Case Study · Geospatial · SaaS Product Architecture

Greenfield Multi-Tenant Geospatial SaaS

Solutions Architect from RFP to running platform for a national subsurface-mapping and utility-locating company — multi-tenant schema, geospatial ingestion, a microservice suite on Azure AKS, and the onshore/offshore team that built it.

Role: Solutions Architect Team: onshore + offshore delivery team Scope: greenfield, RFP to running platform

01 The Challenge

A national subsurface-mapping and utility-locating company digitizes what’s under the ground — scanning and mapping buried utilities across the country. Its field data was valuable, but it wasn’t yet a product. The company wanted a customer-facing SaaS platform: an interactive map viewer, a digital plan room for file and folder management, notifications, sharing, and subscriptions.

This was a true greenfield — no existing platform to extend, no architecture to inherit. Everything from the tenancy model to the map provider to the billing integration had to be decided, defended, and turned into work a delivery team could build from. Leopard Data was engaged as Solutions Architect from the RFP onward.

02 The Approach

Leopard Data designed the multi-tenant database schema — tenancy, sharing, roles and permission types, and organization-group trees — along with the KML/KMZ/GeoJSON-to-PostgreSQL ingestion design that turns field-collected geospatial files into queryable map data. On top of that sat a microservice suite on Azure AKS: a map-object service (~30 endpoints for lines, polygons, markers, layers, features, and tracks), a digital plan room service, notifications, a sync service providing offline mode for field crews, settings, tags, and blob file management — with Azure AD B2C for identity and CI/CD built on Terraform.

Platform decisions were made on evidence, not preference. We ran the Mapbox vs Google Maps evaluation including real pricing quotes from both vendors, designed the NetSuite, Workato, and Chargebee integrations for ERP, automation, and billing/subscriptions, ran a WCAG 2.1 accessibility review, produced the TCO analysis for the client, and presented the architecture at steering committees.

Architecture only matters if a team can build it. Leopard Data ran daily onshore/offshore coordination, designed the work packages the offshore team executed, ran the hiring pipeline — dozens of architect and developer interviews — and delivered the ERDs and design documents the teams built from.

03 The Delivery

The platform went from a blank page to build-in-progress across web, mobile, and microservices. The multi-tenant schema and geospatial ingestion design were adopted as the foundation, the map-object service’s ~30 endpoints gave the web and mobile clients a common geospatial API, and the sync service design gave field crews an offline mode that survives spotty connectivity at job sites.

The commercial architecture landed alongside the technical one: the map-provider decision was made on hard vendor pricing, the NetSuite/Chargebee/Workato integration architecture connected the platform to the client’s ERP, billing, and support desk, and the TCO analysis gave leadership a defensible picture of what the platform would cost to run. The offshore team we staffed and coordinated was productive on the work packages we designed for them.

04 The Outcome

  • Architecture taken from blank page to build-in-progress across web, mobile, and microservices.
  • Multi-tenant schema and geospatial ingestion design adopted as the platform foundation.
  • Map-provider decision made on hard pricing evidence from both vendors.
  • Integration architecture designed for ERP, billing, and support desk.
  • Offshore delivery team staffed, coordinated, and productive.

05 Tech Stack

Microsoft Azure AKS / Kubernetes PostgreSQL .NET Microservices Angular Mapbox / Google Maps KML / KMZ / GeoJSON Azure AD B2C Terraform NetSuite / Chargebee / Workato Azure DevOps

Have a hard delivery problem like this one?

Senior architecture, hands-on engineering, on-site when it counts. Let’s talk.