Skip directly to search

Skip directly to content

 

Distributed SAFe PI Planning

 
 

Agile | Florin Manolescu |
30 June 2020

Co-authored by Marius Leatu.

When we started to think about switching to a SAFe delivery model in our project, nobody could have predicted the challenges that we would face during the preparation of our first PI planning event. Due to the situation we all now face globally, we had to reconfigure the event to be completely remote. What follows is our story and what we did before and during PI planning to ensure a successful event.

We participated in a PI planning event with several Endava teams, but there were also other suppliers involved. Here is some additional context regarding the facts and constraints involved with this particular event: It took place across nine cities, and more than 130 people were involved. Each person was at home, attending virtually. The main facilitator of the PI planning was a native English speaker from another supplier, and we had Agile champions (local facilitators) in each location – mainly the Scrum Masters of the teams.

Several key factors supported our success:

1. Planning and preparing

We cannot overstate the importance of preparation ahead of time. Product and architecture people prepared and pre-refined up to 10 features and tech enablers with the teams, and six of them were pulled into the final PI plan. Our agile coach, Marius Leatu, spent four sessions of 1.5 hours each walking the teams through what SAFe is, what PI planning involves, what success looks like, and what to focus on in preparation for and during the event.

2. Online tools for every need

The key tool that made all of the difference was Miro – that is, the online whiteboard with a special template for PI planning. It allows for editing in real-time, collaboration, sync, etc. For tickets and documentation, we used JIRA and Confluence. The online meeting software BlueJeans provided audio-video capabilities and 20 virtual rooms available for hot-join. One room was used for plenary sessions, and others for team breakouts, retros, and ad-hoc usage.

For presentation slides, PowerPoint served our needs, but there are also other online office tools that can be helpful with presentations. Additional communication took place through Slack and Microsoft Teams for instant messaging and group chats, and we used the online planning poker tool PlanITpoker for project estimation.

3. A clearly defined agenda

The agenda was walked through several times with champions and leadership roles and was adjusted to meet time zone challenges, as some faced a time difference ranging from one to two hours. Clearly defining the agenda, especially around the start, end, and lunch times was crucial for the event’s organsation.

4. Open communication

The massive use of Slack channels and Microsoft Teams for broadcasts and team organisation – as well as one-to-ones during the event – proved to be very important. Being able to have platforms which maintained open communication lines for direct contact was imperative.

5. Strong leadership

The governance team (agile coaches, CPO, Program Manager, Release Train Engineer) were in constant communication and review. There was not just one, but two management review sessions, as well as an abundance of back-channel communication on Slack.

We learned a lot from the experiences with our PI planning event, and the current circumstances of indefinite dispersed working taught us even more about being thorough and adaptable. As such, there are several key details that must be adhered to for a PI planning event to be successful during these times. Implementing the following practices should provide positive results:

Working agreements: When you have more than 130 people online and working from home, you should have some clear communication rules to avoid chaos. The Release Train Engineer (RTE) should be the curator of those agreements during common sessions. In our case, there was an appointed Agile champion that acted as a local RTE for each location.

Planning agenda: Create an online agenda available for everyone and discuss it with your teams in all locations and time zones. Keep in mind the different time zones and adjust the agenda to accommodate all teams. Plan for more breakout sessions for each team and plan for coffee breaks and lunch breaks. Enable the possibility to synchronise between teams and make sure there will be time slots available for eventual delays that could occur due to dispersed working.

Product backlog preparation: Ensure the product backlog is prepared for the PI planning event, and features are explained and clarified in just enough detail that enable teams to estimate and plan. Make certain the definition of ‘ready’ is fulfilled and teams have had refinement beforehand. Establish (and re-establish) clear priorities and plan according to those, taking into consideration team capacity. For two weeks before the event, Product (PMs, POs, CPO) should have regular catchups to see the status of refinement and narrow down priorities – a.k.a. scope trimming.

Orchestration of the event: Have online facilitators for each team/location and an RTE for the common sessions. Agile coaches should also be available, if possible, to help orchestration and to provide feedback and guidance on the spot. Ensure people like Product Owners, E2E business analysts, and architects are constantly available on Slack to be pulled into online meeting sessions. Hop on and hop off in different sessions if needed for clarifications.

Team preparation: For us, each team had already identified their capacity for the sprint, picked reference stories, and prepared a backlog of platform-level improvements to fill in the planning gaps if capacity remained available. This process is highly recommended.

People: Ensure people, especially leaders, are focused and committed. Check that people are aware of what they must plan, and which are the PI planning objectives. Confirm that no other project-related activities can impact the focus of the people involved and are not happening in parallel (e.g. deployments, testing, etc.).

