A methodology for identifying the most limiting factor (constraint) in a system and systematically improving it. Your system is only as strong as its weakest link.
A chain is only as strong as its weakest link. Strengthening any other link doesn't increase the chain's strength. The same applies to product workflows—improving non-constraints won't improve overall throughput.
Theory of Constraints (TOC) was developed by Dr. Eliyahu Goldratt in the 1980s. The core insight: every system has at least one constraint that limits its performance. To improve the system, you must find and fix that constraint.
The system's performance is limited by a small number of constraints
Improve the constraint, not everything else. Local optimizations often harm global throughput.
Once you fix one constraint, another appears. This is progress!
Your constraint might be:
TOC provides a systematic process for managing constraints. Follow these steps in order:
Find the bottleneck—the step where work piles up or waits the longest.
Looking at the Research to Backlog flow, you notice:
Constraint identified: PM capacity to write PRDs
Get maximum output from the constraint with existing resources. Don't add resources yet—just use what you have better.
Result: PRD writing time reduced from 6 hours → 2 hours. Wait time from 7 days → 2 days.
Align all other processes to support the constraint. The constraint sets the pace.
Don't maximize efficiency everywhere. That creates inventory (WIP) before the constraint.
Key insight: It's OK if researchers have idle time. Their "efficiency" doesn't matter—only system throughput matters.
If exploiting isn't enough, add capacity to the constraint. This usually costs money or time.
After elevation: PRD creation is no longer the constraint. A new constraint emerges (maybe engineering capacity).
Once you break one constraint, another appears. Go back to step 1. This is continuous improvement.
After elevating, watch out for policy constraints—rules and habits from when the old constraint existed.
Apply the Five Focusing Steps to the Research to Backlog flow.
User research scheduling, participant recruitment
PM capacity, stakeholder alignment, PRD reviews
Senior designer reviews, design system updates
Engineering capacity, code review, testing
QA cycles, approval processes, release windows
| Method | Focus | Best For |
|---|---|---|
| Theory of Constraints | Find & fix the one biggest bottleneck | Maximizing throughput quickly |
| Value Stream Mapping | Visualize entire flow, eliminate all waste | Understanding current state, identifying multiple improvements |
| Agile/Scrum | Iterative delivery, team collaboration | Organizing team workflow and ceremonies |
| Kanban | Limit WIP, visualize flow | Managing steady flow of work |
Use Value Stream Mapping to visualize your workflow, then apply TOC to focus on the biggest constraint. Use Kanban/Agile for day-to-day execution.
Deepen your understanding with these complementary frameworks