Atlassian-Platinum-Solution-Partner.svg

Why Scale Agile? Basics of Scaling Agile with SAFe Principles

Trundl_Why_Scale_Agile

Teams usually want to Scale Agile with a view to amp up speed, quality and add value to existing outputs that they are achieving with their existing practices. Easier said than done, as it is also not uncommon to come across companies bluffing customers in the name of Agile. We’re pretty sure you must have worked with or come across companies who call themselves Agile, but in reality, they were just ‘old fashioned waterfall’ teams who conduct scrum ceremonies.

Don’t worry, we do not want to scare you away from Scaling Agile, but the reason we’re starting on this note is to first help you get the basics right. You cannot get away by saying “Agile is a mindset”, and then follow anti-patterns. Think about it, if the foundation is wobbly, scaling it for more teams would only invite disaster.. wouldn’t it?

To begin with, we would recommend doing an honest introspection of your practices, processes and toolsets, and once your teams feel ready, let’s get the ball rolling!

How to Scale Agile ?

Understand that it will require significant effort from leaders and team members to setup and function in a Scaled Agile environment. Teams that wish to implement SAFe® should be backed by the top-management, and more importantly, believe in their reason for change.

Now, there are many ways you can choose to Scale Agile: LeSS, DaD, SoS, etc. In this blog, we will focus on Scaled Agile Framework or simple SAFe®.

Created by Dean Leffingwell in 2011, Scaled Agile framework has undergone many evolutions from its origin to where it stands today at v5.1. The framework provides a structured approach which can be incorporated using 4 configurations – Essential SAFe®, Large Solution SAFe®, Portfolio SAFe® and Full SAFe®.

The framework provides users with a set of proven principles and core values which can be leveraged to build practices for teams who wish to deploy the same.

...But why principles? It’s just theory, Right ?

Well, Oxford dictionary defines principle /ˈprɪnsɪp(ə)l/ noun/ plural: principles as “a fundamental truth or proposition that serves as the foundation for a system of belief or behaviour or for a chain of reasoning.” To sum up, it is basically a set of rules to conduct yourself in a certain manner.
So pay attention to the following principles, as these should define your processes to deploy SAFe®. Keeping that in mind, let’s look at the 10 principles as defined in the framework.

Trundl_SAFe®_Agile

#1 Take an Economic View

What is the goal of Lean?
To provide best quality and value in the shortest sustainable lead time. SAFe® suggests that economics should drive the entire decision making system. It further talks about achieving this using two practices:

a) Deliver early and often:
Delivering a solution incrementally rather than doing a ‘big bang release’ has direct economic benefits.

Understand that value begins to accumulate in the incremental delivery system after the first iteration, but starts only at the end when it comes to the waterfall model.

Img 3

b) Applying a comprehensive Economic Framework:
This can be achieved by considering the following parameters –
i) Operating within lean budgets and guardrails: The goal is to move from cost center based budgeting systems towards funding portfolio value streams.

ii) Understanding solution economic trade-offs: Understand the linkages between development expense, lead time, product cost, value and risk.

iii) Leveraging suppliers: The relationship between suppliers and buyers is crucial, and deciding between transactional and partnership can make or break an engagement. Learn more here

iv) Sequencing jobs for maximum benefit: SAFe® is a flow based system in which one picks a work to minimize the cost of delay based on WSJF (weighted shortest job first)

WSJF = Cost of delay / Job duration (size)

Learn more here.

#2 Apply Systems Thinking

Understand that when you optimize a component, it does not optimize the system.
Keeping this at focus, here are three of its aspects –

a) The solution itself is a system:
The “value” of a system passes through its interconnections, and it can evolve only as fast as its slowest integration point.
b) The enterprise building a system, is itself a system:
Leaders need to cultivate an environment for a better instrument which will in turn deliver the solution. Understand the interconnections here, and evolve suppliers/customers relationships as a part of systems as well.
c) Optimizing the value streams:
To go from concept to cash, organizations need to holistically look at the entire value stream, especially when it crosses tech and org boundaries.
Read more here.

#3 Assume Variability; Preserve Options

Whenever we are designing a system or software, there is a lot of uncertainty involved. SAFe® suggests bringing in a set based design approach, which recommends keeping multiple requirements/design options through the dev cycle. As teams learn over time, they can eliminate the options and choose the best path for the customers.
More here.

