The "Argh" Moment, Not the "Aha" Moment

The Friction Points Hidden Behind the Data

The "aha moment" refers to that exact instant a new user first feels the value of a product. Because this heavily impacts retention, getting new users to experience that "aha" as quickly as possible is often the core of the new user onboarding journey. However, last quarter, our team realized that our onboarding journey had an "argh" moment instead of an "aha" moment. After conducting several user observation sessions, we realized that before we could even unearth our aha moment, we desperately needed to eliminate the "argh" moments.

An "Argh" Moment...?

An "argh" moment is a roadblock where new users, while using the product, let out a sigh of "argh" instead of an exclamation of "aha." It means something isn't smooth and friction has occurred. Common examples include moments of cognitive overload from an excessive dump of information, cognitive dissonance due to an inconsistent information architecture and hierarchy, or feelings of annoyance and unpleasantness triggered by an unfriendly UI/UX. Of course, there's no way we created these moments on purpose. These are cases where we tried our absolute best to impress and amaze, but unintentionally ended up causing frustration and sighs. How are your products doing right now? This is something you can easily check for yourself.

The Spontaneous Observation Camera

One day, our Head of Engineering, Jason, was having dinner with an industry acquaintance and asked them to install and try out Alarmy. With their consent, he recorded the entire journey and shared it with the team. What we saw in that recording was pretty embarrassing. The user kept getting stuck in places where they absolutely shouldn't have. They repeatedly tried to swipe on screens that didn't support swipe gestures, and after the onboarding was entirely finished, they couldn't even tell if they were actually done, feeling unnecessary confusion and anxiety. If someone working in the tech industry couldn't properly understand our journey, just imagine how difficult it must be for everyday users.

Prompted by these somewhat shocking results, this "observation camera" activity turned into a bit of a relay race for a while. While the specific details varied slightly, the overall content of the recordings was consistent. Users were finishing onboarding without really understanding what value our app provides or how to use it, and pop-up purchase screens along the way made them mistakenly think it was a paid-only app.

How did our hard work end up leading to "argh" moments instead of "aha" moments?

I think the root cause was trusting the numbers a little too much.

The "Argh" Moments Data Couldn't Show Us

Numbers tell us how many users dropped off, but they don't tell us why they left. By the same token, they tell us how many users stuck around, but they don't tell us why they stayed.

The completion rate of onboarding might not have differed depending on how early the purchase screen was shown, but the reasons users left without completing it could have been totally different. It could have been "I looked around and don't think this will solve my problem," or it could have been "Seeing this purchase screen, I guess this is a paid app." Even if the next-day app return rate increased, the reasons why users returned more could have changed. It might have been "I tried it once and wanted to use it more," or it might simply have been "I unintentionally had an alarm set."

Among our purchase screen entry points, there was one particular point that had an oddly high payment ratio. It was an entry point acting as a placeholder for about 1–2 seconds before an ad loaded in the ad space. What was unusual was that this entry point also had a high refund rate. As it turned out, users who were half-asleep right after waking up were frequently misclicking, which brought up the purchase screen, and then misclicking again to accidentally complete the payment—this, too, was something we only found out by actually listening to our users' voices.

Like this, even if the immediate data shows no difference or even points to better numbers, qualitatively, those numbers might not mean something good. They could be numbers that only look good in a very brief snapshot of the journey—born out of users being "unwittingly tricked," "having no other choice," or doing something "by mistake."

Winner + Winner != Winner

People often call this a Frankenstein product. If you don't look at a product macroscopically and only optimize specific areas piecemeal, the individual parts might be optimal, but when you put them all together, it often turns into a weird product. Simply put, it gets ugly. After going through a long period of localized growth under the guise of Agile, our product seemed to have fallen into that same trap of local optimization. We took another step back to look at our most important journeys: the new user onboarding journey and the alarm set/dismiss journey. The culmination of the "winners" from our past experiments didn't look like a winner at all when stitched together. We started identifying the poorly constructed journeys caused by local optimization, defined the problems, and began resolving them one by one.

Now Aiming to Eliminate the "Argh" Moments

It's a bit tough to set this as a goal since it's a rather qualitative area, but nevertheless, we decided to sprint toward the goal of eliminating "argh" moments. When we meet friends offline, we ask them to install Alarmy for tomorrow's wake-up, and if we observe the process and they don't get stuck, it's a success. We could also try to force-quantify it, like "if we test 10 people and 9 of them don't get stuck, it's a success." Or we could create a questionnaire to measure their understanding, declaring success if they pass a certain score. The important thing here is that we are obsessing over the actual usability (use cases) of our users that exists behind the data. Perhaps the very process of eliminating these "argh" moments will turn out to be one of the best strategies for getting users to experience the "aha" moment quickly.

In our observation cameras at the end of this year, I truly hope we see exclamations of amazement rather than sighs of frustration. :)