This is a sample agenda for a sprint planning meeting. Depending on your context you will have to change the details, just make sure the outcomes stay the same.

Meeting purpose: Plan and prepare for the upcoming sprint
Meeting duration: ca. 1 hour for a 2 week sprint (it will be less if you’re an experienced team and know your domain well)

Part 1:

1. Intro & set the scene

Remind everyone that the purpose of the meeting is to plan the work for the next sprint.
Remind everyone of the sprint duration and demo date.

Share the desired meeting outcome:
– A sprint goal
– A sprint backlog including tasks

Share the planned agenda:

  • Define the sprint goal
  • Decide how much to chew off
  • Decide which stories to do this sprint
  • Create tasks for stories

2. Define the Sprint goal

Come up with a sprint goal: The purpose of having a goal is to be able to select user stories that support the goal, i.e. help you work towards something bigger than just delivering a collection of stories or unrelated features. It’s also really useful to assess at the end of the sprint whether you have been successful or not. Success is achieving the sprint goal, not just delivering all the stories you had planned to get done.

Examples could be:

  • Build edit delivery feature
  • Make the app hotter!
  • Demo advanced property search” (It’s okay to have a goal that is to collect feedback through a demo/prototype)
  • Fix push notifications and improve motors search” (One goal is better but if it is way too small it’s okay to have a dual goal. Just make sure you don’t end up with just a collection of unrelated user stories)

3. Decide how much to chew off

If you are using story points or count the number of stories delivered as a measure of velocity it’s a good idea to decide on a target velocity next. Base that decision on looking at a combination of past velocity, people’s availability (any holidays?) and just plain gut feeling.

Make sure you take past data into consideration, not just wishful thinking about how much you feel you “should” get done. This is also the reason why I like to discuss target velocity before choosing stories for the sprint: It is all too easy to get carried away with enthusiasm and optimism.


If you don’t have any data yet, don’t worry – just go by gut feeling and start collecting data now.

4. Decide which stories to do in this sprint

Take your product backlog and discuss for each candidate story what it means, what you’re trying to achieve and what success looks like. Do this for all the stories you think you might be able to schedule into the sprint.

Choosing which stories to consider for your sprint is mainly determined by the sprint goal. Other considerations are:

  • What if we can do more than just the stories that support the goal?
    You could have a secondary goal (“Fix push notifications and improve motors search”) but make sure these are prioritised against each other. Or you could “fill” with bug fixes, enhancements and smaller, not necessarily related requests. Make sure to communicate that these are less important though than things that support the sprint goal and will get dropped if the goal is in danger.
  • What if there’s stuff we really want to do such as re-paying some technical debt?
    I like agreeing on one or two of such items for each sprint. I have found ca. 20% of a sprint’s work dedicated to the team’s wishes to be a good thing.
  • What about high risk items?
    It sometimes makes sense to prioritise a risky story higher to just learn, gather knowledge and be able to plan. Early prototypes for hairy features are a good idea.

What it comes down to is to have a discussion between the team with their technical expertise and the Product Owner as the representative of end users and stakeholders and to decide together what you want to do during the next couple of weeks.

Also, make sure not to “fill up” your entire sprint with planned work but to leave some room for unexpected opportunities or complexities.

Once you have decided which stories to aim for during the next sprint, move on to part 2.

Part 2:

5. Task planning

I like planning individual tasks for each user story and to make work visible at a granular level. This makes it easier for people to collaborate around user stories and to help each other reach a shared goal.

As an example for the user story “As a photographer I want to make some of my albums private so that no one else can see them” we identified the following tasks:

If you find it hard to make tasks ask yourself how you would get started. And what you would do next? And then? You’ll find you can come up with quite a lot of useful tasks that need to be done.

Try to keep tasks to less than a full day’s work so that stuff moves through on a daily basis and there is a sense of progress and opportunity for collaboration during the daily standup. Also, try to split them in a way so that in theory different people could do different tasks.

Remember that you don’t need to get the tasks right upfront: It’s perfectly fair to add new tasks to a story or to delete tasks that are no longer needed during the sprint.

And don’t assign tasks upfront! It will just make you wait for each other and cheat you for opportunities to collaborate and help each other.

Part 3: Set up your visual workspace

This is the outcome of your sprint planning meeting: A visual workspace with a list of user stories and associated tasks.

Here’s an example:

If you aren’t co-located you can of course achieve the same in Jira or some other tool. I really like the tactile and highly visible nature of the post-it note board though, so if you have the option I’d definitely use one of these.

