The ReleaseTEAM Blog: Here's what you need to know...

How to Map Your Value Stream

This is Part 3 in our Value Stream Mapping series. Did you miss Part 1 or Part 2? A value stream is anything that delivers a product or service. You may certainly map the entire software delivery life cycle, from the first feature idea to deployment, or you can set narrower start and endpoints (called “fence posts”) to focus on a couple of areas more closely.

Identify the people and teams involved in the value stream you’ve selected. Have you accounted for business stakeholders, developers, testers, operations team, security engineers, and systems architects?
Agree on what’s in scope and what’s out of scope for your value stream mapping exercise.
Working together, identify each step in the value stream. You can complete this step with sticky notes on a whiteboard, virtual boards, or various tools out there. As you discuss each step in the value stream, make sure the team understands their chronological order. Some steps may be done in parallel rather than sequentially. If you aligned your value stream steps to the SDLC, your steps might look like:
Determine how much time it takes to complete one unit of work for each step (the Process Time metric from Part 2 in this series). It’s much easier to compare steps and visualize the roadblocks if you choose one unit of time for measurements across all of the steps – whether that is minutes, hours, days, or weeks.
After agreeing on the process time for each step, determine the wait time between steps. (Wait time before the current step + Process Time for the current step) = Lead Time for the current step.
Add all of your process time and wait time to get your cycle time (or Flow Time in flow metrics). This is your baseline before you start eliminating roadblocks and inefficiencies.
Identify the teams associated with each step.
Is each step a value add to the customer? Are there any steps that are less important or can be removed to speed up time to value?
Identify the amount of rework that happens in each step. Are downstream steps receiving accurate workpieces, or are they sending a high number back due to errors?

Now that you have your initial value stream map, which steps are the least efficient or have high percentages of rework? These are good candidates for closer examination and modification to improve accuracy and productivity. Use this information to create a “future” value stream map that you can implement to improve quality and cycle time.