Commercial Quoting in Minutes, Not Hours

CHALLENGE

XOi provides asset intelligence technology for field service teams. Technicians capture detailed HVAC equipment data including make, model, condition, and service history. That data supports two internal workflows: sales teams build detailed quotes by hand in spreadsheets, often taking up to 15 hours each, while planning teams use separate tools and processes to turn asset data into capital forecasts.

XOi wanted to launch a new platform, Advisor, to bring those workflows together on one shared foundation.

Two technical challenges shaped the work:

Quoting required a configurable calculation engine. The solution needed to consistently translate equipment attributes, configured services, labor rates, material costs and markups, service needs, and adjustments into a final price. The calculation logic also needed to handle incomplete equipment data and project-level overrides without producing inconsistent results.
The architecture needed to support complex calculations across large equipment portfolios. XOi's evolving equipment dataset introduced many combinations of equipment, services, sites, pricing configurations, and data completeness. Advisor needed to preserve a common equipment foundation while efficiently applying workflow-specific calculations across thousands of records.

The goal was to build a shared equipment foundation that could support two distinct decision systems: a configurable engine for near-term quoting and a lifecycle-based model for long-term capital planning.

Implementation Details

Developing an Ambitious Plan

Six Feet Up worked closely with XOi's team to refine initial designs into an implementation plan. Prioritized epics were broken into user stories tied to milestones, with functionality, acceptance criteria, and dependencies clarified before development.

That refinement process was especially important because Advisor was set to launch on an ambitious timeline. Surfacing scenarios early helped keep the timeline realistic without compromising the data model.

Designing Configurable Pricing

Advisor needed one data foundation with configurable logic layered on top. Pricing logic became the core architectural challenge.

Early modeling explored direct mappings between equipment characteristics and pricing recommendations. As more scenarios emerged, it became clear that a more flexible structure would better support the range of equipment and service combinations.

The team moved to a modular pricing model. Rather than one rigid formula, quotes were built from configurable components such as parts, labor, and services that could be combined differently depending on the scenario. That allowed new cases to be handled through configuration rather than requiring new application logic each time.

Building on Reliable Data

Both quoting and capital planning depended on a well-structured equipment foundation. Advisor was designed so the two workflows could share enriched equipment records, site metadata, permissions, and project infrastructure while applying different business logic to that data.

As new scenarios emerged, the model needed clear validation and a schema that could evolve with the dataset. The service layer used FastAPI with SQLAlchemy models, Pydantic validation, and Alembic for schema changes as the model developed.

That maintainable foundation also supported a light-lift handoff. Six Feet Up worked closely with XOi's engineering team and documented the codebase throughout development, supporting the transition to the team that now owns and extends Advisor.

RESULTS

Advisor significantly reduced the time required to produce a quote while establishing a shared foundation for pricing and capital planning.

Quoting now takes minutes instead of hours. Reusable pricing components and structured data cut quote preparation from up to 15 hours of manual spreadsheet work to minutes.
Pricing adapts across asset types. New scenarios can be supported through configuration instead of custom development.
One equipment foundation supports two decision models. Quoting applies configurable service and pricing logic to enriched equipment data, while capital planning uses lifecycle and condition data to model long-term replacement needs.

The broader lesson: when multiple workflows depend on the same domain data, a shared model with configurable logic can create consistency without limiting flexibility.

Working with complex operational data? See how Six Feet Up turns real-world data into infrastructure teams can build on: https://sixfeetup.com/big-data.