Sogiry

Before writing a single line of code, I conducted deep-dive interviews with e-commerce store managers, pricing analysts, and executive stakeholders. The resounding feedback was fear. They had all been burned by automated tools that aggressively dropped prices based on flawed competitor data.

I used AI to rapidly synthesize transcripts, identifying 'Fear of the Black Box' as the #1 barrier to adoption. Stakeholders wanted automation, but they demanded transparency and control over the execution.

Mapping these pain points into an affinity diagram revealed that giving users visibility into the algorithm's decisions was just as important as the algorithm itself.

My initial hypothesis was that users wanted full 'set-and-forget' automation to save time. Early wireframes and testing quickly disproved this.

Store managers felt they were losing control over their pricing strategy. I pivoted away from full automation to a 'human-in-the-loop' approach, introducing semi-automatic modes and strict approval workflows.

The core design principle for Sogiry became 'Trust through Transparency.' Automation is only usable when the stakeholder trusts it enough to turn it on.

I achieved this by ensuring every automated decision was auditable. The Executive Dashboard isn't meant to be decorative; it is a tactical triage center designed to answer three critical questions immediately: Where am I losing margin? Where am I uncompetitive? What should I do next?

Tracking thousands of SKUs daily creates a massive amount of noise. The challenge was distilling this data into a digestible format that allows store managers to instantly understand macro market movements.

I designed the Daily Market Overview to highlight extreme market shifts at a glance. Instead of digging through rows of data, managers immediately see if a specific competitor has started aggressively discounting, or if there's a sudden trend of price increases across the market. This allows them to proactively adjust their strategies rather than simply reacting to individual product changes.

Enterprise users migrating from Excel demand speed and safety. To ensure the UI matched their existing workflows, I focused on high-density ergonomics at the micro-level:

• Keyboard-First Triage: Allowing analysts to navigate data tables and approve or reject pricing suggestions via keystrokes instead of endless mouse clicks.

• 'Blast Radius' Guardrails: Bulk actions instantly calculate and display the estimated financial impact before execution, preventing catastrophic user errors.

• Accessible Data Visualization: Critical margin warnings use distinct iconography and high-contrast typography—never relying on color alone—to reduce eye strain and accommodate colorblindness.

Pros don't want 'simple'—they want 'efficient.' Every unnecessary click in a multi-channel workflow multiplies by a hundred. By delivering a 'glass-box' system, I successfully mapped and simplified complex pricing strategies.

• Reduced time spent managing pricing adjustments by 75% for early adopters.

• Achieved a 0% critical error rate on automated repricing thanks to the margin guardrail system.

• Successfully onboarded key enterprise retail partners within the first quarter of launch.

Looking back, there are two major areas where I made missteps that are currently being addressed:

1. UX Over UI: In the initial build, my focus was entirely on UX over UI. For a tool this critical, customers care far more about the engine not breaking their business than having a perfectly polished interface. I intentionally shipped functional but raw visuals to test the mechanics and break things early. Now that the core UX is validated, the entire platform is undergoing a comprehensive UI and design system overhaul.

2. The Frictionless Onboarding Trap: I initially assumed the best onboarding was the easiest one. I spent time building a complex self-serve UI allowing users to easily search, filter, and select products to track on our system. My mistake was not doing enough research on enterprise buying psychology. For a tool with this much financial impact, store owners actually want friction. They want to speak with the people behind the software, understand the mechanics, and build trust before handing over control. We abandoned this self-serve flow and pivoted to a high-touch, concierge onboarding model where we set up the store for them.