As an entrpreneur, product and people manager, I often find myself needing to design incentives that affect people’s behaviors. If you read the last sentence and the image formed in your head was that of a dangling carrot on a stick, i’ll request you to ignore it. I am talking about creating an environment where some really smart people become open to changing their behaviour ever so slightly to achieve larger goals. I wouldn’t blame you for the image that formed in your head though, as incentives are often poorly designed and are counter productive. In his book Mixed Signals, Uri Gneezy, describes in detail how incentives can be used to achieve desired outcomes or mixed signaling be avoided that leads to incentives that aim to improve one thing but focus on another. I have tried to summarise the key learnings from the book below.
Not the image you should have when you think of incentives
What is signaling?
Signaling in this context refers to the social cues that we send out to others. Signals are a credible way to indicate your values, abilities, aspirations and preferences. We all indulge in one or the other type of signaling but if the signal is easy to adapt it can be copied and if it is too loud it can come across as inauthentic. A rich person may not want people to know that they want them to know they are rich. Thus enter, Rolexes. The real Biker culture has strong signals like conspicuous tattoos on the neck and arms. This is to identify real bikers and eliminate casual bikers who can also buy bikes & riding gear which are, in comparison, cheap signals.
Through policies we can force a person to send a very strong signal or a weak signal for specific things. Disney World, for example, allows free tickets for kids under 3. People often lie about their kids’ age but if DW began asking children their age directly, two things can happen
- Kids will tell their true age and DW will eliminate the problem
- Parents will instruct their kids in advance on lying about their age which will sort of persist the problem to some extent.
Notice what is really happening with the second point. Parents are being asked to teach lying to their kids which is a strong signal to the parents that the price of the ticket is worth teaching kids to lie. This is a very strong signal and most parents won’t do it. This is a very crude example but it illustrates the point of how one can design policies and incentives that use the signaling mechanism to achieve desired outcomes.
Avoiding mixed signaling
Another important aspect of using signals is avoiding sending mixed signals. It is usually borne out of encouraging something but incentivising another. For instance, encouraging quality of deliverables but giving bonuses for quantity or encouraging long term goal alignment but praising short term milestones.
This is usually sorted out by adding more dimensions to the incentive structures. For instance, adding a rating system for the end user satisfaction along with evaluating people on quantity. This will ensure that people don’t just focus on quantity but also on quality.
Self v Social Signals
Most signals are social in that they allow one to communicate their values to others. Driving fuel guzzling semis or choosing to go with hybrid cars is a social signal that you care about the environment. In the early days of hybrid cars, they were seen as pricier options with limited comfort and safety features. However, one shining factor set them apart – their eco-friendly nature. Consumers began embracing hybrid vehicles not just for their fuel economy but to send a powerful message to the world: “We care for the environment, and we’re willing to sacrifice some comforts for it.”
Enter the Toyota Prius and the Honda Civic, both based on existing models. However, the Prius had a distinct and conspicuous look that drew consumers who wanted to stand out and signal their commitment to environmental causes. As a result, Prius won big. The importance of signaling was underscored by a 2007 survey, revealing that a remarkable 57% of Prius owners chose it to make a statement about themselves, far outweighing those who mentioned fuel economy or reduced emissions.
Self signals are signals that you send to yourself. Self signaling allow people to tell themselves that they are a good person or have certain values. Pay what you want coffee shops that are popping up in the US are a good example of self signaling. People readily pay more than the cost of the coffee because they want to feel good about themselves. The same people will not pay more than the cost of the coffee if they were asked to pay a specific amount. This is called the warm glow effect, what we feel when we help, volunteer or donate. The self signaling mechanism is very powerful and can be used to achieve desired outcomes.
Mental Accounting
We all have different mental accounts for various expenses. Think about it; we might spend 20 minutes searching for parking deals to save a few bucks but don’t hesitate to splurge on dessert afterward. The money across these accounts is not interchangeable and the story we tell ourselves to justify our choices are entirely different. While saving on parking is a sound financial choice, ending a family meal with a lovely dessert is a way to elevate a moment and make it memorable. To make incentives truly appealing, we must use the right currency that shapes the narrative we want.
Consider the story of a car review site that offered a $450 discount on car purchases. Surprisingly, this incentive didn’t have the positive impact they expected. Why? The $450 discount on a high-value purchase like a car got lost in the documentation and seemed insignificant. However, when they began offering the same $450 discount as “prepaid gas cards” for the same amount, demand improved significantly. Why? $450 is $450. Or is it?
When it comes to incentives, they can often sit in different mental accounts. In the gas card scenario, buyers know they will need gas and the $450 incentive seems to come across as something extra. Moreover, every time buyers filled up their gas tanks using the free card, they felt smart for making a clever buying decision. This simple shift in the perception of value made all the difference.
Regret Minimisation
Regret is caused by comparing what we did v what we could have done. Regret can also be caused by not exploring a new option. In that sense, exploration is it’s own reward. This is related to the loss aversion concept from Thinking Fast & Slow. We love to avoid regret at all costs. This makes regret a great tool for incentives and driving behaviours.
A classic example of this is the case where someone keeps buying a lottery ticket with the exact same numbers cause he thinks if he stopped buying and those numbers came up as winner, he would regret it immensely 🤷. A similar interesting case was of a lottery company that picked a random postcode from where a final winner will be chosen. This forced people in that postcode to think of a scenario in which someone from their address won the lottery but they didn’t. To avoid that regret, people flocked to buy lottery tickets.
I run a startup called Harmonize. We are hiring and if you’re looking for an exciting startup journey, please write to jobs@harmonizehq.com. Apart from this blog, I tweet about startup life and practical wisdom in books.