Client
Onsite Analytics Mobile App
Overview
Onsite Analytics clients were conducting Annual and Move‑In/Move‑Out (MIMO) inspections with a first‑version internal app, then spending after‑hours time cleaning data and assembling reports. Photos weren’t tied cleanly to items/areas, and identifying asset age/warranty from serial plates was slow and inconsistent. Leadership wanted a cleaner, faster mobile experience that: - Streamlined Annual/MIMO inspections end‑to‑end on device. - Allowed default answers per unit to reduce taps. - Captured photos natively and linked them to the exact item/area. - Introduced Serial Plate Photo (SPP) capture that instantly returns model, age, warranty status, and resources; when the model wasn’t in our DB, fall back to ChatGPT for a short‑term answer. - Scaled for large portfolios (10k+ units)
Client
Onsite Analytics Mobile App
Service
Mobile App Design
SaaS Product
The Challenge
1. The first‑version app required manual re‑entry and produced inconsistent photo linkage, slowing reporting and audits. 2. Identifying asset age/warranty from serial plates was slow and unreliable. 3. Large portfolios (10k+ units at MVP, now 25k+) demanded performance, clarity, and simple flows for Annual, MIMO, and Due Diligence inspections. 4. Scope discipline was required: ship a dependable inspections core first; defer automated PDF/CSV emails and Tag/NFC to later phases.
The Solution
- Flutter rebuild + component library for consistency, speed, and accessibility. - Default answers at unit start to cut taps and cognitive load. - Native camera with item/area linkage; resilient upload states; photos stored in S3 with stable references. - Serial‑Plate Photo (SPP) pipeline: capture → parse → enrich → show model/age/warranty/resources for known models; LLM fallback when unknown, flagged for verification. - Data sync to dashboard at completion; exports handled in the reporting layer (no auto‑emailing in v1). - Scalable lists and caching tuned for large portfolios and low‑connectivity environments.
Projects
