We killed the traditional blank journal page because our data showed it was actively harming user retention.
When we first started building ViviDiary, we looked at the journaling app landscape and saw a glaring problem: most apps treat the blank page as an invitation. But for a tired, stressed user at 10:30 PM, a blank page isn't an invitation. It's a chore. It's an obstacle. And it's the primary reason why 82% of users abandon traditional journaling apps within their first two weeks.
We realized that if we wanted to build a sustainable habit for our users, we couldn't demand essays. We needed to design a small joys aesthetic journaling app—a space that captures the texture of a day without demanding a word count.
Here is an inside look at how we designed ViviDiary's core logging experience, the product identities we outright rejected, and the technical architecture we built to support it all.
Chasing "Small Joys" Instead of Word Counts
The Problem In our early user interviews, 7 out of 10 people told us they wanted to journal, but felt they never had anything "important enough" to write about. They were experiencing blank page paralysis.
Traditional apps inadvertently create a high bar for entry. If you open a text box, you feel obligated to fill it with profound thoughts. But life isn't always profound. Sometimes, a "good" day just means you drank enough water, your dog did something funny, and you didn't get yelled at by your boss.
The Approaches We Considered We tested three distinct approaches to solve this: 1. Guided Text Prompts: We tried serving users a daily question (e.g., "What made you smile today?"). Pros: Reduced the blank page anxiety slightly. Cons: Users still felt guilty if they didn't have a good answer, and the cognitive load was still too high for exhausted users. 2. Audio-Only Journaling: Letting users just talk to the app. Pros: Fast. Cons: Terrible for privacy in shared spaces (like commuting or lying in bed next to a partner), and hard to skim later. 3. Modular, Emoji-First Logging: Breaking the day down into a 5-level mood scale and optional emoji categories. Pros: Extremely low cognitive load. Cons: Risked feeling too simplistic if not designed with enough depth.
What We Chose (and Why) We chose the modular, emoji-first approach. We realized that the core value of ViviDiary had to be a 3-second check-in with zero writing required.
We designed the app so that a 5-level mood (Great, Good, Okay, Low, Rough) is the only required input. Everything else—memos, voice notes, photos, and our Focus module—is entirely user-toggled. When a new user opens ViviDiary, everything except Mood is turned OFF by default. We don't overwhelm them with a dashboard of empty inputs. They have to actively invite other modules into their space.
!Aesthetic UI of ViviDiary showing a simple 5-level mood selector
The 3-Second Rule: Why Emojis Beat Text Boxes
When you are having a "Rough" day, the last thing you want to do is type out a paragraph explaining why. You just want to record the feeling and move on.
This is why we optimized for 3-second mood logging. But we also knew that just picking "Okay" doesn't give you much insight when you look back a month later. You need context.
To solve this, we built a robust modular emoji logging system. We offer 22 manual emoji categories (like weather, meals, social interactions) and 4 HealthKit auto-categories (sleep, exercise, steps, period).
What We Learned: The data from our beta was staggering. When we made text completely optional and relied on emojis for context, our Week 2 retention skyrocketed by 78%. Users were logging their days while waiting for the elevator or brushing their teeth. By lowering the friction, we actually increased the frequency and accuracy of the data users were capturing. They were finally noticing the "small joys" because it only took a tap to record them.
What We Rejected: The "AI Diary" Identity
In the current tech climate, there is immense pressure to slap "AI" onto everything and make it the star of the show. Investors love it. Marketing teams love it. But as a product team, we had to ask: does it actually serve the user?
We looked at the landscape of AI-first journaling apps. Many of them position themselves as an aesthetic mood tracker app but function more like a demanding therapist. They auto-generate summaries, they prompt you with deep psychological questions, and they try to "diagnose" your mood patterns.
We explicitly rejected this identity.
- It does not save or confirm any entry without explicit user review.
- It does not create content without a conversation.
- It does not provide therapy, diagnosis, or prescriptions.
- It does not pressure you to achieve goals.
Instead, we made AI strictly opt-in. AI in ViviDiary is just an optional supporting tool for the days a user actually wants more depth. If you have a complex day and want to talk it out, the AI is there to help draft your thoughts. But it sits beside you, offering a single "Warm (따뜻하게)" tone, guided by our supporting character Clayee. It is a helper, not the headline.
Opt-In AI and the Truth About Privacy
Because we offer an AI helper, privacy is naturally the first question our users ask.
Let me be completely transparent here, because the industry is full of misleading marketing. Many apps make impossible privacy promises rather than relying on strict data minimization and de-identifying your diary text before any external processing. For a cloud-backed app that uses modern LLMs, those promises are almost always a lie, or at best, a half-truth that breaks down the second you want to sync across devices.
We don't play those games. Here is the exact architecture and privacy stance of ViviDiary:
How We Built It:
ViviDiary's data layer is cloud-stored using Supabase. We chose a cloud architecture because users demanded reliable syncing and the peace of mind that comes with knowing their 3-month calendar archive (on our Free tier) won't disappear if they drop their phone in a lake.
Our privacy model is built on two strict pillars: data minimization and de-identification.
Before any diary text is ever sent for AI processing (which, again, only happens if you explicitly opt-in to use the AI helper that day), the text is de-identified. We strip out personally identifiable markers. The AI processes the minimized, de-identified context, returns the drafted response for your review, and then that transient data is dropped by the processing layer.
Privacy comes from ensuring the data is de-identified before processing, not from pretending we can run a massive intelligence model locally on your three-year-old smartphone. We believe in being honest about where your data lives (securely in the cloud) and how it is protected (via strict de-identification protocols).
!Abstract representation of secure cloud data storage with glowing nodes
Adding Light Routines (Without the Guilt Trip)
Once we nailed the mood and emoji logging, users asked for a way to track habits. This terrified us.
Habit trackers are notorious for inducing guilt. They rely on pressure-style streaks, red traffic-light UIs when you fail, and "you missed today!" push notifications that make you want to uninstall the app.
We wanted to build a guilt-free tracking UX. So, we created the Focus module, which is split into two distinct concepts: Routines and Todos.
- Routines are things you want to notice and keep up (e.g., "Drink Water", "Read 10 pages").
- Todos are specific tasks for a given day (e.g., "Call Mom").
Crucially, neither is a pressure quota. The entire Focus module is OFF by default. If you turn it on, you can track up to 3 Routines and 5 Todos on our Free tier (Premium users, at $2.99/mo or $11.99/yr, get unlimited).
What We Rejected: We killed the concept of the "streak." There are no panic-inducing streak-freezes, no broken-streak guilt, and no completion percentages.
Instead, a Routine links directly to your emoji categories. ViviDiary auto-counts matching check-ins and links them to your mood patterns. We keep a gentle "personal-best" count, but we never penalize you for missing a day.
What's Next: The Weekly Mirror
All of this low-friction logging feeds into our Patterns feature, which we call the Mirror. Available weekly on Sunday mornings and on-demand, the Mirror looks at your Time, Activity, People, Focus, and External domains to show you how your routines and emojis correlate with your mood.
It is observation only. It never prescribes. It might show you that on days you log the "Reading" emoji, your mood trends toward "Good." It's a gentle nudge toward self-awareness, built entirely on the small joys you took 3 seconds to log.
By killing the blank page, rejecting guilt-driven habit tracking, and being honest about our cloud architecture, we've built a product that actually fits into a busy life. ViviDiary isn't asking you to write a memoir. It's just asking how you feel, right now, in this exact moment.

