Incentives are at the core of consumer behaviour, and as product managers, understanding them is key to designing successful products. Every decision a user makes comes with a trade-off; exploitation vs exploration (sticking with what works vs trying something new) or efficiency vs equality (choosing speed and convenience vs fairness and inclusivity). These trade-offs influence not just individual choices but also how users interact with your product over time.

Let’s look at a few abstract examples and dive into important takeaways.

<aside> 🚔 The Prisoner’s Dilemma: Two people are jailed, they can either frame each other or keep quiet. If both people frame each other, they both go to prison for 10 years, the one who frames gets to walk, and the one who stays quiet goes to jail for the brunt of the crime (15 years). If they both stay quiet, they go into jail for one year each.

The game is symmetrical, for both person 1 and 2 - the best option is to frame to avoid the 15-year maximum penalty. Therefore the predicted outcome is they will both most likely frame each other.

</aside>

<aside> 🛣️ Braess’s Paradox: Should we have more nodes on a computer network to connect to different computers; and different nodes so that traffic can move quickly between pairs of nodes?

Similar to our daily lives, we have 2+ routes to get to work. Our goal is to use the shortest time possible and you just do what the app tells you to do. You can take a T road where T is the number of drivers here, it will take you T minutes to get through this patch to expressway A. So say if there are ten drivers there, it will take you ten minutes. And then from expressway A to the destination, it will take you 45 minutes, no matter how many drivers are on that patch of road. Or you can go to freeway B where things are flipped. It takes you 45 minutes, no matter how many drivers there are. And then it takes you T minutes where T is the number of drivers on that particular patch of road.

Before adding a shortcut, the predicted outcome is that 50% of drivers will take expressways A &B respectively. (Similar to a supermarket where the lines will even out in length)

When we add a bidirectional shortcut, where drivers can cross from either expressway for 0 minutes, drivers then think they can take the T road the entire way. Every driver thinks about minimising their time of travel and as a result, more people take the shortcut and thus take longer to get to work.

</aside>

Practical Takeaways

1. Mitigate Negative Outcomes: Similar to the Prisoner’s Dilemma, users might make suboptimal decisions when acting in self-interest. We need to implement features that encourage collaboration or positive outcomes, such as transparent reward mechanisms or peer validation systems.

   1. Kickstarter (Crowdfunding platform):
      1. Backers might hesitate to fund a project if there’s a risk of it not reaching its goal. To mitigate Kickstarter implemented “all-or-nothing” funding, where funds are only collected if the goal is reached. This encourages collaboration and positive outcomes.
      2. Transparent reward mechanisms (like showing a leaderboard of top backers) and peer validation systems (where backers can leave supportive comments) create a community environment that motivates contributions.
   2. X savings feature**,** in ****neobank app:
      1. Users may be tempted to withdraw savings early if they see an immediate financial need, which could negatively affect their long-term financial goals. To mitigate this, the neobank can introduce features like “savings goals lock,” where users earn higher interest rates if they keep their money in for a set period.
      2. A peer validation system, such as social savings challenges where friends encourage each other to reach financial goals, promotes collaboration and positive behaviour.

2. Anticipate Self-Interested Behaviour: Assume users will make decisions that are beneficial for them in the short term. Design your product to align incentives accordingly, ensuring that user interests also benefit the overall product ecosystem over time.

   1. Uber (ride hailing app)
      1. Drivers may prioritise accepting shorter, high-demand trips to maximise their earnings in the short term. To align their self-interest with the company’s ecosystem, Uber implemented incentive structures, such as surge pricing or bonuses for completing a set number of rides. These incentives ensure that drivers' short-term goals also contribute to a well-functioning and balanced service over time, benefiting both drivers and riders.
   2. Pay As You Go**,** in MDF ecosystem:

      1. Users might be inclined to miss or delay payments on their device financing if they perceive it as financially beneficial in the short term, especially if they are facing financial constraints. To address this, the MDFs implement incentives that align the user's short-term needs with their long-term interests. For instance:

         Offering small, tiered rewards for on-time payments, such as loyalty points that can be redeemed for discounts on future devices, airtime top-ups, or access to lower interest rates on future loans.

         By providing these immediate benefits for responsible behaviour, the app ensures users are motivated to make timely payments, benefiting both the user (through tangible rewards) and the app's ecosystem (by reducing default rates).

3. Optimise for Symmetry: Where applicable, ensure that the design of interactions is fair and predictable, similar to how users in a symmetrical game expect equal consequences. This builds trust and reliability.

   1. Ebay (peer to peer market place)
      1. Both buyers and sellers expect equal treatment and consequences for their actions. eBay has symmetrical feedback systems where both parties can leave reviews for each other. This fair and predictable interaction design builds trust and reliability because users know that both parties are held to the same standards.
   2. Neobank’s referral programme
      1. Both the referrer and the referred person could receive equal rewards, like a monetary bonus or waived fees. This symmetry in reward structure makes the programme predictable and fair, encouraging users to participate without feeling one-sided. Additionally, having a transparent set of rules that apply equally to all participants builds trust in the bank’s services.

4. System Design and Bottlenecks: Braess’ Paradox highlights how adding more options or shortcuts can unintentionally create inefficiencies. Be mindful when designing product flows or networks ( recommendation engines or content delivery, and consider how they might introduce complexity)

   1. Netflix
      1. Adding more content recommendation categories might seem beneficial. However, Braess’s Paradox suggests that this could overwhelm users and make finding something to watch more difficult. Netflix addresses this by simplifying its recommendation system, often limiting categories to what is most relevant to the user. They also avoid adding too many options that could create decision fatigue or slow down the user experience.

5. User Routing and Choice: When providing multiple pathways for user action (e.g., different workflows or options for task completion), simplify decision-making to avoid choice overload and inefficiencies. Test and model user journeys to optimize for real-world scenarios.

   1. In a neobank’s bill payment feature, offering numerous ways to pay bills (like paying manually, setting auto-debits, or scheduling reminders) might overwhelm users. To simplify this, the neobank could introduce a default suggestion, such as auto-debit for recurring bills, with a simple one-click setup. Modelling and testing these pathways ensure that users can manage their payments efficiently, reducing the cognitive load and enhancing user experience.

6. Utilise Predictive Modelling: Just as drivers or users act based on perceived benefits, use data to model and anticipate user behaviour. Build simulations or A/B tests to understand how users will react to new features or changes.

   1. A neobank can use predictive modelling to anticipate when a user is likely to need a short-term loan based on their spending habits and upcoming bill payments. By testing personalised loan offers through A/B testing, the neobank can determine which messaging and loan terms (e.g., interest rates or repayment plans) are most effective in converting users, ultimately helping users while maintaining a strong loan repayment rate.