Context (About Poster Maker):

The Core Problem:

Poster Maker was evolving from a poster-only app into a content-first app.

According to data : video content consistently drove higher engagement and retention than static visuals

But this created two critical risks:

• We couldn’t scale video creation fast enough to meet the growing diversity of user needs and categories.

• Going fully video-first would slowly devalue static posters, which were still the fastest and most trusted creation format in the app.

Staying static meant falling behind.
Going fully video meant breaking speed, simplicity, and existing habits.

The challenge wasn’t choosing between posters or video

It was finding a middle path that allowed both to coexist and grow.

How the problem emerged?

• Users increasingly paused longer on video content

• Video posters had higher saves and reshares

• Static posters were still the fastest path to daily posting

• Requests for more video categories grew faster than delivery capacity

Demand for video was growing faster than we could responsibly build it - both for users and the product

Key Insights

Most Poster Maker users are small business owners.

• They’re busy running their shops, handling customers, and managing daily operations.

• Hiring a designer or video editor for regular promotions is expensive.

• Learning video creation tools is time-consuming and often unrealistic for them.

What they needed wasn’t “video creation”

They needed:

• Content that looks dynamic

• That grabs attention on social platforms

• Without extra learning, setup, or production effort

They didn’t want to make videos.
They wanted video-like impact : motion, freshness, and visibility ; with the same speed and simplicity as static posters.

Video wasn’t the goal, Relevance without complexity was.

The Tension We Had to Solve

As Poster Maker evolved into a content-first app, we faced clear tensions:

• Users wanted video content because it performed better on social platforms.

• The product was built for speed, simplicity, and repeat daily usage.

Fully shifting to video meant:

• Higher learning cost for users

• Slower creation flows

• Breaking the habit of “open → pick → share”

Staying only with static posters meant:

• Falling behind content trends

• Lower long-term engagement

• Reduced visibility on social feeds

This wasn’t a format problem. It was a trade-off between speed and richness.

Breaking motion into smaller problems

Video was inevitable, but shipping it responsibly required breaking “motion” into smaller, testable steps

Early signals showed three clear risks:

• Cognitive load would increase

• Learning curve would rise

• The core habit could break

Instead of asking users to suddenly
“Add motion to your static posters”

we stepped back and reframed the problem.

What’s the smallest step toward video that users will accept instantly without changing how they already work?

Not all motion needed to arrive at once.

This led to a deliberate decision: Start with the least disruptive layer first

Phase 1: Music as the MVP for motion

Music introduced a sense of movement without asking users to learn anything new.

• No new creation flow

• No extra decisions

• Posters stayed posters

This phase helped us validate:

• Whether users responded to richer content

• If engagement increased without slowing creation

• Whether motion-adjacent features fit daily usage

Music acted as a low-risk bridge between static posters and full video.

Phase 2: Video effects as the next step

Once we had confidence in user behavior, we introduced video effects not as video creation tools, but as an extension of posters.

• Effects layered on existing designs

• No timelines or complex editing

• Speed and familiarity preserved

We didn’t replace posters with videos.
We let posters evolve.

Why this approach mattered?

This phased rollout allowed us to:

• Move toward a content-first future responsibly

• Reduce cognitive load at every step

• Keep static posters relevant and widely used

• Scale video capabilities without breaking existing habits

Constraints That Shaped the Solution

Before deciding what to build, we had to be clear about what we couldn’t afford to break.

Phase 1: Introducing Audio Without Changing the Poster Flow

Our goal wasn’t to “add a feature”. It was to add motion without changing behavior.

Users could:

• Start with the same static poster

• Add background music

• Export and share exactly like before

No new tools to learn. No new workflows to understand.

Why we split audio into two parts?

Early feedback showed something important:

Small business owners didn’t just want engaging content. They wanted to say something.

That led to a second version.

1. Background Music

• Choose from in-app library

• Or upload custom music

• Works passively, no effort

2. My Message (Voice / Text-to-Audio)

• Text-to-speech for most users (default path)

• Voice recording kept optional (secondary tab)

• Default message provided to reduce hesitation

• Multiple voices to experiment without pressure

This respected real user psychology:

• Most people don’t like hearing their own voice

• But they do want to promote their business

So we optimized for comfort, not capability.

Intentional trade-offs we made

1. We prioritized text-to-audio over voice recording → Faster adoption, lower embarrassment barrier

2. We kept audio optional → Static posters still worked exactly as before

Once users were comfortable adding sound to static posters,
motion was no longer a foreign concept.

