Automation Builder

Overview

Giving brands a workflow engine — not a workaround

Before Automation Builder, Aspire's brand users had no structured way to act on creator relationships at scale. Communication, status changes, approvals, and campaign triggers all lived in manual, one-off workflows — meaning as brands scaled to hundreds of creators per campaign, the complexity became unmanageable.

The ask: build a visual automation engine from scratch. Something that felt as intuitive as Zapier for non-technical brand managers, but powerful enough to handle Aspire's domain-specific logic — creator statuses, offer conditions, campaign triggers, and communication sequences.

The Problem

Scale breaks manual processes

Aspire's customers were managing hundreds of creator relationships per campaign. Actions that should be automatic — sending a welcome email when a creator accepts an offer, moving them to a new list when they ship content, flagging them for review if they miss a deadline — were being done manually, one by one.

• 1.Inconsistency at scale: Manual workflows meant different team members applied different standards. There was no reliable 'if X then Y' system, so campaign outcomes varied wildly.
• 2.Bottleneck in the brand team: Brand managers were spending hours per week on repetitive outreach and status management instead of strategic work. Aspire's value prop was scale — this was its ceiling.
• 3.No visibility into what was running: Even when brands set up basic automations, there was no way to see what ran, what failed, or why. Debugging was essentially impossible.

My Role

Owned the product from whiteboard to launch

I was the sole designer on this 0→1 product. That meant I wasn't handed a PRD — I authored significant parts of it. I ran discovery calls with brand users, mapped edge cases, defined the trigger taxonomy, designed every screen, prototyped the canvas interactions, and was the primary point of contact between PM and engineering throughout.

Key Design Decisions

Four decisions that shaped the product

Challenges

The hardest parts

Canvas-based UIs are deceptively complex. The interaction model — drag, connect, zoom, pan, undo/redo — has to feel effortless even though the underlying data model is a graph.

• ↳Representing grouped flows: Some automations span multiple phases. Showing hierarchy on a flat canvas without visual noise took several iterations — we landed on collapsible group containers with a distinct visual treatment.
• ↳Failure states that don't panic users: When a run fails, brand managers need to know what happened and what to do. I designed failure states with specific context: which step failed, why, and one-click retry or skip options.
• ↳Trigger taxonomy: Aspire has a rich data model — creator statuses, offer states, campaign milestones, content deadlines, payment events. Designing a trigger system that surfaces the right events without overwhelming users required multiple sessions and close engineering collaboration.

Impact

From zero to a platform's most-used feature

Automation Builder launched and became the highest-adoption feature Aspire had shipped. The numbers tracked over 18+ months told the story clearly:

The SVP of Product cited the Automation Builder directly in the annual review as one of the products driving company growth. It was also one of three features I presented to the CEO as part of the company's product roadmap review.