NOTE: This version is being updated to reflect the Agile Stream book, the changes will be available when it is complete
Agile Stream Framework is a management approach to streamline enterprise operations along Agile Streams. It builds upon the principles of Scrum and Lean to target management contributions in agile transformations at lowering organizational drag around the SDLC. Agile Stream Framework (ASF) consists of roles, events, artifacts, and the rules that bind them together.
Stream Team (ST) mirrors the composition of an Agile Team and its mission to shorten the time a work item travels along an agile stream by removing organizational impediments. The Stream Team is accountable for prioritizing and resolving impediments within the organization, along the Agile Stream.
Stream Teams are composed of individuals ubicated at organizational touch points along the Agile Stream and Scrum Masters for the teams they support, with the primary role to resolve impediments for them in a timely matter.
Stream Teams include business representatives, managers and individuals that have authority within the organization to resolve issues related to work crossing organization boundaries. Team also includes an agile coach to help the team maintain an agile approach in their interaction with teams and understanding the value behind their actions.
When Stream Teams actively participate in removing impediments and manage streams versus teams, they become de facto ambassadors and champions of a sustainable agile transformation within the organization.
Stream Teams serve two to nine Agile Teams, however its members can belong to more than one Stream Team as long as they are committed to meeting working agreements.
Stream Team responsibilities include:
Stream Owners (SO) team mirror the composition of an Agile Team and its mission to improve organizational agility. Stream Owners team is accountable for identifying and addressing organizational agility along Agile Streams.
Stream Owners team is composed of representatives from Stream Teams and leaders with primary role to align activities and procedures to accelerate delivery and responsiveness along Agile Streams and improve organizational agility.
Agile Stream Framework events are appended to or placed around exiting Scrum meetings. These events include Daily Stream Meeting, Stream Experiment Meeting and Stream Review Meeting.
Agile Teams meet daily for their Daily Scrum Meetings, and when an impediment is identified as outside of the team’s control, the information is transferred to the Daily Stream Meeting of its Stream Team.
Daily Stream Meetings occur daily, as long as there are items on the Active Stream Board, for a maximum of 45 minutes. During that meeting Stream Team members review and update their Active Stream Board. The goal of the Daily Stream Meeting is to identify which impediments to resolve on that day. This meeting generally occurs after team standups and it is divided into three activities:
Stream Experiment Meetings occur before each Sprint and has a one-hour time-box. During that meeting Stream Team members define experiments designed to streamline workflow along Agile Stream. The goal of the Stream Experiment Meeting is to design experiments with testable hypothesis with goal to optimize an Agile Stream. This meeting generally occurs before start of a Sprint.
Stream Review Meetings occur after one to two Sprints and has a one-hour time-box. During that meeting Stream Team members identify action items that would reduce impediments for Agile Teams. The goal of the Stream Review Meeting is to identify which impediments to reduce or eliminate along the Agile Stream. This meeting generally occurs after Agile Team Review meetings.
This meeting mirrors Scrum retrospective in purpose to find ways to permanently improve process around SDLC based on what was learned. The team generally engages in:
Artifacts in Agile Stream Framework represent the tools needed to identify, rank and visualize organizational impediments. They provide transparency for Agile Teams, Stream Teams and Stream Owner Teams and used to surface opportunities for inspection and learning.
A Stream Backlog contains a list of actionable items called Stream Backlog Items (SBI). SBIs include Impediments and Improvement Items, that when addressed by organization would accelerate future work or unblock impeded work along the Agile Stream for one or more teams. Stream Backlog Items are:
Empirical learning is a key principle of agile, and managers can use Stream Experiments to explore the effect of changes to the organizational ecosystem.
Stream experiments follow the premise of viewing an Agile Stream as a scientific experiment and to think creatively about how to test operating assumptions. Stream Teams measure the effects of backlog planning, story workshops, training, coaching, compliance, tools, etc. and validate hypotheses based on empirical data to enable teams to hyper perform and increase organizational agility in the process.
Stream Team use Story Coefficient of Drag (SDC) to quantify impediment impact on teams’ ability to deliver an increment in a sprint, and Stream Turbulence Index (STI) to gauge fluctuations in delivery stream. These metrics lend themselves to be clustered into teams, products, silos, etc.
The first ASF metric is Story Drag Coefficient (SDC) and it is an agile-friendly constant used across Agile Teams and reflects the effect of organizational drag, observed as Impeded Story Points (ISP), on Teams’ Planned Capacity (TPC) to deliver a subset of story points within a Sprint (SL).
SDC = Impeded Story Points / (Team Planned Capacity x Sprint Length )
The resulting coefficient, SDC, can then be multiplied by the number of days since the impediment has been raised at the team level, to compute the Cumulative Drag in days.
Cumulative Drag = SDC x Days Unresolved
Daily Drag Index (DDI) is the Cumulative Drag, adjusted to the organizational impediment removal strategy. It is used to provide instant empirical feedback to Stream Teams as to which impediments carry the highest value and to be ranked accordingly.
