Thought Leader Interview:

Dilip Soman

You have said that every organization is fundamentally in the same business. Please explain.

Whether it be a for-profit business or a government agency, every enterprise is in the business of behaviour change. In each case there is a collection of individuals interacting with other individuals with a particular goal in mind. It could be a sales team dealing with external customers to sell a product or service; HR managers dealing with an employee base to optimize productivity; or a CFO dealing with shareholder concerns. All of these interactions involve some form of desired behaviour change.

Some scholars have focused on the persuasion aspects of changing behaviour, but I would argue that what we are really doing is helping people get things done and make welfareenhancing decisions more easily. Once you accept this, understanding human behaviour moves to the forefront of your challenge. And in many cases, time is of the essence, because we want people to do things sooner rather than later. We want them to open a retirement account before it’s too late; we want them to switch to a healthy diet before it’s too late; and we want them to make environmentally friendly choices today. Within organizations, the challenge is the same. Maybe you want your employees to be more productive, more efficient or more inclusive — and the sooner the better.

The problem is, even when people want to do something, they often don’t do it. Describe how the ‘intention-action gap’ gets in the way of behaviour change.

This is one of the most robust phenomena from the behavioural research, and it is true of individuals as well as organizations. The basic idea is that we often don’t do what we intend to do, and this happens for a number of reasons. Nobel Laureate Richard Thaler and his co-author Hersh Shefrin boil it down to the dichotomy between the planner and the doer within each of us. They argue that each of us has two entities within our minds: A planner who thinks about the future, seeks continuous improvement and wants long-term goals to be addressed; and a doer who actually has to go out and get things done. And for individuals and organizations alike, the doer often doesn’t get the job done.

At an organizational level, most leaders will say they want to be long term in their thinking and actions. They clearly state this in their vision statement and annual reports, yet ultimately, many of them end up making myopic choices. That’s because there are also lots of shorter-term considerations to deal with — like quarterly reporting. For individuals, the challenge is similar. The planner within sets out these lofty goals, but as she lives through day-to-day life, the doer within has competing opportunities and challenges. Just as a business has a quarterly report to file, we all have a daily to-do list: we have kids to drop off at school, classes to teach. In short, life gets in the way.

This is a really important aspect of behavioural psychology that has significant implications for the modern enterprise. When we set out to attempt to change behaviour, our emphasis tends to be on convincing people to do things. We spend time and resources trying to persuade them with advertising campaigns or other messaging. But as indicated, the challenge isn’t persuasion per se as much as it is helping people get things done. That is where our focus should be placed.

By now, most of our readers are familiar with the term nudge. But what on earth is sludge?

Small factors in a context — whether it be in a government service, a financial institution or a retail environment — either facilitate or impede the end user in accomplishing an objective. If they are designed to facilitate, these variables are called nudges; but there is also an entire category of situations where the contextual variables actively impede activities that are in a consumers’ best interest. This is called sludge, and it can be seen as the evil cousin of nudge. Sludge exists in all walks of life — oftentimes, not intentionally.

I have a degree in Engineering, so I tend to think about things in those terms — including behaviour change. The fact is, we can model behaviour change as a ‘plumbing problem’. Think about water flowing through a pipe, and the behaviour change component is that you want to get someone from point A to point B of the pipe. In Nudge, Richard Thaler and Cass Sunstein essentially make the argument that if the pipes are clean and people are motivated, the fluid will flow right through. Sludge is all the stuff that blocks the pipeline.

In and of itself, sludge is not harmful to an organization. It is the fact that it feeds into a lot of negative behavioural tendencies that makes it so dangerous. For example, imagine you are the kind of person who procrastinates. In the heat of the moment, you will be supremely motivated to get something done, so chances are, you will get it done; but as time passes or pain points arise, you might not be able to complete the task. This happens all the time, on many levels.

A couple of years ago, I was eager to sign up for online tax filing because I was fed up with all the paper forms. In my frustration I was highly motivated to sit down and get my online account up and going. Everything was humming along — until I got to a point where the tax authorities needed to send me a web code to access my new account; and they would be sending it via ‘snailmail’. Obviously there were privacy reasons for this, but by the time that web code arrived a week later, I was no longer in a motivated state. You can guess how the story ended.

Anything in a particular context that impedes choice and action can be viewed as sludge. If people do not have any patience in a particular context, offering them a long, multi-stage process is sludge; if they are faced with too many options, that can also be sludge because people struggle with overchoice. What complicates matters is that sludge is insidious, because two different people might view the same aspect of a particular context as sludge.

As you indicate, nudging sets out to make choosing easier for people. But there are potential negative effects to facilitating choice. Please elaborate.

