Most AI initiatives we're asked to evaluate start with a solution already in mind — "we want a chatbot" or "we want to use AI for X" — rather than a clearly defined problem. An opportunity assessment flips that order: start with where the business is actually losing time or money, then evaluate whether AI is the right tool for it.
Step 1: Map the workflow, not the department
We look at specific workflows — a support queue, an onboarding process, a document review step — rather than entire departments. Specificity is what makes the next steps measurable.
Step 2: Quantify the current cost
How many hours, how many people, how much delay does this workflow currently cost? Without a baseline, there's no way to know afterward whether an AI investment actually worked.
Step 3: Evaluate volume, structure, and error tolerance
High-volume, structurally consistent, error-tolerant workflows are strong AI candidates. Low-volume, highly variable, or high-stakes workflows usually aren't — at least not as a first project.
Step 4: Sequence, don't boil the ocean
We recommend starting with the highest-confidence, lowest-risk opportunity first, proving the approach works operationally, and expanding from there — rather than a single large program that has to succeed everywhere at once to show any value.