Stream Team in collaboration with Stream Owners and Agile Teams decide on an impediment strategy that best aligns the business and Agile Teams goals as to how to rank impediments for removal. Some of the common strategies include:
Stream Turbulence Index (STI) is constant that reflects the deviation between optimal and actual team’s plan delivery. It is used to observe fluctuations in the Agile Delivery Stream and used as an early warning indicator of organizational drag observed at the level of Agile Teams.
STI = (Forecasted End Day / Sprint End Day) - 1
STI relies on a Forecast End Day for the Sprint based on the team’s actual burn-down trend. FED can be calculated using linear regression and used for comparison against optimal plan.
STI provides a simple indicator on the deviation from plan in Sprints, where an STI of 2.5 indicates that team’s delivery is off by 2.5 sprints, whereas an STI of zero means that the delivery and plan coincide. STIs can be aggregated to estimate uncertainty of plans and can be grouped by teams, releases, products, etc.
Stream Backlog Items are posted on the Active Stream Board, which integrates Work In Progress (WIP) limits. Digital versions of ASB would integrate Daily Drag Index (DDI) to facilitate impediment ranking by ST. Active Stream Board boards may use WIP limits on silos or the whole board. Figure 4 and Figure 5 display two styles of the ASB with and without silos respectively.
Figure 4 Agile Stream Board with separate silo lanes and WIP limits
Figure 5 Active Stream Board with WIP on all impediments on board.
ASF Analytics, although not part of the framework itself, provide the agile stream practitioner with examples on how to use the information produced by the framework to understand the different ways they can be used to track progress and understand the overall effect of removing or keeping the impediments.
Organizational drag from impediments can be visualized by grouping Daily Drag Index (DDI) across teams, organizational silos, technologies, etc. Stream Teams and Steam Owners can predefine how to measure DDI when conducting experiments or understand the impact of impediment over a set of variables. In the following examples, will showcase few options and explain how to read the graphs.
Visualizing DDI grouped by teams versus a group of impediments showcases its effect on each team. While some teams can overcome certain organizational impediments, a subset might not be able to for different reasons. By comparing these, you can look for ways to empower, train or coach the teams that are struggling.
Visualizing DDI grouped by silos or organizational areas versus showcases which silos produce the highest number of impediment on the teams. Stream Owners in collaboration with organizational leaders and Agile Teams can address those areas to accelerate adoption and agile transformation.
Figure 6 Aggregate DDI by team over a period of time.
|
Figure 7 Aggregate DDI over silos and organizational areas. |
Visualizing DDI trend lines of raised and resolved impediments provide Stream Owners, Stream and Agile Teams with a total view of what is happening on the organization or group level. While it is normal for teams to experience impediments, spikes and accumulation of impediments levels, act as brakes that delay deliver and increase cost of each increment of work.
Figure 8 DDI Open and closed trends
Managing Agile Streams requires a transparent approach to determine the effect of changes on Agile Teams capacity to deliver software within a Sprint. Additionally, by empowering teams to self organize, managers need a tool that can allow sampling of progress, without adversely affecting performance or apply undue pressure that will force teams to revert to waterfall practices.
In 2015, VersionOne reported in its 9th annual State of Agile Survey that 58% of respondents measured agile success by On-Time Delivery. Stream Turbulence at its core, provides the organization with both a macro and micro viewports into organizational ability to deliver on time. Stream Turbulence Index is the main unit of measure and can be inspected at several stages in the software development lifecycle.
The same VersionOne report states that 59% of respondents measure day-to-day success based on Velocity. It is worth mentioning that Velocity was not intended to be used as a measure of team’s success, and its use negatively impacts agile transformation, where management focus is diverged into increasing velocity and using velocity to compare teams, resulting in increased dysfunction and delays in agile transformation initiatives.
On the other hand, Stream Turbulence can be used to support team needs and shifts focus on supporting their daily goals. This frees managers from using velocity as an absolute unit of measure, and replaces it with STI as a tool to provide empirical feedback, learn from successful teams and engage struggling ones.
Figure 9 provides an example of micro viewport of team’s ability to deliver on time. Radar charts of STI can be inspected during Stream Review and Experiment Meetings to identify deviations in forecasted delivery and allows Stream Teams to see which teams need help most. In this example, most of the teams are on-time and considered On Stream, while a couple of teams could benefit from coaching and considered in Turbulent Stream, and one team has deviated considerably from their delivery goals for the time period observed and are considered Off Stream.
Variations of the approach in Figure 9 would include clustering of STI along Product Streams to identify delivery delays and provides Stream Team with tools to target areas to optimize in Agile Stream.
Figure 9 Stream Turbulence Index radar at a point in time for a group of teams.
Figure 10 provides an example of macro viewport of organization’s ability to deliver on time and provides Stream Owners and Stream Teams with a bird’s eye view of the entire Delivery Stream. This graph reflects organizational changes and can be team or product focused. Macro viewports are sanity check tools to overlay on the short and long term effects of changes made to the Agile Stream.
Figure 10 Stream Turbulence Index sampled over several Sprints
Copyright © 2013 - Agile Stream Framework, 131 Daniel Webster Hwy #712, , Nashua, NH 03060 , USA. P:(844) 842-2488 Powered by Ayna | Manage