MoFlo
•
Shipped: Dec 2025
Despite strong AI generation, content tone varied session to session. SMBs wanted consistency across platforms, but manually retyping brand instructions into every prompt created friction. I designed MoFlo’s persona system to encode brand identity into the generation flow.
Role
Product Designer
Team
Product Designer (Me)
CTO
Lead Developer
Timeline
3 Weeks
Tools
Figma (Design, make, Jam)
Claude Code
Lottielab
V0
The Problem
Brand voice wasn’t sticking and AI could generate captions and visuals, but identity drifted.
Users re-typing tone instructions every session
Different emotional tone across platforms
Manual edits to “fix voice” after generation
Confusion about who the content was actually for
Example of a user prompt:
Signal
As adoption grew, we began monitoring qualitative feedback alongside usage data.
One message stood out.
This wasn’t an isolated complaint.
We were already seeing:
High draft generation rates
Low direct publish rates
Frequent manual edits before scheduling
The message gave language to what the data hinted at.
The AI was producing content. Users were still doing the thinking.
If every draft required rewriting, the system wasn’t reducing cognitive effort. It was shifting it.
That observation triggered a deeper investigation into how SMBs were actually interacting with drafts, and why execution stalled after generation.
Speaking with Users
Speaking with more of our users what I learned:
Most didn’t think in adjectives.
They thought in:
• Who they were speaking to
• What they wanted to be known for
• How they wanted to be perceived
• What they avoided saying
What Users Said:
“I want it to sound like us, not like AI.”
“We’re premium but approachable. I don’t know how to explain that to a model.”
“Sometimes it’s too salesy. That’s not our brand.”
“It forgets who we’re talking to.”
The Initial Solution
What Didn’t Work
Despite shipping the persona builder:
Persona selection rates remained low
Users continued typing brand rules manually
Many personas were created once and never reused
Regeneration requests due to “tone mismatch” remained high
Behavioral Evidence
After launching the initial persona system, we analyzed usage patterns over 4 weeks.
38%
of users created at least one persona.
But the problems still existed,
17%
only reused a persona
60%
still included manual tone instructions
30%
Users regenerated content
50%
Persona selection was skipped entirely
After launching the initial persona system, we analyzed usage patterns over 4 weeks.
“I just type what I want.”
“I don’t remember what this persona does.”
“It’s faster to just tell it again.”
Rethinking the Model
If structured personas weren’t naturally adopted,
what would make identity feel intuitive?
If we improved the quality and usability of the creation experience, engagement should increase.
The Behavioral Shift
From Rewriting Identity to Selecting It. The goal wasn’t to eliminate prompting.
It was to eliminate repetition.
Before the redesign, users recreated their brand voice inside the prompt box every session. Tone adjustments lived in memory, not in the system.
Final System
The solution had to preserve user control, while quietly reinforcing consistency.
Instead of treating personas as a separate setup feature,
I moved identity directly into the generation layer.
Lightweight Persona Creation
The initial persona builder required users to manually define tone, audience, writing rules, and stylistic preferences.
It assumed users could articulate their brand identity upfront.
Instead of asking users to construct identity from scratch, I redesigned persona creation around base archetypes.
Active Persona Visibility
During generation, the selected persona was always visible.
Users could:
• See what rules were being applied
• Edit or refine them inline
• Understand why output looked a certain way
Identity was no longer invisible.
It became explainable.
Prompt → Persona Bridge
If a user typed additional tone rules in the prompt:
“Make it less salesy.”
“More direct.”
“Target investors.”
The system surfaced a subtle suggestion:
“Add this to your persona?”
Manual behavior became structured data.
Instead of fighting user habit,
the system absorbed it.
Impact
Within one month:
2.3×
Increase in persona reuse across sessions
– 34%
Reduction in manual tone instructions inside prompts
– 21%
Decrease in regeneration due to tone mismatch
+ 18%
Increase in direct publish rate after first draft
But the biggest change wasn’t in metrics. It was in behavior.
Users stopped rewriting their identity every session. They started selecting and refining it.
Learnings
This project reinforced a core principle:
AI features don’t fail because they lack capability.
They fail when they misalign with behavior.
Three key takeaways:
Structure Must Absorb Habit
Reduce Repetition, Not Control
Visibility Builds Trust