Understanding the challenge
The firm's people were sold on AI in theory and stuck on it in practice. Software vendors pitched AI for design, for proposals, for project management, and everyone had an opinion, but no one could say which of it was worth the disruption to a firm that runs on billable hours and hard deadlines. "We did not want to gamble a project team's time on a tool that might not pan out," the firm's principal said. The partners worried about client confidentiality and about tools that did not fit their design and project systems, so the firm kept experimenting in pockets and never committed, spending on subscriptions that went unused.
Our approach
The firm did not need a build. It needed clarity on where AI would actually pay off before committing a project team to it. We ran our AI Strategy engagement on the principle the firm shared: value before hype, and strategy before software:
• Use-case discovery across design production, proposals and bids, project management, and back-office operations, surfacing where the real time went and where AI could help.
• Opportunity prioritization that ranked every use case by value and effort, so the firm could see what to pilot first and what to leave alone.
• A roadmap that named the right tool for each opportunity, including where workflow automation would remove repetitive project admin and where AI agents could speed research and first-draft proposals, all inside a governance framework built for client confidentiality.
• Clear next actions and quick wins the firm could start on immediately, matched to its actual design and project systems.
Technical innovation
The core of the work was a prioritization model that scored each opportunity on value and feasibility together, weighing the time a task consumed against the data readiness, integration effort, and confidentiality risk of automating it. Because we run more than fifty operational AI tools inside our own agency, the recommendations came from what works in production rather than vendor marketing, and every high-value use case was pressure-tested against the firm's client-confidentiality obligations before it made the roadmap. The result was not a wish list, it was a sequenced plan the firm could act on with confidence.
Outcome
In a matter of weeks, a year of scattered experiments became one prioritized plan: 14 high-value AI use cases ranked by value and effort, the quick wins the firm could start immediately, and a clear line around the ideas that were not worth the confidentiality risk yet. For the first time the partners knew exactly where AI belonged in the firm and where it did not, and they could commit a project team without gambling. "We stopped guessing and finally have a plan we trust," the principal said.
