The Formula That Helped Alarmy Drive User Behavior (feat. HCI)

As Alarmy wakes up 50 million mornings and grows into a global alarm app, there's a question we hear often.

"What makes Alarmy so different from other alarms?"

To this question, I usually give a simple answer: "We just thought a bit more about how to wake users up effectively," but this alone doesn't quite convey the actual depth of our deliberation. In a previous post, I introduced the various things we do behind the scenes that aren't visible on the surface, but I didn't really touch on product planning, so I want to take this opportunity to share one of the frameworks we considered while improving the product.

The formula that drives desired behaviors

Back when I was juggling my master's degree and running Alarmy, I read an interesting paper. At the time, I was researching the field of Persuasive Technology—how to change people's behavior—and I came across a paper by Dr. BJ Fogg, a pioneer in this field, which got me thinking about design methods that could better trigger desired behaviors.

To be fair, it's quite a common concept now since it was introduced in Hooked, a book widely read during product planning, but encountering this model back then helped me find the seed for why Alarmy wakes people up better than any other alarm app, and how we could wake them up even better in the future.

This model, called the FBM (Fogg Behavior Model), is a powerful and intuitive framework that simplifies how behavior occurs into three elements, and can be explained as follows:

1) When there is a motivation to act (Motivation)
2) When the behavior is easy to do (Ability)
3) When a cue occurs (Trigger)
→ The behavior happens (Target Behavior)

To put it in a fancy way, this is expressed as B = MAT (Behavior = Motivation * Ability * Trigger). It's easiest to understand with a simple example; if we look at a situation where the target behavior is "practicing guitar" from a B = MAT perspective, it can be interpreted like this:

• Motivation: The drive for the target behavior could be the desire to 'play the guitar well.' What if there is no motivation to play well? This results in a Low Motivation state, making the behavior unlikely to occur.

• Ability: Also referred to as Simplicity, this indicates how easy it is to perform the target behavior. Here, 'how easy is the environment for practicing guitar?' could be an example. What if there's no guitar at home and you have to go to a studio just to practice? This results in a Low Ability state, making the behavior unlikely to occur.

• Trigger: Even if you have the desire to practice and a guitar at home, if you don't actually think to yourself 'let's practice guitar!', the behavior cannot happen. Here, a repeatedly scheduled reminder could be an example of a Trigger.

What if you applied this to your service?

Whether you're a PM, product planner, designer, or engineer, if the people building the product gather internally to discuss what the target behavior is and which part of MAT is lacking, you'll be able to see things from a new perspective (in fact, we tried ideating together like this in a workshop, and everyone easily came up with ideas regardless of their roles and responsibilities).

I've summarized some points below that can help when examining your own service from a B=MAT perspective. I hope this helps when discussing which elements to leverage and how with your internal teammates.

There are 3 types of Motivation

• Pleasure / Pain
  These typically act as immediate, rather than future-oriented, motivational factors. For example, there's the pleasure of a reward you get when you perform a behavior, or the pain you endure when you don't.

• Hope / Fear
  These are more future-oriented motivational factors, involving the hope that something good will happen in the future if you act, or the fear that something bad will happen. For example, working out with the hope of building a good physique, or getting vaccinated to avoid the fear of catching the flu.

• Social Acceptance / Social Rejection
  This relates to motivation based on how you are perceived socially. Examples include trying to be socially accepted by dressing well, or avoiding swearing or harsh words in everyday life so as not to be socially rejected.

There are 6 types of Ability (Simplicity)

• Time
  How long does the target behavior take? If it takes a lot of time, Ability is low.

• Money
  How much does the target behavior cost? If it costs a lot of money, Ability is low.

• Physical Effort
  How much physical exertion is required to perform the target behavior? If it requires a lot of effort, Ability is low.

• Brain Cycle
  How much cognitive load is required to perform the target behavior? If it takes a lot of thought, Ability is low.

• Social Deviance
  How normative is the target behavior from the perspective of others? The weirder it looks to others, the lower the Ability.

• Non-Routine
  How regular is the target behavior? The more non-routine the behavior, the lower the Ability.

There are 3 types of Triggers

• Spark
  Used when Ability is high, but Motivation is low. It serves to remind or emphasize Motivation. For instance, you could send a motivational photo along with a workout time notification.

• Facilitator
  Used when Motivation is high, but Ability is low. It serves to help increase Ability. For example, you could send a message alongside an update notification stating that the app update is easily completed with just one click.

• Signal
  Used when both Ability and Motivation are high. You simply need to provide a reminder so they don't forget.

Motivation and Ability are complementary.

• If Motivation is extremely high, the behavior can occur even if Ability is low (and vice versa). For instance, if you encounter a service with a ridiculously complex sign-up process, you normally wouldn't bother, but if your drive to use that service is massive, you'll endure the tedious process. Therefore, I recommend figuring out which element is easier to elevate in your own service and targeting that first (as a side note, it's generally easier to improve Ability than Motivation).

Finally, how exactly did Alarmy utilize B=MAT?

Let's apply it to Alarmy

Alarmy's core user base consists of people who dismiss their alarms and go back to sleep, and one of the core features for them is the "photo mission," where the alarm is only dismissed if they walk to a designated spot and take a picture when it goes off.

Comparing the state of dismissing an alarm in other alarm apps versus dismissing an alarm in Alarmy clearly reveals why Alarmy is better at waking people up. As shown in the image below, using Alarmy's photo mission increases the Ability for the target behavior after dismissing the alarm compared to a standard alarm.

• Target Behavior: Waking up from sleep

• Ability: Lying in bed (standard alarm) VS Standing far away from the bed (Alarmy)

Between lying in bed and standing far away from it, in which situation are you more likely to wake up? Naturally, it's when you're standing outside your bed. Ultimately, because the user is put into a state where it's easier to wake up, they can effortlessly shake off sleep when using Alarmy. I used increasing Ability as an example here, but depending on the case, there are plenty of points to deeply consider across both Motivation and Trigger as well.

Wrapping up

Depending on what your service's target behavior is and which part of MAT you aim to improve, the direction for enhancing your product can vary dramatically. Even if you don't apply it to a service, plugging in a personal habit you want to maintain can help you strategize ways to stick to behaviors you usually struggle to follow through on.

Now, let's circle back to the answer to the initial question.

"We just thought a bit more about how to wake users up effectively."

To unpack this a bit: we deliberate on what target behavior our service fundamentally needs to focus on, and to improve it, we continuously form hypotheses, test them, and progressively enhance our product. Personally, I believe these efforts consistently compound to create our differentiating edge against other alarm apps.

At DelightRoom, we build our product by relentlessly pursuing its essence and navigating these various challenges together. If you want to think fiercely with us and build a service that impacts a massive number of people, please check out our hiring page.