Sketching the Big Picture of Our Product

Pausing our sprint velocity to envision 10x growth

When you're constantly running short, fast sprints, you often lose the thread connecting where you've been to where you need to go. We might think we've been running well without getting lost, but sometimes when we finally arrive, it's not the destination we expected, or our overall journey ends up looking completely zigzagged.

That being said, trying to map out the entire journey from the get-go isn't efficient either. For the people doing the work, it demands too much effort, and our prediction accuracy plummets. Plus, it makes it incredibly difficult to stay flexible and adapt to what users in the market actually need.

At DelightRoom, we've always grown our product with a combination of "small and fast." However, as we recently decided to undergo a massive refactoring of Alarmy's core alarm logic, we were able to secure a solid four weeks of time for planning and design. Taking this opportunity to pause and reflect on our product, I started noticing quite a few issues that had piled up from our sprint-heavy engineering approach.

1) Clunky user journeys

As one-off flows and screen-level growth initiatives piled up, the friction of stringing multiple features together became severe. This awkwardness stood out even more in our core flows, like user onboarding, the purchase screen, and setting up or dismissing alarms.

2) Inconsistent feature hierarchy

Many features had been added without considering the overall balance of the product. It wasn't uncommon to see relatively unimportant features taking up way too much screen real estate, or features that should have been grouped together scattered all over the place.

3) Lack of planning for 10x growth

Because we were just repeating small-scale planning within already established flows, the impact we could create was inherently limited. It was hard to think about product expansion from a much broader perspective.

During the refactoring period, each squad decided to sketch out an "End Picture" to solve these exact issues. Our goal was to seamlessly redesign the macro flows, balance the product, and come up with planning that could deliver massive impact all at once.

The subscription squad decided to overhaul the entire subscription journey, the core squad aimed to revamp alarm types and their respective editor views, and the engage squad began envisioning the future of the Morning Tab that users see right after dismissing an alarm.

And the squad I'm leading ended up designing a brand-new journey aimed squarely at "10x growth."

Namely, a reward system and characters.

01. How can we dramatically increase retention?

The starting point of our 10x growth journey

The main problem we wanted to solve was "retention." To achieve a massive leap in revenue, global expansion was essential, and that meant we desperately needed to boost retention in countries outside of the US and Korea. So, we started gathering references to see how other services were successfully driving up their retention metrics.

02. Two battle-tested methodologies

The surefire strategies everyone is choosing

Upon digging in, two methodologies with clearly proven effectiveness stood out.

The first approach leverages extrinsic motivation. This is heavily used by "app-tech" (reward-based) services, and in Korea, many apps—led by Toss—have applied similar planning. It's a strategy with a crystal-clear effect on increasing retention.

The second approach stimulates intrinsic motivation.

This involves quantifying or visualizing a sense of "accomplishment" and "growth" to make the user feel proud, or using characters to build emotional attachment. Both are fantastic ways to motivate a user to stick with a product for a long time.

03. How do we melt this into our product?

Harmonizing with Alarmy's unique value

Alarmy is a service with a very clear value proposition: "reliably waking up heavy sleepers." Pasting a generic reward system onto this is like mashing two delicious foods together. (It's like mixing curry into seaweed soup...)

Furthermore, the sense of accomplishment from waking up in the morning hits differently than it does with studying or exercising. With learning and fitness, you can map out a growth roadmap: "Finish A, then move on to B." Waking up doesn't really work like that. "I woke up at 6 AM this week, so let's aim for 5 AM next week!" just doesn't resonate the same way. Plus, the idea of "waking up multiple times a day to grind for more XP" makes zero intuitive sense.

As for building attachment through characters, it seems like an easy fit for the product—if executed well. But doing it "well" is incredibly hard. Forming emotional bonds with characters requires building an entire universe and designing a life cycle, which makes the difficulty level sky-high. In reality, very few of the many services that introduce characters actually pull it off successfully.

04. Planning rewards & characters that fit Alarmy

Finding the sweet spot

Ultimately, we decided to introduce extrinsic motivation (rewards), but rather than becoming a reward-centric app-tech service, we framed them strictly as a "tool to help you wake up." We didn't want users getting rewards just for breathing; they absolutely have to complete "wake-up" related quests to earn anything. That's why we named this feature "Quests" instead of "Benefits" or "Rewards."

While it's tough to design a long-term growth roadmap for waking up, short-term challenges are definitely doable. For instance, you could run a challenge to wake up at 6 AM every day during a 2-week exam period, and offer a leaderboard based on success rates. So, we decided to add a Challenge feature within the Quest tab to tap into that intrinsic motivation as well.

After looking at various case studies for characters, we realized they don't necessarily need an ultra-complex life cycle. Depending on the nature of the product, a simple step-by-step growth system is often more than enough to build attachment and boost retention. Drawing from this, we mapped out our own simplified character growth system.

05. The result of diverging and converging

The End Picture comes together

After countless rounds of expanding and refining ideas, we finally finalized the perfect reward & character planning for our product. It’s the result of asking "Is this really the best way?" broadly and deeply over the span of 3 to 4 weeks. Testing such massive hypotheses was completely different from our usual "fast and small" approach, so it felt a bit awkward, but at the same time, incredibly thrilling.

We broke this plan down into appropriately sized MVP chunks and began our design work. Because we had such a crystal-clear vision of what things would look like post-MVP launch this time around, the meaning and impact of this work within the product felt much deeper and more solid. By the nature of zero-to-one projects, you inevitably keep asking, "Is this really the right way?" But this time, thanks to the confidence we built by repeatedly asking and answering that question while painting the End Picture, we were able to push forward with unwavering conviction.

Along with our squad, the other squads have also started wrapping up their respective End Pictures one by one.

While I don't know the exact details of everyone else's journeys, I'm absolutely certain they've established natural and seamless flows in their areas, nailed the product's consistency and balance, and baked in the planning elements required for 10x growth.

When those End Pictures start melting into the product one by one,
I can already feel how excited I am to see Alarmy take its next massive leap forward. :D

And that’s a wrap on our product End Picture sketching journey!