Bill Ryan Product manager with B2C and extensive marketplace experience. As a steadfast follower of the Jobs-to-be-Done methodology, I strive to understand customer struggles as the primary catalyst for innovation and optimization.

What is sprint planning? Guide and meeting agenda cheat sheet

7 min read 2072

What Is Sprint Planning? Guide And Best Practices

Sprint planning is an important scrum ceremony in which the scrum team decides what work it will commit to in the upcoming sprint. Whether your teams work in two- or four-week sprints, the sprint planning ceremony should take place at the beginning of each sprint.

In this guide, we’ll define sprint planning at a high level before diving into the details, including who participates in sprint planning, what occurs during the ceremony, and best practices for running a sprint planning meeting.

Scroll to the end to see a sprint planning meeting agenda template you can use as a cheat sheet when conducting your own sprint planning ceremony.


Table of contents


What is sprint planning?

sprint planning meeting is a session in which the product manager and development team plan the work the team will commit to for the upcoming 2–3 weeks.

Sprint planning meeting are usually full-day events for a month’s sprints but can be proportionately shorter for shorter sprints managed by a scrum master.

Who attends a sprint planning meeting?

The sprint planning ceremony is usually led by the product manager, product owner, or scrum master. All members of the scrum team should be in attendance. This includes engineers, QAs, and designers, each of whom is critical to the sprint planning process.

In addition to leading the meeting, the product manager helps provide business or user context and prioritizes tradeoffs.

Designers can help clear up any last-minute confusion about designs and acceptance criteria.

Engineers define implementation complexity and how much work it can commit to. QA is also important in this regard. Just because a story is dev-complete doesn’t mean it can be released; there needs to be testing capacity too.

We made a custom demo for .
No really. Click here to check it out.

What do you do during sprint planning?

During sprint planning, the scrum team reviews work from the previous sprint, determines its capacity for the upcoming sprint, and commits to the highest-priority work to fill that capacity.

Begin by reviewing the work completed in the previous sprint. Were all the committed stories completed? If not, identify the amount of effort left on any stories or tasks that are rolling over. Taking these into account ensures you won’t overcommit to new work by approaching it as if you have full capacity (because you don’t).

Some notes on sprint rollover:

  • Sprint rollover is not always a bad thing! Many things can contribute to sprint rollover, including critical bugs and unknowns in the committed work (more on these later)
  • Sprint rollover can even be a good thing when the team completes the committed work earlier than expected and can begin pulling from the prioritized backlog
  • Whatever the situation, use sprint rollover as a learning opportunity for velocity adjustments and more accurate planning in the future

After you’ve assessed the work and effort rolling over, the team can discuss new stories to pull in. Again, as the product manager, you should define priority and provide context, but the rest of the team should decide how much of that it can complete.

Tradeoff decisions may need to be made during these discussions. This is where your product management skills are critical.

For example, let’s say you’re hoping to fit in a three-point and five-point story, but there’s only capacity left for five points. You can decide to pull in the five-pointer or opt for the three-pointer with another small story.

When faced with a scenario like this example, you should discuss the business, user, and technical tradeoffs with the team. Don’t delay any future work that’s blocked by the story you’re deprioritizing.

Below is a detailed example of a product backlog groomed and prioritized as a result of sprint planning:

Example: Product Backlog Groomed During Sprint Planning
Source: Lucidchart

Do you need to set a sprint goal?

Sprint goals originated in the early days of scrum as a way to measure success for each completed sprint. While some teams still hold sprint goals in high regard, they’re not always necessary. In some cases, they’re not even possible.

If you’re building a new product that may take several sprints to be MVP-ready, it’s tough to define useful sprint goals. It can be tempting to set a goal to deliver all committed stories in the sprint, but this puts your focus on output rather than outcomes.

As long as you’re prioritizing the most important work along the path to release and the entire team feels comfortable with each sprint’s commitments, you might be better off forgoing sprint goals altogether.

Best practices for sprint planning

Sprint planning is more than just pulling stories into the to-do list until the team says “stop.” Let’s review some key considerations and best practices to help you commit to the highest quality and proper quantity of work while preemptively avoiding roadblocks.

Estimate stories before sprint planning

Use sprint planning to decide what stories to commit to and don’t get lost in the weeds. Diving into the requirements for those stories and estimating how long they may take is a surefire way to derail sprint planning and trap the team in an unnecessarily long meeting.

The backlog grooming ceremony is a much better time to scope and estimate stories before they’re pulled into a sprint. This gives the team time and space to dig into the acceptance criteria and fully understand the effort without the added pressure of figuring it out immediately. This can also surface tradeoffs related to complexities or dependencies, which may need to be evaluated before estimating or committing to the work.

Estimating stories ahead of sprint planning can help you avoid sprint delays and unrealistic expectations.

Account for unknowns in the committed work

Address as many unknowns as possible ahead of the sprint. This can often be done through grooming ceremonies, technical spikes, or other pre-work investigations. However, some unknowns are unavoidable until the actual work begins.

You should plan for these scenarios to avoid overcommitting to complicated work. The best way to do this is to account for these unknowns ahead of time when estimating stories during your grooming ceremony.

For example, if a story feels like it’s probably three points, it’s a good idea to overestimate the time it will take to account for unforeseen issues.