One of Thaler and Sunstein’s mantras is that making things easy is always good because it helps people accomplish tasks. But I would argue that making things easy is not always a good thing — and I’m pretty sure they would agree. There are many domains in which friction or impedance can actually be welfare-enhancing for the individual.

Take something like divorce. You would never want to make that too easy — whether it be for personal relationships or organizational partnerships. You don’t want to be in a situation where you fight with your partner, you get home, push a one-click button and before you know it, the marriage is over. The fact is,

Anything in a particular context that impedes choice

and action can be viewed as sludge.

oftentimes the same impatience that motivates us to do things can also drive us to do silly things without thinking enough about it. That’s why some form of friction is often helpful.

Think about overspending. We know that when it becomes easier to spend, people tend to spend more. It is all too easy to simply tap a credit card and make a purchase; so sometimes, we might want to introduce some friction into the context. My former PHD student Amar Cheema and I wrote a couple of papers on the idea of introducing such frictions. We call them ‘decision points,’ and we found that simply by interrupting a consumption process, you can get people to be more mindful about their consumption. A decision point is any intervention that adds friction to a process with the objective of getting the individual to pause and think.

Friction shows up in lots of places in life. In negotiations, there is often a ‘cooling-off’ period where the thing that you negotiate doesn’t become binding until you ‘sleep on it’ and decide that you truly want to proceed. That’s because quite literally, we often need to cool down. We sometimes agree to things in the heat of the moment and later wish we hadn’t.

Making things too easy has its pitfalls, and sometimes these are insidious. For example, subscription traps. I’m sure many readers have been accidentally defaulted into purchasing a subscription to a magazine or a news feed when they didn’t intend to. We often need a bit of friction in our systems.

You and your team have identified three main sources of friction or sludge: the process itself, the nature of the communication and different levels of inclusivity. Please summarize each.

This past summer I worked on this with BEAR researchers Daniel Cowen, Niketana Kannan and Bing Feng. We interviewed people and analyzed their interactions with organizations, and we concluded that friction can exist on three levels.

The first is the process that people have to complete to accomplish a task. Think about consumer services: How many steps are required to complete the task? How many redundancies are there in these steps? Does the user have to provide the same information to multiple people during a single transaction? Does she need to fill out a form online as well as speak to someone on the phone? These physical characteristics of a process are one of the most common sources of sludge.

I remember back in the day when we used to have Palmpilots — these little handheld computers that preceded the smartphone. At one point I learned that there was a person working at Palmpilot whose entire job was to make sure that no process on the device required more than two or three taps. Just imagine if every organization embraced this philosophy! ‘Can our user accomplish the task/pay the bill/get a refund/make an exchange/or update their driver’s licence in no more than two or three steps?’ How amazing would that be?

The second source of sludge is communication. Are you speaking the language of your target consumer? Is your message comprehensible to them? Are you providing people with the right information at the right time? Sometimes, we give people plenty of information, but it comes either too soon or too late. For example, people might be psychologically committed to doing something, like choosing a credit card, and then they receive the terms and conditions information too late in the decision-making process. Also, is the information available in the right mode that the consumer wants? Can they easily access it? Simplicity and relevance of communication is key, as is making sure that there is no shrouding of the pertinent data.

The third aspect of sludge involves varying degrees of inclusion. In many cases, your process and/or your communication might — intentionally or otherwise — exclude certain segments of the population from accessing your products and services. As a simple example, if I require everyone to submit a form digitally online, I am automatically excluding people who don’t have access to the Internet as well as those who are simply not comfortable doing things online.

Sometimes we see more insidious types of exclusion. For example, if you set up a service centre that is only open in the middle of the work day, clearly people with full-time jobs cannot use it. Or, say I am a fast-food restaurant manager who only decides the night before who will occupy the various positions in the restaurant the next day. There might be perfectly good operational reasons for this — but in fact, I will end up excluding single parents, who will find it impossible to find childcare that quickly.

We also sometimes exclude people because of psychological factors. In our country, we have a welfare program for children’s education called the Canada Learning Bond, which requires people to go to a bank in person and open an account. What we found in our research is that many people refused to do that, because they felt they were publicly acknowledging that they needed help. For some cultures, that was a deal breaker. Oftentimes, emotions like embarrassment, guilt, anxiety or awkwardness create exclusions and as a result, people don’t engage with the service.

Sludge is extremely common in the realm of customer service. Why is that?

Put simply, we have made it difficult for people to interact with our organizations. Sometime during the Industrial Revolution, organizations became fixated on efficiency and throughput. Leaders started thinking about breaking down the customer service process into small manageable pieces and assigning a different department to manage each aspect. Over time we got to the point where most frontline agents don’t have access to the entire file for a given consumer, and as a result, they don’t understand the consumer as a human being. This is why, when you call up your cable provider or your mobile phone company, you end up speaking to someone who can only help you with one aspect of your service. If you have a question or a concern that is slightly off-topic from their area of expertise, they have to transfer you. That is sludge personified.

