisions.

Easily quantifiable targets are powerful motivators. Numbers are probably so compelling because they are, or at least seem, objective. They are mental shortcuts that are easy to “get.” They focus the mind, guide strategy, and help plan for the future.

Beware of the quantitative bias trap

What’s not to like about them? Well, quantitative targets, if not used carefully, and when not balanced by other goals, can lead to distorted outcomes. They can affect behavior in pernicious ways as well. The wrong incentives – divorced of context and isolated from other, less quantifiable goals – can have negative unintended consequences.

So it is that studies of the benefits of sleep tracking apps, that quantify hours of rest and create benchmarks in users minds (I have to get my eight hours in!) have found using them often worsens sleep quality.   People become more anxious and stressed when the app shows they are not hitting their sleep target, even while the ability of these apps to correctly measure sleep has been called into question.

The observation that using numbers as targets has negative side effects has been made famous by the British economist Charles Goodhart. In what has become known as Goodhart’s Law, he states that “when a measure becomes a target, it ceases to be a good measure.” Why is this so? One of thinking about it is that ways are found of hitting the target that hurt or distort other important goals. Other aspects of a plan or policy may get deprioritized, given less funding. Anything that doesn’t have a target on its back loses the appeal of the hunt, and risks falling by the wayside.

Campbell’s Law, named after the social psychologist and experimental evaluation pioneer Donald Campbell, is a variation of this: “the more any quantitative social indicator is used for social decision-making, the more subject it will be to corruption pressures and the more apt it will be to distort and corrupt the social processes it is intended to monitor.”

This is just not a management issue – it can have life and death consequences.

Back in 1993, Domino’s Pizza scrapped its 30 minutes or its free guarantee when it became clear the company was were putting drivers’ safety at risk (and after they lost a $79 million court judgment).

Which brings us to the current upheaval shaking the country and the world and the world: the bolt of awareness around racial injustice and police brutality sparked by the May 25 killing of George Floyd.

A recent Vox article on key police reforms  in the wake of the Michael Brown killing in Ferguson, Missouri, reported that when the Justice Department investigated the Ferguson Police Department in response to the protests, it found that police “were encouraged to ticket as many people as possible with the explicit goal of raising as much revenue as possible from fines and fees. But to do this, police targeted the most vulnerable — mainly, black residents — with frivolous charges.”

While quotas of this type have been declared in some states, and departments won’t admit to using them, New York Police Department whistleblowers reported that, in order to meet their quotas, officers would often go after groups that have little political power.

This is one way that the numbers game turns ugly: targets and distorted power relations are a formula for disaster. A man with a gun – and a number in his head – versus a group without a voice.

It is unclear how much unwritten quotas, or the mere desire to get more “points,” drives police behavior. Maybe death is not the most common outcome, but it is not hard to see how this potent cocktail pricks at the wounds of racial injustice. 

To put it plainly – when quantitative targets are the sole priority, qualitative aspects such as, well, dignity, freedom, and safety, to name a few, get undermined. These are unintended consequences, which make life a little worse, in the best case scenario. However, as we have seen, they also have potential strengthen systems of inequality and oppression, leading to the death of at least some.

Numbers matter. They drive us toward goals. I was a young counting aficionado and it did me little harm and maybe some good. But equally important are putting guardrails in place so that we don’t let numbers blind us, or others, to what actually matters.