Path Dependence and the Danger of the Top 10

Path Dependence and the Danger of the Top 10

by Grant Wenzlau


There is a common misconception that advertising doesn’t work on you—it works on other people—but not you. Many of us try to be conscious of the forces shaping our decisions, but even beyond advertising, many of the things we interact with warp our behavior in subtle ways.

The truth is, a lot of decisions you think you are making rationally were already made for you. Even if you are a “conscious consumer,” your decisions have been formed by information gleaned from a particular perspective. The deck is nearly always stacked in a certain direction.

There is value in being aware of the unseen forces that are always around us. As marketers, we need to understand the science of decision making, and as consumers, we need to be cognizant of these forces and their implications.

Path dependence is the idea that the set of decisions one makes today are shaped by decisions made in the past. Understanding this requires stepping back and exposing the hidden forces that drive decisions. When you start to be conscious of path dependence, you start to see these previously determined paths everywhere. Even things as simple as deciding where to eat lunch or what song to play can be driven by path dependence.

One of the most common types of this bias can be found in top ten lists.

Top ten lists are useful. It is how Google and Yelp structure results for anything you search for–from the best cafes or bars to news and top stories. Increasingly, we consume content in lists that are ordered by popularity. On Spotify, top songs are at the top of each artist’s page just below a big green play button.

Top ten lists can be useful shortcuts to finding what we are looking for. But they also can end up distorting our reality.

Because content is ranked by popularity on many of our digital platforms, it ends up skewing the data towards the top of the list. The mere placement that content occupies on the list changes the way we think of and consume it. Ordered lists like this end up creating a power law distribution, which is a relationship between quantities where one is exponentially higher than the next. So, the #1 option on the list becomes exponentially more visited, played, shared, than the 10th.

The danger of these paths is that at a certain point, the top slot becomes cemented at the top simply because it is the top slot. There is a substitution in our question: what is the best (which can mean any number of things) to what is the most popular? We no longer listen to the top song because it is good–we listen to it because it is the most popular. And it is the most popular because it is positioned at the top of the list. It becomes famous for being famous and will stay at the top because of the real estate it occupies.

Top Ten Leon Top Songs Graph edited
Chart 1
Top Ten Leon Cominghome Graph
Chart 2

Check out how path dependence leads to an extremely uneven distribution in the above example. The first chart shows the top 5 most popular songs on Spotify by Leon Bridges, whose debut album Coming Home was a big hit. His top song, “River,” has almost 70 million plays, followed by “Coming Home” with over 58 million plays.

Next, see chart 2, which shows the number of plays for all of the songs that are on the same album—but not featured on the top of the page in the Most Popular Songs list. Most of the other songs on the album float between 4 to 7 million plays.

The song “River” (shown on Chart 2 for scale) very well may be Leon Bridges’ best song, but this amount of difference surely distorts the reality—exaggerating the value and significance of “River” when compared to Bridges’ other tracks. In this chart, you can see that “River” has more plays than the rest of the songs on the album combined. Is “River” 10x better than all of his other songs? Or is its position as the top song the driving force behind its number of plays? Is it the simply the top song because it’s been placed in the top song position?

Path dependence is a shortcut that can obscure the full story. It has a misleading air of objectivity and of having rigorously studied all of the options. But in fact, it is a snowball that has gained momentum and overtaken the other options. Path dependence and power law would never let you discover “Daisy Mae” by Leon Bridges, which is a great song.

This may be harmless when it comes to music, but this same process and design bias can influence all of the content we see, consume, create, and share. It can extend to the people we interact with and the ideas that gain legitimacy. Path dependence drives everything in our world from the people we meet, the places we go, and the ideas we discuss.

The simple design choice of a top-anything list can have the effect of creating a bias in our behavior. It has the effect of driving towards homogeneity instead of push- ing us towards the edges to discover new things. Path dependence is just one of the biases we all encounter every day—on and off digital platforms. As marketers, we should realize the forces that direct our audiences towards or away from us. We should be conscious of how we drive an audience’s behavior and where we are pushing them.

As consumers—and human beings seeking connection—we must be aware of the forces guiding our decisions. Escaping path dependence and other biases may mean deviating from the well-worn road and heading towards the uncomfortable, uncertain edges. It can be scary. It can be lonely. But who knows, by deviating, you might find what you didn’t even know you were looking for.