Delivering Measurable Business Value with Scrum
Ryan Shriver
Managing Consultant
Dominion Digital

Introduction:

Working with many agile teams, I've discovered that agile methods such as Scrum and XP are very good at helping you “ do the thing right ” but don't really help organizations “ do the right thing ”.

By this I mean that these methods don't provide a framework for measuring business value and helping organizations make strategic decisions about how and where to invest limited resources. They don't help focus the organizations on “the right thing” per se, rather they assume you're working on the most important project(s), which may not always be the case. Traditional agile metrics are oriented towards features, team productivity and planning, but not “measurable value delivered” or “percent to objectives”.

I think there is a missing gap in current agile thinking about the type of information and metrics needed by executive business sponsors to make important decisions. This presentation addresses a way to bridge the gap by incorporating a framework for defining and quantifying business objectives (Evo and Planguage by Tom Gilb) together with agile methods such as Scrum. The result is a way to measure your agile projects with respect to how they've helped you meet key business objectives and ensuring your agile projects are aligned with the right set of objectives, while utilizing your current Scrum approach practically unchanged.

Participants will learn:

Participants will learn how to do the following using a combination of lecture and hands-on exercises meant to help the participants learn-by-doing in a group setting with presenter assistance:

  1. Define and quantify business objectives
  2. Achieve value alignment with Scrum
  3. Measuring “value delivered” each release including “progress towards key objectives”

Outline:

The session is presenter-lead with a combination of lecture and small group exercises (hands-on application of material covered in lecture). Participants are encouraged to ask questions, agree, disagree and otherwise engage in discussion on the topic.

Small group teams will work on one example for the duration of the presentation, applying the concepts just covered in the lecture to their problem. We'll work in three stages and at the end of each stage the groups will share their results with others. The presenter will provide oversight and guidance during the exercises, answering questions and helping teams along (likely with an assistant).

  • My background on using XP, Scrum and Evo and how I became interested in this subject.
  • My assertion that currently there is a “gap” between current agile in practice and the needs of executive sponsors (demonstration of measurable business value)
  • Propose a method to “bridge the gap” using a combination of Evo and Scrum
  • Teach this method using examples (provided by me) and three stages of learning, applying lecture concepts to hands-on example.
  • Stage 1: Defining and Quantifying Business Value:
    • What is business value?
    • How do we define and measure it?
    • What is success?
    • What is failure?
  • Stage 1: Small group exercise:
    • Take the initial real-world goals and objectives from one of my projects (scrubbed of client-sensitive data) and derive the top few business objectives and how to measure them.
    • Define target and failure performance levels for each objective
  • Stage 2: Identifying winning design ideas for achieving objectives:
    • Identifying winning design ideas (aka strategies) for achieving target performance levels
    • Comparing design ideas along performance and cost dimensions to determine the strategy that represents the “best bang for the buck”
  • Stage 2: Small group exercise:
    • Continuation of example in stage 1; Small-group brainstorm to determine winning design ideas for achieving target performance levels
    • Group compares top few design ideas to determine which one represents the “best bang for the buck”, comparing estimated performance gain vs. resources required to implement
  • Stage 3: Turning the winning design idea into a Sprint Product backlog:
    • Creating a product backlog with user stories and tasks for implementing the design idea (including meters for measuring our performance against business objectives)
    • Creating an initial Sprint release plan with prioritized set of stories that make up first release
  • Stage 3: Small group exercise:
    • Using winning design idea from stage 2, create a Scrum product backlog of user stories and tasks to implement the design idea (including meters to measure our performance against business objectives)
    • Prioritize backlog and create first release plan that could be implemented in an 8-week release (4 x 2 week sprints)
    • Simulate a release being implemented then update the top objective metrics to reflect simulated results. Calculate business value delivered against objectives in first release
  • Present case studies where this approach has been used:
    • Fortune 500 Financial Services company
    • Fortune 1000 Insurance company
  • Summary, Resources and Q & A

Biography:

Ryan has been working with internet software since 1995, developing web-based solutions since 1997 and practicing agile since 2000, leading teams for the past five years designing, building and shipping products using agile.

His background is in software architecture, principally web-based transactional systems. More recently Ryan has been applying systems engineering principles to measuring business value, applying Evo methods with Scrum to create what he believes is an improvement on Scrum.

Ryan has presented on large-scale agile development at conferences (Scaling Agility - Agile Business Conference – London 2005) and on incorporating measurable business objectives with Scrum (Smart Decisions – Gilb conference London and No Fluff Just Stuff Virginia - 2007). Hecurrently focus on measure business value using Scrum and Agile Engineering practices for improving software and teams.