SaaS Company — Taiwan
AI Integration · 6 weeks · Cross-functional (marketing, product, support)
Situation
The company had deployed AI tools across marketing, product, and support but usage was inconsistent — some team members used AI daily with no quality standard, others avoided it entirely. Revision cycles on cross-functional deliverables were running three to four rounds.
Approach
Ran the Workflow Map stage across three functions to identify where AI was adding value and where it was creating rework. Redesigned the integration around the workflows that mattered most — drafting, review, and cross-functional handoffs. Built role-based prompts and quality checkpoints into each workflow.
Outcome
Revision loops dropped by half within six weeks. Teams moved from inconsistent individual experimentation to a shared standard for AI-assisted drafting with clear quality gates.
Durability
At the three-month check-in, the workflows and quality checkpoints were still in active use without external support.
"We thought the problem was that people weren't using AI. The diagnostic showed the problem was that everyone was using it differently."
— Operations Lead
Takeaway Inconsistent AI use across teams is a workflow design problem, not a training problem.