The AI Bottleneck Lens
What it is
The AI Bottleneck Lens is the framework leaders use to decide where AI lifts organisational throughput. Three questions, mechanical by design. What does this organisation produce, and how is throughput measured? Not revenue; the rate at which inputs become the thing the customer pays for: matters delivered, engagements closed, patient episodes completed, features shipped. Where does a single unit of work wait the longest on its way through? Almost never the step that takes the most effort; the step with the longest queue. Would lifting this step lift the system, or shift the wait elsewhere? The constraint moves when addressed.
What it is
The AI Bottleneck Lens is the framework leaders use to decide where AI lifts organisational throughput. Three questions, mechanical by design. What does this organisation produce, and how is throughput measured? Not revenue; the rate at which inputs become the thing the customer pays for: matters delivered, engagements closed, patient episodes completed, features shipped. Where does a single unit of work wait the longest on its way through? Almost never the step that takes the most effort; the step with the longest queue. Would lifting this step lift the system, or shift the wait elsewhere? The constraint moves when addressed.
Why it happens with AI
Vendor dashboards instrument adoption (active users, sessions, time-saved estimates) because that is what licences are sold by. Throughput at the constraint requires the leader to know the actual shape of the work, define a metric, and track it monthly. Four forces compound the pull toward dashboards: adoption metrics are easy and instrumented by default; vendor incentives reward licence count, not throughput; adoption is collective and reports as a single number; constraint work is local and harder to defend upward. Loss aversion does the rest.
What working on it does, impact and benefits
A leader using the Lens stops asking where can AI help and starts asking where is the work waiting, and can AI lift the wait. AI deployments then move the metric the business is measured by: matters per associate per month, engagements closed per consultant per quarter, patients discharged per bed per week, even when the licence-count dashboard goes red because most of the workforce did not need the tool. The discipline retrains the board over three or four meetings to ask the right question by default.
Canonical framework: virenlall.com/ai-bottleneck-lens, the full ~600-word treatment of the three mechanical questions and the constraint dashboard.
“AI’s organisational impact is decided not by how many people use it, but by whether it lifts the one step that controls the flow.” — Viren Lall, Managing Director, ChangeSchool LDN (2026).