Wrap-up: Finish with a retrospective, debrief, and a virtual drink at the end to celebrate.

For the sake of further instruction, let’s look at the client’s outlook when we started to think about switching to a SAFe delivery model in our project and prepared for our first PI planning event. The client was concerned with several things:

  • Preparation of requirements: To ensure this, we maintained regular catchups on requirement preparation for PMs and POs to focus their effort during a very busy period.
  • Online communication: Since communication is crucial, there was a large time investment for selecting and configuring the right online tool. We started with Trello and JIRA, and eventually landed on Miro as the tool of choice for planning. Additionally, we created ample BlueJeans sessions and Slack channels, and had key facilitators involved more than was initially necessary to guarantee fluid and open correspondence. Finally, we documented how to use the tools and circulated the information often.
  • Managing dependencies between teams and producing an integrated plan for at least five features: It helped massively that the main facilitator was very skilled and had help from two agile coaches and a Program Manager for coordination. Also, Miro helped a lot with visualisation during alignment sessions.
  • Ensuring success in a pilot for SAFe implementation and first-time PI planning: This involved many senior managers investing time in preparation and dedicating time to joining the sessions. It even included the CPO of Merchant Services BU joining for 10 minutes during Day 1 and giving a stirring speech of encouragement.

We hope this information has been useful. Please don’t hesitate to contact us to know more, or if you want additional details related to our actions and preparations.

Florin Manolescu

Agile Project Manager

Florin is a passionate and experienced Project Management professional with over 22 years of experience in the IT Software field, with half of them in the role of Project and Program Manager. His journey with Endava began on March 2019 when he joined as a Project Manager and Scrum Master. When not in the office, Florin is nurturing his love for music and travelling.

 

Related Articles

  • 08 July 2020

    A Virtual Hackathon Together with Microsoft

  • 30 June 2020

    Distributed SAFe PI Planning

  • 09 June 2020

    The Twisted Concept of Securing Kubernetes Clusters – Part 2

  • 15 May 2020

    Performance and security testing shifting left

  • 03 September 2019

    Creating A Visual Culture

  • 06 August 2019

    Evaluating the current testing trends

  • 09 July 2019

    Developing your Product Owner mindset

  • 11 February 2019

    Distributed Agile – Closing the Gap Between the Product Owner and the Team

Most Popular Articles

A Virtual Hackathon Together with Microsoft
 

Innovation | Radu Orghidan | 08 July 2020

A Virtual Hackathon Together with Microsoft

Distributed SAFe PI Planning
 

Agile | Florin Manolescu | 30 June 2020

Distributed SAFe PI Planning

The Twisted Concept of Securing Kubernetes Clusters – Part 2
 

Architecture | Vlad Calmic | 09 June 2020

The Twisted Concept of Securing Kubernetes Clusters – Part 2

Performance and security testing shifting left
 

Testing | Alex Gatu | 15 May 2020

Performance and security testing shifting left

AR & ML Deployment in the Wild – A Story About Friendly Animals
 

Augmented Reality | Radu Orghidan | 30 April 2020

AR & ML Deployment in the Wild – A Story About Friendly Animals

Cucumber: Automation Framework or Collaboration Tool?
 

Automation | Martin Borba | 16 April 2020

Cucumber: Automation Framework or Collaboration Tool?

Challenges in creating relevant test data without using personally identifiable information
 

Testing | Alex Gatu | 25 February 2020

Challenges in creating relevant test data without using personally identifiable information

Service Meshes – from Kubernetes service management to universal compute fabric
 

DevOps | Oleksiy Volkov | 04 February 2020

Service Meshes – from Kubernetes service management to universal compute fabric

AWS Serverless with Terraform – Best Practices
 

Architecture | Vlad Cenan | 10 December 2019

AWS Serverless with Terraform – Best Practices

 

Archive

  • 08 July 2020

    A Virtual Hackathon Together with Microsoft

  • 30 June 2020

    Distributed SAFe PI Planning

  • 09 June 2020

    The Twisted Concept of Securing Kubernetes Clusters – Part 2

  • 15 May 2020

    Performance and security testing shifting left

  • 30 April 2020

    AR & ML Deployment in the Wild – A Story About Friendly Animals

  • 16 April 2020

    Cucumber: Automation Framework or Collaboration Tool?

  • 25 February 2020

    Challenges in creating relevant test data without using personally identifiable information

  • 04 February 2020

    Service Meshes – from Kubernetes service management to universal compute fabric

We are listening

How would you rate your experience with Endava so far?

We would appreciate talking to you about your feedback. Could you share with us your contact details?