Coordinate dependencies with other teams

Dependencies are a catalyst for roadblocks if not coordinated ahead of time. And they can go both ways.

Maybe your team is building a feature that’s dependent on another team completing API updates. Or maybe the marketing team is waiting to launch a new email campaign and your team is working to build and release the associated landing page.

As a product manager, it’s your responsibility to coordinate and prioritize these dependencies with stakeholders or other product managers prior to bringing the work to sprint planning.

Know your team’s true capacity for the upcoming sprint

Knowing your team’s capacity for the work they can commit to is critical for successful planning. This should not be based solely on past sprint velocity because each sprint is different.

We already talked about accounting for sprint rollover when determining your capacity for each sprint, but there are other factors to weigh: Are there any holidays or vacations coming up? Are there company or department offsites? Does your company do dev days for non-sprint related pursuits? All of these considerations are important in understanding your true capacity.

Anticipate unplanned work and know how to respond

Unplanned work can arise in the middle of a sprint and set your commitments off track, but it doesn’t have to be that way. As a product manager, it’s your job to assess and decide whether something is important enough to be pulled into the in-progress sprint.

This isn’t to say you should leave capacity open on the off-chance that something arises, but you should head into the sprint knowing that some of the committed work could potentially be sidelined by unplanned work.

There are three main types of unplanned work you may encounter:

  • Critical bugs — these are drop-everything bugs that the team should pivot to immediately (e.g., the homepage isn’t loading, payments aren’t processing, etc.)
  • Less severe bugs — beyond critical bugs, severity can range from low to high. Discuss these with your engineers and stakeholders and collect enough information to evaluate the tradeoffs of pulling them in now versus later
  • Urgent feature requests — these can come from leadership, clients, or stakeholders and should be evaluated carefully. Very rarely is a feature request urgent enough that it needs to happen immediately

How to prepare for a sprint planning meeting

Running an effective sprint planning begins with intentional and upfront preparation.

Prior to planning, you should have a clear idea of what you hope to achieve in the sprint. Early steps include product discovery, research, user testing, and other methods to determine the most impactful work. If you’re walking into sprint planning with a new idea you haven’t properly vetted, that is potentially a recipe for disaster.

You should also ensure the work is truly ready to go. Have the tickets been groomed and prioritized by impact or urgency? Have the necessary spikes, tradeoffs, and dependencies all been addressed?

If the answer to any of these questions is no, consider whether the work is truly ready for development. In some cases, a “no” may be unavoidable (e.g., a last-minute bug that should be addressed but hasn’t been scoped yet), but this usually isn’t the case.

How to run a sprint planning meeting

Once you’ve prepared for sprint planning, it’s time to run the meeting. We’ve already discussed how to determine capacity based on rollover work, resource availability, and other factors, which are all important to running a successful sprint planning.

If there’s a single golden rule for product managers to effectively run a sprint planning, it’s this: listen to your team. Developers know their capacity and understand the complexity of the work you’re discussing. Don’t push or challenge their expertise. If they say they can commit to no more than 20 points, don’t push or persuade them to increase it to 25.

Worst case scenario, you miss your sprint commitments and lose trust among your team. Best case scenario, you hit all your deadlines and your team still doesn’t trust you. You should be leading, not dictating.

Sometimes, this means the stories you had planned for during preparation don’t all make it in. If you’re in tune with your team’s work and typical velocity, your wish list should be pretty close to the actual result, but don’t be frustrated if the stars don’t all align.

If you’re listening to your team, considering capacity, willing to make tradeoffs or scope cuts, doing the upfront work for dependencies and prioritization, and understand the acceptance criteria in and out to help answer questions, your sprint planning should be a success.


Sprint planning meeting agenda cheat sheet

As described above, a lot goes into leading successful sprint planning meetings. The sprint planning meeting agenda template below is designed to help you keep track of the most important items to achieve that success.

Pre-meeting preparation

Stories are groomed, estimated, and ready to be worked on

Dependencies are coordinated with other teams and/or stakeholders

Backlog is prioritized by accounting for impact, urgency, dependencies, and blockers

Review of prior sprint

Review work that was completed and shipped during the sprint

Define effort left on stories or tasks rolling over

Identify opportunities for velocity adjustments or ways to make planning more effective

Define sprint capacity

Account for sprint rollover and other factors that may reduce capacity, including holidays, vacation time, offsites, and dev days

Decide on sprint commitments

Optional: Set a sprint goal (avoid making it “ship x number of stories”)

Discuss which of the prioritized stories the team can commit to

Facilitate tradeoff discussions (e.g. two 3-point stories over one 5-point story)

Account for unknowns in any of the stories


 

LogRocket helps you speak the same language as your developers

Thanks for reading about Product Ops. This is an ad for LogRocket.

Not sure what we do? We just won Product School's "Proddy" for best Heatmaps & Session Replay, beating out a lot of great solutions that you probably already use. We make it so much easier for you to work with your developers by diagnosing bugs and catching revenue-killing snags in your app's UI.

See what you're missing - try LogRocket today.
Bill Ryan Product manager with B2C and extensive marketplace experience. As a steadfast follower of the Jobs-to-be-Done methodology, I strive to understand customer struggles as the primary catalyst for innovation and optimization.

Leave a Reply