Sprint is a feedback loop. Sprint in agile is normally 1 week to 1 month. The cadence should be constant from Sprint to Sprint for a given team. However, the cadence of a Sprint might vary from team to team.
The shorter the Sprint is, the better – Sprints are to exert pressure on teams to innovate and deliver quicker. Do not overthink the length of your Sprint; choose 1, 2, 3- or 4-week cadence. Then adapt the Sprint duration, once you become familiar with the project or product.
For those who like to act deliberately, here is a detailed guide on how long a sprint should last. A “good” Sprint is Long enough to produce results but short enough to limit risk.2
Why 1 to 2 weeks?
Although scrum.org claims “one month or less”, there is a growing consensus that shorter Sprints are better:
- The shorter the Sprint is, the faster feedback arrives from clients and the market. Also, the gut feelings of team members surface quicker.
- Sprints were designed short to break down silos and deal with turf wars. How? Short sprints promote cross-functional teams with no outer dependencies. Otherwise, the teams wouldn’t be efficient enough to get the work done within 7 to 14 days.
- Newly forming teams get to their full performance within a certain number of Sprints, be it 1- or 4-week Sprints. Actually, longer Sprints even increase the number of Sprints teams need to reach full potential.
- When story points are used rather than time units to estimate tasks, short Sprints let teams learn their velocities earlier.
These tips apply to fixed-length timeboxes of other agile project and portfolio management frameworks. For instance, SAFe® recommends that the iteration should last two weeks1. Read Sprint vs. cadence vs. iteration.
Interestingly, software teams seldom choose 1-week long sprints. Instead, 2-week sprints are the norm. This has to do with the “architectural challenges” of software development and with stakeholders not always attending Sprint Reviews. Another counterintuitive truth about Sprint length is as follows: those new to agile should stick to shorter Sprints – ideally 1 week.
What omens do indicate too short or too long cadence of Sprint?
Table 1. Signs of too short or too long Sprint
While these are pretty objective indicators, it is also good to be aware of biases. Teams will generally favor long Sprints, for the sake of comfort, while it is in the best interest of management to have shorter Sprints so that obstacles get revealed sooner.
Also, industries with stable requirements and established technologies get away with longer Sprints.
Does this sound as if you needed to review the length of your Sprint periodically? Nothing could be further from the truth. Reviewing the length of Sprint should occur incidentally, if at all, since it resets tracking at the initiative level. Velocities invalidate; forecasting and planning of future Sprints start from scratch; teams escape performance metrics for some time.3
How to fit into 1-2 week Sprints?
- Practice splitting user stories in your backlog into pretty small tasks and sub-tasks.
- Re-plan over the course of a Sprint. Initially, fill Sprint to 50-60% capacity and add the rest later.
- Acknowledge that the end of Sprint is not a deadline. Rather it is an opportunity to pause, reflect, learn and improve.2
21 Tips for Choosing a Sprint Length, Mishkin Berteig, Jun 2014, berteig.com
Frequent dilemma: what sprint duration is best for your team, Mar 2014, eylean.com
Sprint Length: What Length is the Right Length? Dan Rawsthorne, 3back.com
1 Iterations, scaledagileframework.com
2 Choosing a Scrum Sprint Length – Shorter Beats Longer, Mark Levinson, Nov 2019, agilepainrelief.com
3 Changes to Length of Sprint and Its Impact, Aug 2020, SCRUMstudy