That confidence created space for the next step:
visual motion through video effects.

Phase 2: Introducing Video Effects

After audio adoption stabilised, we introduced video effects as the next layer of motion.

The goal was simple: Add movement without adding effort

Users could:

• Start from the same poster

• Keep the same edit flow

• Apply motion using one tap

• Export without configuring timelines, layers, or durations

No learning curve, No editing mindset.

Why presets, not timelines?

Instead of giving users control-heavy tools, we chose presets.

Reasoning:

• Business users don’t want to design motion

• They want something that “just works”

• Presets remove decision fatigue

This is why:

• Effects are previewable

• Names are simple (Spin, Fade, Tumble, Confetti)

• One primary CTA: Apply Preset

Speed > control

Design × Engineering: Building Video Effects as a Scalable System

As the scope of video effects grew, the challenge wasn’t just what motion to design but how to make it scalable and implementable.

Instead of treating each effect as a custom animation, we collaborated closely with engineering to design JSON-based motion effects presets

By collaborating with engineering early, I:

• Designed effects that were lightweight and performant

• Ensured consistent rendering across devices

• Made the system extensible for future effects

Building Presets for Scale

To move fast without locking engineering into rigid implementations, I used LottieLabs as a motion playground

Effects were designed using image placeholders, with motion applied independently of real poster content. These animations were exported as JSONs that engineering could directly adapt.

The approach was deliberate:

• Motion was designed independent of final images

• Placeholders represented dynamic poster content

• Engineering replaced placeholders with real image URLs from the backend

This separation allowed design and engineering to work in parallel, without blocking each other.

Designing Presets for Multiple Formats

Posters are shared across different platforms, so motion couldn’t be locked to a single size.
Effects were designed to scale naturally across 1:1, 4:5, and 9:16 formats, instead of creating separate animations for each.

While exporting from LottieLabs, animations were tested across formats to ensure motion stayed balanced and content never felt cropped or awkward.

This approach kept motion reusable, consistent, and future-ready as new formats were added.

How I used AI in this project

AI wasn’t the solution here. It was a supporting tool to move faster while keeping design decisions intentional.

I used AI to quickly explore motion directions and timing ideas before locking presets. This helped filter out effects that felt distracting or heavy, early in the process.

For video effects, I designed image-agnostic Lottie templates with placeholders. AI helped speed up experimentation, but the final motion logic, constraints, and structure were designed manually. Engineers later replaced placeholders with backend images, keeping the system scalable and easy to implement.

AI also helped in refining microcopy

Impact

1️⃣ Product impact (behavior-level)

• Users gradually moved from static-only posters to posters enhanced with audio and video effects

• The transition from music → video effects felt natural, not overwhelming

• Posters started behaving more like video content, while keeping the same creation flow

• Users explored effects without needing tutorials or guidance, validating the preset-based approach

2️⃣ Engagement & retention impact

• Motion-enhanced posters encouraged repeat usage, especially for daily and promotional posts

• Users spent more time previewing and experimenting before export

• Video effects helped maintain engagement while video content supply continued to scale in parallel

• This supported the broader shift toward a content-first app, without overloading users.

3️⃣ Business impact

• Video effects became a strong premium trigger, as users could clearly see the value before downloading

• The phased rollout reduced risk while still unlocking monetization opportunities

• Effects helped balance demand between static posters and full video content, protecting both workflows

4️⃣ System & scalability impact

• Motion effects evolved into a reusable, JSON-based system, not one-off animations

• New effects could be added without redesigning user flows

• Support for multiple formats (1:1, 4:5, 9:16) scaled without rework

• Design and engineering could iterate independently, improving delivery speed

Learnings

● Users don’t want “video tools”, they want video impact

People weren’t asking to create videos. They wanted motion, attention, and better reach — without changing how they already worked.

● The smallest acceptable step matters more than the boldest feature

Starting with music before video effects reduced cognitive load and helped users accept motion instantly.

● Breaking problems into layers accelerates adoption

Separating motion into audio → effects made experimentation feel safe instead of overwhelming

● Scalability is a design decision, not just a technical one

Designing effects with placeholders and format flexibility upfront prevented rework later.

● Iteration revealed the real problem, not the first solution

Music MVP validated intent., Video effects solved scale, Each phase made the next decision clearer.

● Reducing system complexity unlocks scale

Instead of custom animations per poster, preset-based JSON effects made motion faster to build, easier to maintain, and predictable to ship for both design and engineering.