Kanban best practices are the foundation of effective flow-based project delivery. Originating from Toyota’s production system in the 1950s and adapted for knowledge work by David Anderson in the 2000s, Kanban has become one of the most widely adopted Agile frameworks globally — particularly for operations teams, support functions, and any project environment where work arrives continuously and unpredictably rather than in planned batches. Unlike Scrum, which prescribes specific roles, ceremonies, and sprint cadences, Kanban provides a minimal set of core practices that can be applied to existing processes without requiring a structural transformation. This guide covers the Kanban best practices that consistently produce the most significant improvements in flow, throughput, and predictability.
The Six Core Kanban Practices
David Anderson’s Kanban Method identifies six core practices that, applied together, transform a chaotic or inefficient workflow into a managed, measurable, continuously improving system. These practices build on each other — each one enhances the value of the others.
1. Visualise the Workflow
The Kanban board is the foundation of everything else. Before you can manage a workflow, you must make it visible. A Kanban board maps the team’s actual workflow as columns — each column representing a stage through which work items pass from request to completion. The board reveals what work exists, where each item is in the process, and how much work is accumulating in each stage. This visibility is itself transformative: teams that make their workflow visible consistently discover queue buildups, bottlenecks, and work distributions that were invisible — and therefore unmanageable — before the board existed.
Effective Kanban board design starts with mapping the actual workflow as it currently operates, not as it theoretically should operate. Discovering that “review” is actually three separate stages — technical review, security review, and business acceptance — is a valuable insight that the board design process surfaces.
2. Limit Work in Progress (WIP)
WIP limits are Kanban’s most powerful and most counterintuitive practice. A WIP limit is a maximum number of work items allowed in a specific workflow column at any time. When a column reaches its WIP limit, no new items can enter that column until an existing item exits. This constraint forces the team to focus on finishing work in progress before starting new work — changing the default behaviour from “always be starting something” to “always be finishing something.”
The impact of WIP limits on flow is explained by Little’s Law: Cycle Time = WIP ÷ Throughput. When WIP is reduced, cycle time decreases proportionally (assuming throughput remains stable). Teams that implement WIP limits consistently report significant reductions in average cycle time — often 40–60% within weeks of implementation. The most common WIP limit starting point is n+1 (one more than the number of people in the team), revised through experimentation based on observed flow data.
3. Manage Flow
Managing flow means actively monitoring and improving the movement of work through the system. Flow management involves: tracking cycle time for each work item and monitoring trends, identifying where items are blocking or accumulating (bottlenecks), using flow metrics — throughput, cycle time, WIP — as the primary performance indicators, and taking daily action to remove blockers and restore flow when it stalls. The daily Kanban standup (“walk the board” from right to left, focusing on items closest to completion) is the primary flow management ceremony.
4. Make Policies Explicit
Explicit policies are the agreements that govern how the team operates — what criteria must be met for an item to enter each column (entry criteria), what conditions must be satisfied for an item to leave each column (exit criteria), how different types of work are prioritised, what the team’s WIP limits are and why, and how expedite items are handled. Making these policies explicit transforms implicit, individually held rules into shared team agreements that are visible on the board, consistently applied, and open to collaborative revision as the team evolves.
5. Implement Feedback Loops
Kanban’s feedback loop cadences provide regular opportunities to inspect the system and adapt. The key cadences are: the daily standup (operational feedback — what is blocking today?), the weekly replenishment meeting (demand management — what new work enters the system?), the monthly review (service delivery — are we meeting our commitments?), and the quarterly operations review (strategic improvement — what systemic changes should we make?). Each cadence addresses a different time horizon and different questions about system health.
6. Improve Collaboratively and Evolve Experimentally
The sixth practice is the cultural and scientific foundation of Kanban: use data from the system to form improvement hypotheses, test them at small scale, measure results, and standardise successful changes. This is the Kaizen spirit applied to Kanban — continuous, evidence-based improvement rather than periodic restructuring driven by opinion or management mandate.
“Stop starting, start finishing. WIP limits are the single most impactful Kanban practice because they force the behavioural change — focus over multitasking — that produces all the other flow improvements.” — David Anderson, Kanban: Successful Evolutionary Change
Kanban vs Scrum: Choosing the Right Framework
Kanban and Scrum are both Agile frameworks but optimise for different contexts. Scrum works best when work can be meaningfully batched into time-boxed iterations, when the team can commit to a sprint goal, and when regular cadences of planning, review, and retrospective add value. Kanban works best when work is continuous and unpredictable (support, operations, maintenance), when items vary widely in size and complexity making sprint commitments unreliable, and when the team needs to start delivering improvements without a structural transformation. Scrumban — a hybrid combining Scrum’s cadences with Kanban’s WIP limits and flow focus — has become popular for teams that want elements of both.
Kanban Metrics Reference
| Metric | Definition | How to Improve |
|---|---|---|
| Cycle Time | Work started to work done | Reduce WIP; remove blockers faster |
| Throughput | Items completed per time period | Reduce cycle time; remove bottlenecks |
| WIP Level | Items currently in active stages | Enforce WIP limits; finish before starting |
| Lead Time | Request received to delivery | Reduce queue wait time; improve WIP |
| Flow Efficiency | Active time ÷ total cycle time | Reduce wait times between stages |
Key Takeaways
- The six core Kanban practices — visualise, limit WIP, manage flow, make policies explicit, implement feedback loops, and improve experimentally — build on each other and must be applied together for maximum impact.
- WIP limits are Kanban’s most powerful practice — they change team behaviour from “always be starting” to “always be finishing,” reducing cycle time through Little’s Law.
- Walking the board right to left in the daily standup focuses discussion on items closest to completion — the highest-value improvement any PM can make to a standard Kanban standup.
- Explicit policies transform implicit individual rules into shared team agreements that are consistently applied and openly revisable as the team learns.
- Kanban’s feedback cadences — daily standup, weekly replenishment, monthly review, quarterly operations — address different time horizons and different aspects of system health.
- Flow efficiency (active time ÷ cycle time) is one of the most revealing Kanban metrics — most teams discover that items spend 80–95% of their time waiting, not being actively worked on.
