Terms and definitions
Most of the planning vocabulary used in this article was set out by Mike Cohn in Agile Estimating and Planning (2005), which remains the most complete treatment of the subject. Other terms are credited where they apply.
| Term | Definition |
|---|---|
| Baseline | A plan approved at the start of a project, against which deviations are measured and corrected. |
| Bounded context | In domain-driven design, a part of a system with its own model and vocabulary, owned by one team (Eric Evans). |
| Burn-up chart | A chart of work completed over time, drawn against a separate line for total scope. |
| Cone of uncertainty | The observation that estimates are very uncertain early in a project and become more precise as work progresses (Barry Boehm; popularised by Steve McConnell). |
| Definition of Done | The quality standard every increment must meet, shared by the whole team and not negotiable from sprint to sprint. Part of Scrum. |
| Definition of Ready | A checklist some teams use to decide when a story can enter sprint planning. Not part of Scrum; useful as a checklist, harmful as a gate. |
| Domain event | A published fact about something that has happened in one area, to which other areas can react (Eric Evans). |
| Enabler | Backlog work that builds a lasting technical capability, such as infrastructure or architecture, rather than a user-visible feature. |
| Feature buffer | Capacity filled with work that can be dropped, so that a fixed date holds when progress is slower than expected. |
| Forecast | The current best estimate of what will be delivered and when, updated as knowledge improves. |
| Hardening sprint | A sprint devoted to stabilising the product before a release. Usually the sign of a Definition of Done that was not respected. |
| Ideal day | A unit of estimation: the work one person could do in a day without interruptions. |
| Kano model | A classification of features by their effect on satisfaction: basic expectations, performance features and delighters (Noriaki Kano). |
| MoSCoW | A prioritisation scheme from DSDM: must have, should have, could have, won't have this time. |
| Parking-lot chart | A one-page summary of a release: one box per theme or area, with its percentage complete and a status colour (Mike Cohn). |
| Planning horizons (planning onion) | The nested levels at which planning happens: strategy, portfolio, product, release, iteration and day (Mike Cohn). |
| Planning Poker | An estimation technique in which estimators reveal their estimates at the same time and discuss the differences. |
| Refinement | The ongoing work of clarifying, splitting and sizing backlog items ahead of sprint planning, so they are ready to be selected. |
| Relative sizing | Estimating the size of work by comparing items with one another rather than in units of time. |
| Relative weighting | A prioritisation technique that scores each feature on the benefit of having it and the penalty of not having it, divided by its cost. |
| Release | A version of the product that users receive and can adopt, as distinct from a deployment. |
| Release burn-down bar chart | A release chart in which completed work lowers the top of each sprint's bar and scope changes move its bottom, so progress and scope change are shown apart (Mike Cohn). |
| Release planning | Forecasting which capabilities a release will contain and roughly when it will be available. |
| Schedule buffer | A single, visible time margin added to a release, sized from the difference between likely and safe estimates. |
| Spike | A short, timeboxed piece of work whose purpose is to reduce uncertainty; its output is knowledge. |
| Sprint Goal | The single objective of a sprint, which gives the team focus and the Daily Scrum something to inspect. |
| Sprint commitment | In the current Scrum Guide, the team commits to the Sprint Goal; the selected stories are a forecast of how to reach it. |
| Story point | An abstract unit of size combining effort, complexity and uncertainty. |
| Velocity | The amount of work, usually in story points, a team completes per sprint, measured rather than promised. |
| Walking skeleton | A minimal end-to-end implementation that connects all the main parts of a system's architecture (Alistair Cockburn). |