How AI Is Transforming Everyday Expense Management Operations

Over the past year, I've been integrating AI into different parts of my mobility expense

management workflow. Not through massive projects, but through small, practical steps that make daily work faster, cleaner, and more scalable. One of the biggest wins has been using AI APIs to automate tasks that used to require a lot of manual review.

Automating Rate Plan and Device Classification

I first used AI to identify rate plan allowances by analyzing plan descriptions. Normally, this means opening each line, reading through the text, and manually figuring out the allowance. With AI, I can automate that entire review. The API reads the plan description, pulls out the allowance, and returns the information instantly.

I also built another AI-driven function to identify device types directly from plan descriptions. Typically, device type is available in vendor inventory reports but that requires downloading an additional file, searching for matches, and reconciling everything line by line. AI allowed me to skip that extra report entirely.

Why Device Type Matters

Device type isn't just a "nice to have." When we run rate plan optimizations, we need to know whether a device is a phone, tablet, router, or wearable because different device categories may not be eligible for the same plans. Misalignment causes errors, incorrect recommendations, and unnecessary back-and-forth with the client.

Since billed optimization tools don't include device type in their output, I built an AI summarization step that takes the optimization results as input and generates:

• A clean breakout by device category

• A summarized view of plans and usage

• A more understandable explanation of the recommended changes

This gave us a clearer picture of what we're actually optimizing and why.

Real Impact on Client Communication

This summarization made a huge difference. It helped us quickly understand the patterns in the data and explain plan changes in a way that clients could immediately understand. Instead of just showing a list of recommendations, we could clearly show:

• Which devices were on the wrong plan

• Why certain changes mattered

• How much the client would save

• How the plan structure aligned with device usage

That clarity directly translated into higher adoption rates. When clients understand the "why," they move forward with confidence.

The Bigger Lesson: AI Doesn't Need to Be a Big Project

The bottom line is that AI can support operations in ways that are surprisingly simple but extremely effective. You don't need a huge AI transformation program. You don't need months of planning or a dedicated engineering team.

Incremental AI adoption, one workflow at a time, is often the most powerful approach.

These AI tools now save me hours per week, reduce manual errors, help me scale my work, and improve the quality and clarity of the insights we deliver. It's a perfect example of how AI can quietly, consistently elevate operations without disrupting the business.