Trundl_SAFe®_Agile

#4 Build Incrementally, with Fast Integrated Learning Cycles

SAFe® recommends using Shewhart’s Plan-Do-Check-Adjust cycle for continuously improving the quality. The goal is to build in iterations, take feedback from customers, and mitigate risk and variability. The process and architecture are designed to focus on cadence based integration points.

Discover more here.

img 5
Agile PDCA

#5 Base Milestones on Objective Evaluation of Working Systems

In traditional systems, teams work in their silos and even when there are gate reviews at phase ends and a problem occurs, it is too late to go back and solve them. SAFe® suggests basing milestones on objective evidence using demos and on a cadence, thereby enabling systems thinking and building confidence with customers.

Discover more here.

#6 Visualize and Limit WIP, Reduce Batch Sizes, and Manage
Queue Lengths

To move features from concept to cash in a system, SAFe® suggests enabling ‘flow’. This can be achieved by three aspects –

a) Visualize and limit Work in Process (WIP)
Make sure you are limiting the overlapping work. Use a Kanban board to see how many items you have ‘in progress’. This helps teams to identify bottlenecks in the system.

b) Reduce the batch sizes of work items
Remember, small batches go through system with less variability. Rather than pulling in big chunks of work, break them down to ‘bite sized’ work which will enable work to be done at faster speed.

c) Manage queue lengths
While using automation and efficient test practices are important, if you keep the queue of committed work short, the wait time will also reduce. This allows work of higher priority to get in and out of the system faster.

Find out more here.

#7 Apply Cadence, Synchronize with Cross-Domain Planning

Cadence is a rhythmic pattern which puts uncertainty at ease for all routine tasks, creates space for variable ones and instills collaboration. On the other hand, synchronizing these activities with others using cross domain planning enforces quality in the system.

Discover more here.

Trundl_cadence_vs_synchronization

#8 Unlock the Intrinsic Motivation of Knowledge Workers

Team members want to be valued, and compensated well, but is money the only motivation ?
Ideation and innovation cannot be achieved by money itself, and rarely by fear. Leaders need to empower their teams with autonomy, a sense of purpose and remove constraints coming their way. Hence, it is imperative to understand the importance of negotiating, compromising, agreeing and committing.

Read more here.

Trundl_knowledge_workers

#9 Decentralized Decision Making

When decisions are made from the top, problems arise due to lack of context, and causes delays.

This advice could come handy during decision making:

Centralize when decisions are infrequent, long lasting and demands significant economy of scale.
Decentralize when decisions are frequent, time critical and requires local context.

Learn more here.

#10 Organize Around Value

The latest entrant with SAFe® 5.1 into the principles section is focus on value.
It urges businesses to build technology portfolio around development value streams, and suggests the formation of cross functional and cross discipline ART (Agile release trains) and Agile teams to deliver value directly.

Explore more here.

Img 8

These fundamental concepts are updated with references from successful deployments, systems thinking and Agile methods which actually work in real life organizations. Keeping these at the heart of your execution will provide teams with the right direction for focusing on achieving agility at a scaled level.

At Trundl, we focus on the basics as much as the sweet execution of processes. Explore our Agile offerings here.

Want to know more about SAFe? Read our blog on Program Increment or PI Planning

Sajit Nair

Sajit Nair

(SAFe Program Consultant (SPC) | Atlassian Community Leader | Program Manager at Trundl Inc) Sajit Nair is an Agile consultant working with technical teams to deliver quality solutions with expertise in Atlassian PM toolsets like JIRA and Confluence.

Share This Post

Share on facebook
Share on twitter
Share on linkedin
Share on pinterest
Share on email

More To Explore

Blog

Atlassian Community… Security matters!

For those who read ‘The Phoenix Project – A Novel about IT, DevOps and Helping Your Business Win,’ Chapter 21 stands out when it comes to security

We drop Atlassian knowledge.

Once a month we send out Atlassian product updates, recent use cases, helpful videos, Atlassian security advisories, and other resources for the Jira-using community.

It’s our way of keeping you a happy Atlassian customer & bringing new ideas to your technology users.