HOW TO RUN A SPRINT PLANNING MEETING (THE WAY I LIKE IT)的更多相关文章

  1. OneZero第一次站立会议&Sprint Planning Meeting(2016.3.21)

    会议时间:2016年3月21日 11:40~12:00 会议成员:冉华,张敏,王巍,夏一鸣. 会议目的:列举第一周工作内容,确定第一周工作分配.即挑选出一个Story作为本次迭代完成的目标. 会议内容 ...

  2. 一张图看懂Sprint Planning Meeting

    本文主要参考<Scrum精髓>这本书的内容 每个Sprint都是从Sprint Planning Meeting开始,Scrum团队成员聚集在一起商定下个Sprint目标,并且确定在Spr ...

  3. 敏捷软件开发实践-Sprint Retrospective Meeting(转)

    介绍: 在敏捷开发模式中,Sprint Retrospective Meeting 也是一个必不可少的环节,它通常发生在每个Sprint的结尾,其主要作用是对于当前的迭代周期做一个阶段性的总结,包括好 ...

  4. Scrum的三个仪式:Sprint规划会,Scrum每日站会,Sprint评审会

    转自:http://blog.sina.com.cn/s/blog_6997f01501010m21.html Sprint Planning Meeting(Sprint规划会) 根据Product ...

  5. A Practical Guide to Distributed Scrum - 分布式Scrum的实用指南 - 读书笔记

    最近读了这本IBM出的<A Practical Guide to Distributed Scrum>(分布式Scrum的实用指南),书中的章节结构比较清楚,是针对Scrum项目进行,一个 ...

  6. [转载]敏捷开发之Scrum扫盲篇

    现在敏捷开发是越来越火了,人人都在谈敏捷,人人都在学习Scrum和XP...      为了不落后他人,于是我也开始学习Scrum,今天主要是对我最近阅读的相关资料,根据自己的理解,用自己的话来讲述S ...

  7. iOS项目分析及优化

    iOS项目分析及优化  来源:吴白的简书   从代码看一个程序员的笔力 从代码的整洁度上就可以看出一个程序员的实力,规范其实就是让你养成一种良好习惯的标杆,在此面前我们应该顺从.本篇我们以OC为例,统 ...

  8. ACCEPTANCE CRITERIA FOR USER STORIES

    One of the teams I have recently coached quickly got a grasp of how to phrase user stories but found ...

  9. [转] Agile Software Development 敏捷软件开发

    原文作者:kkun 原文地址:http://www.cnblogs.com/kkun/archive/2011/07/06/agile_software_development.html 敏捷是什么 ...

随机推荐

  1. dubbox rest服务

    1.xml配置: web.xml 定义监听的contextpath,见rest协议定义 <servlet> <servlet-name>dispatcher</servl ...

  2. windows下nginx安装、配置与使用(转载)

    目前国内各大门户网站已经部署了Nginx,如新浪.网易.腾讯等:国内几个重要的视频分享网站也部署了Nginx,如六房间.酷6等.新近发现Nginx 技术在国内日趋火热,越来越多的网站开始部署Nginx ...

  3. Linux学习记录

    ---恢复内容开始--- linux与unix的关系 linux是借鉴了unix设计思想,也称linux位类unix系统. Linux常用命令 1.命令基本格式 命令[选项][参数] 注意:个别命令不 ...

  4. javaSE基础04

    javaSE基础04 一.三木运算符 <表达式1> ? <表达式2> : <表达式3> "?"运算符的含义是: 先求表达式1的值, 如果为真, ...

  5. django学习记录

    1.参考资料问题: 现在django发布了1.11版本,离线文档下载引擎地址 文档下载地址 在线文档:https://docs.djangoproject.com/en/1.10/intro/tuto ...

  6. Linux更改计算机名称

    1.修改:vim /etc/hosts 2.修改:vim /etc/sysconfig/network 3.重启:reboot 如不重启可以输入:hostname  新改的计算机名称,然后su

  7. js实现返回顶部功能的解决方案

    很多网站上都有返回顶部的效果,主要有如下几种解决方案. 1.纯js,无动画版本 window.scrollTo(x-coord, y-coord); window.scrollTo(0,0); 2.纯 ...

  8. 两种方式testng dataprovider结合csv做测试驱动

    方式一: 第一.读取csv数据源码 import java.io.BufferedReader; import java.io.File; import java.io.FileInputStream ...

  9. Ext.js的store里放model,还是field?

    按别人的经验, 一般来说,如果通用性强的应用,STORE里存放MODEL,便于重用代码. 如果通用性较弱的(报告,图表),则考虑使用field进行定制.

  10. T-SQL 基本语法

    --查询 select DB_ID('B2C') --检查数据库是否存在 if DB_ID('B2C') is not null --使用数据库 use B2C --单引号表示字符串,双引号则不是 U ...