Projects / Monstrum

UX / UI · Product Design

In Progress

Monstrum AI Growth Engine

AI-powered growth engine that helps businesses become visible in AI search results like ChatGPT, Perplexity, and Google AI Overviews.

Role

Product Designer
Sole designer

Tools

Figma · Claude · Notion

Platform

Web

Company

monstrum.digital
Monstrum dashboard preview

The Product

What Is Monstrum?

Monstrum is a design agency building an AI-powered growth engine that helps businesses become visible in AI search results like ChatGPT, Perplexity, and Google AI Overviews. It runs three layers simultaneously: SEO for Google rankings, AEO to get cited in AI-generated answers, and GEO to become the default source AI engines pull from. Twelve agents run a continuous pipeline covering research, content writing, optimization, publishing, and performance monitoring. The system compounds over time because performance data automatically feeds back into strategy without any manual work from the client.

Design Challenge

Making a 12-Agent Pipeline Feel Simple

The core tension was making a 12 agent technical pipeline feel simple and trustworthy for a marketing manager who just wants to know if their brand is showing up.

I solved this in two ways. First, I abstracted the pipeline into 9 human-readable steps with three clear states: complete, active, and locked. Second, I replaced all agent references in the UI with plain language activity labels like Writing, Awaiting your review, and Goes live Mon.

Key Positioning

Designing Without a Reference Product

Monstrum's AI powered platform operates in a space with no established interface pattern to design against. Profound, Peec, and Gushwork each solve one piece of the problem, but none of them run SEO, AEO, and GEO as a single compounding client facing pipeline. There was no existing dashboard to benchmark against for the exact problem I was solving.

That meant the information architecture, the core metric, and the interaction model all had to be defined from scratch. I couldn't adapt an existing UI pattern, so I had to work from the pipeline logic and the business model directly, and build the product's vision alongside its interface. Every structural decision in this case study, from the 12 to 9 step abstraction to the AI Citation Score, came out of that starting position.

Research

Competitor Analysis

I spent two weeks on competitive analysis before touching any screens. The three main competitors were Profound, Peec, and Gushwork.

Profound competitor analysis

Profound

G2 Winter 2026 AEO Category Leader

Profound is the G2 Winter 2026 AEO category leader with 140+ verified reviews. Strong on monitoring but the execution loop is incomplete.

Peec competitor analysis

Peec

Focused Monitoring, 2,000+ Teams

Peec is the benchmark for focused monitoring with over 2,000 teams and self-serve pricing. The structural limit is that it stops at insight and never executes. Every Peec client still needs a separate content team.

Gushwork competitor analysis

Gushwork

Content at Volume, No Measurement

Gushwork executes content at volume but has no measurement layer and no feedback loop.

The Gap I Found

No competitor runs SEO, AEO, and GEO in one compounding pipeline. That gap is Monstrum's core differentiator and it shaped every design decision I made. Three things follow from that gap that no competitor is building: GEO as a named, executed discipline; a compounding feedback loop where performance data updates strategy automatically; and an agency architecture with no client caps.

UX Flows

Four Flows, Mapped End-to-End

Main Dashboard Content Review & Approval Content Calendar View Performance Drill-Down
View in FigJam
UX Flow — Content Calendar View and Performance Drill-Down UX Flow — Main Dashboard and Content Review and Approval

Iteration

Feedback Applied

The wireframe went through a significant review process with the founder. The feedback was detailed, specific, and required multiple rounds of iteration.

Key Screens

11 Screens, 5 Surface Areas

High-fidelity design in progress — screens below show the interaction model and content hierarchy.

Onboarding

Asks for one thing: a URL. The crawler runs automatically, extracts brand information, and presents findings as editable pill tags for the client to confirm or adjust. Three screens total.

Onboarding — 3-screen flow

Dashboard

Answers three questions in under 10 seconds: Am I showing up? What changed? What do I do next?

Dashboard

Agents

Shows the full 9 step pipeline with step status, scores, and recent activity for each agent.

Agents — 9-step pipeline view

Content

Shows all published and in-progress pieces with tab filters, status badges, SEO scores per row, and Review and Edit actions.

Content — content management surface

Brand Memory

Two states: a read only view showing everything the system knows about the brand, and an edit mode where every field becomes interactive.

Brand Memory — read and edit states

AI Process

Early wireframes were sped up using Claude, fed structured design brief files rather than freeform prompts, turning sketches into starting point layouts faster. Every subsequent decision, structure, hierarchy, content, and iteration based on feedback, was mine.

In Progress

What's Next

High-fidelity design is currently in progress. Remaining work: finalize the hi-fi visual system across all 11 screens, usability testing, and accessibility review.

Visit monstrum.digital
Back to Projects Next: StageUp Mobile App