Designing Operational Visibility for Farm Performance
Designing a scalable visibility layer to turn execution data into reliable, decision-ready signals for HQ and operational teams.
Confidentiality Notice
Enterprise systems under NDA. Production UI & proprietary logic are intentionally omitted.
Happy to discuss details privately 🙌
Key Outcomes
0 → 100% farm performance coverage
Enabled ~1–2 weeks earlier issue detection
Overview
My Role
Led end-to-end product design
Defined what performance meant at a farm and task level
Designed data structures, indicators, and interaction patterns
Collaborated closely with PMs, engineers, agronomists, and operations stakeholders
Team
1 PM • 2 Designers • 5 Engineers
Duration
~4 months
Phased rollout alongside task system adoption
Problems
Execution data existed, but was scattered across individual tasks
No shared definition of “good” or “at-risk” farm performance
HQ lacked a fast way to identify issues without deep inspection
Decision-making relied on manual interpretation and follow-ups
Operational signals were delayed, inconsistent, or missing
The core issue wasn't data availability; it was making execution data visible, trustworthy, and actionable at scale.
Key Constraints
Data quality depended on ongoing task system adoption
Highly variable farming contexts across regions and crops
Different stakeholder needs (Operations, PMs, Management)
Visibility needed to support action, not just reporting
Any solution had to surface insights without oversimplifying reality or exposing raw, unreliable data.
Key Design Decisions & Trade-offs
Signals over raw data
Instead of showing granular tasks, we focused on synthesized indicators that reflected operational health, accepting reduced transparency in exchange for faster comprehension.
Comparative context over absolute numbers
Performance was framed relative to expectations, trends, or thresholds, rather than fixed targets, enabling more meaningful interpretation across different farms and conditions.
Actionability over completeness
The system prioritized highlighting risks and deviations over showing everything, intentionally leaving out low-value data to reduce cognitive load.
Outcomes (cont.)
The visibility layer turned execution data into a trusted, system-level view of farm performance.
By synthesizing task data into clear signals, HQ teams could understand farm health at a glance without manual reports.
This shift led to measurable operational improvements:
Established the company’s first system-driven visibility, achieving 100% farm performance coverage
Enabled proactive intervention, surfacing risks up to ~1–2 weeks earlier
Improved decision quality by reducing manual follow-ups and grounding actions in consistent execution data