What can be done about sludge?

As I said earlier, identifying sludge is difficult because what is sludge to one person might not be sludge to another. That’s why it’s so important to continuously observe interactions and stay close to your customers. Also, whenever you set out to optimize variables, you need to make sure that they are human-centric variables.

Take wait times. It turns out that shortening them is not always the solution to keeping people happy. Sometimes, people are happy to wait longer. Take Disneyland. People will wait for hours in line for a ride that lasts a minute and 20 seconds. But because they’re occupied throughout the wait and are having fun, it’s not a problem. The human variable of interest is not always the total amount of time. If you can keep people sufficiently engaged, nobody’s going to be looking at the clock. That’s why you need to think carefully about which human variables are at play, and then use data to optimize those variables.

It is critically important to develop scorecards or indices that enable a company to audit its processes, identify specific factors that create sludge and then score that sludge over time. And keep in mind that every process will need a different scorecard, because every process is inherently different. Empathy doesn’t come naturally to people in organizations — or to people more generally. Hence, it’s really important to be systematic about measuring it. The good news is, once it’s measured, it can be managed.

By the way, technology is a whole new breeding ground for sludge. It can be immensely helpful in some ways, but it can also be a significant source of friction: websites crash; people forget their passwords; they can’t find what they need. And as indicated, there are entire segments of people who are not familiar with technology at all. Unless you have other channels to help them, they will see your mere use of technology as sludge.

In many organizations, people are rewarded for thinking big. But your work shows how important it is to also think small. Please discuss.

These days we are often asked to come up with breakthrough insights — and that is very important. But I do think that my work and that of my colleagues tells us that the success of those big things is actually contingent upon some very small things — such as the friction in an environment.

Consider the case of an insurance company that set up a beautiful program encouraging people to get an annual health checkup. The friction there wasn’t that people were too busy to get the checkup, it was that they didn’t want to have to call to make an appointment. If you address that friction, it turns out that people will go and get their annual check-up. Many dentists have figured this out: by the time you leave their office, they already have you booked for your next appointment. Likewise, everyone understands the importance of retirement savings, but in many countries you have to open an account for yourself. In the U.S., Prof. Thaler and Shlomo Benartzi found that any

intervention that automatically enrolls you into an account improves pension savings significantly — just by removing the step of having to physically go into a bank to open an account.

Again, this is not about persuading people; it’s about making things easy so that the doer within each of them can get things done. It’s so important to look at the small environmental factors that can trip people up. You can have big ideas for an amazing new offering, but if you don’t pay attention to all the little humancentric factors, it will come crashing down.

You founded Behavioural Economics in Action at Rotman (BEAR) in 2015; and in 2019 you launched Behaviourally Informed Organizations (BI-ORG). What is this new entity?

BEAR is a University of Toronto entity funded by Rotman and the University, while BI-ORG is a global initiative funded by our federal funding program, the SSHRC (Social Sciences and Human Resources Canada). It currently involves 20 academic researchers from Canada, the U.S., Europe and Asia and 17 partner organizations drawn from industry, government and the not-for profit and social welfare sectors. Our joint mission is to help organizations around the world become more behaviourally informed.

The partners involved in BI-ORG have a shared understanding as to what a behaviourally informed organization is: It is an organization that understands that its internal and external stakeholders are humans and not ‘Econs’ (purely rational beings). The behaviourally informed organization understands what it means to be human, and it designs its products and services accordingly. And importantly, it is evidence based: It uses experimental data to learn. It tests continuously, it learns continuously, and it adapts continuously. It looks for and reduces sludge and it tries its very best to be ‘human compliant’.

What exactly does it mean for an organization to be ‘human compliant’?

Organizations must keep in mind that they are designing for human beings who are cognitively lazy, forgetful, emotional and myopic. Many of them spend a lot of their time and energy complying with laws and regulations. My wish is that they would spend a bit of that energy on being human compliant: Designing processes for humans, designing incentive structures that humans respond to, and more generally, designing workplaces where humans of all shapes, sizes and preferences are welcomed. This is the challenge that my colleagues and I have set for every organization.

Dilip Soman is the Canada Research Chair in Behavioural Science and Economics at the Rotman School of Management and serves as Founding Director of the Behavioural Economics in Action Research Centre at Rotman (BEAR). He is the author of The Last Mile: Creating Social and Economic Value Through Behavioural Insights (Rotman-utp Publishing, 2015) and teaches a massive open online course (MOOC) titled “BE101X: Behavioural Economics in Action” on EDX.