Value Stream Management Roadmap: From First Map to a Managed Flow

Disclosure: M365.FM host Mirko Peters works at iFAKT GmbH. This page describes the company factually; paid partnerships and links are labelled as advertising.

By Mirko Peters · 2026-11-11

A value stream management roadmap in five phases: pick one stream, map it, plan 30-60-90 days, review weekly, then scale by revenue share. Includes an example.

A value stream management roadmap is the sequence of steps that takes a plant from its first value stream map to a flow that someone owns, measures and improves every week. It has five phases: choose one stream, map it, schedule the first 90 days, run a weekly review, and then scale to the next streams. This article is for operational excellence managers who must show management what happens in month one, month six and year two. It gives the phases, a calendar, a worked example and the mistakes that stall a lean transformation.

Key points

• A value stream management roadmap moves from one pilot stream to a managed flow in five phases: choose, map, plan, run and scale.
• Mapping one product family takes about two days; the work after the map decides whether anything changes.
• A Lean Enterprise Institute contributor advises short-term wins inside a longer vision of 18 to 24 months, and 30-, 60- and 90-day implementation plans.
• Scale by revenue share: one practitioner targeted streams for 50 percent of revenue within a year and 80 percent after that.
• Redraw the future-state map when the stream reaches it, and review the annual programme against the same measures every quarter.

Why does a roadmap help?

VSM, short for value stream mapping, is a lean technique, and a mapping exercise on its own is easy to forget. A roadmap turns it into a programme. It keeps the end-to-end value flow in view, so each step adds value and helps to reduce waste where cycle time and lead time show it. It also stops you from wasting effort on process mapping that nobody uses.

A VSM helps teams see the value flow: which steps are value added and which are waiting. The mapping exercise with current and future state maps is the lean technique that makes this visible. Adding value to a product or service means doing the steps the customer pays for. Agile teams use the same words, deliver value, for software. In both cases, process improvement starts with state maps and a clear workflow.

In lean manufacturing, such a programme helps you optimize the whole flow, not one process. Teams in software delivery and software development know the same pattern, with a pipeline and a development process in place of a plant. The logic of VSM tools and process flow charts is similar. Keep the first map simple, and expect to improve it with each review: a flowchart with a few symbols on paper is enough for the first stream.

What does the roadmap look like?

A VSM roadmap needs a clear unit. Use the product family. The Learning to See workbook describes four steps for each family: select the family, draw the current state, draw the future state, and prepare and use a work plan. Management adds the routine around these steps.

The five phases of the roadmap:

1. Choose. One product family, one owner, baseline measures.
2. Map. Current state and future state, drawn by the people who work in the stream.
3. Schedule. Implementation steps at 30, 60 and 90 days.
4. Run. A weekly review, a visible board and a map that stays current.
5. Scale. The next families, the annual improvement plan and the link to company targets.

The phases overlap. The workbook notes that future-state ideas often appear while you map the current state. Treat the phases as a guide, not as gates.

Phase 1 and 2: how do you choose and map the first stream?

Pick the family that matters most for customers or revenue. Aaron Hunt, in an LEI article on low-volume, high-mix mapping, reports that significant growth began once streams representing 50 percent of revenue were defined and improved. He targeted about one year for that phase. That is one practitioner's experience, not a rule.

Name an owner and record baseline measures before you draw anything: lead time, on-time delivery and inventory are enough. Without a baseline, you cannot prove progress later.

Then map. The workbook says that mapping one product family takes about two days to reach the point where implementation can begin. The same workbook calls the future-state map the most important one. The LEI ebook on value stream mapping adds a warning: do not jump straight to point kaizen, and agree on the future state across functions first.

Phase 3: how do you structure the first 90 days?

The LEI ebook, Insights on value-stream mapping, quotes Judy Worth with practical advice: aim for short-term wins inside a longer vision about 18 to 24 months out, start rapid learning cycles within two to three weeks of the mapping, and use 30-, 60- and 90-day implementation steps. She also advises experiments before you roll out a solution.

A simple structure:

• Days 1 to 30. Start the weekly review, fix the simple flow problems found on the map, and run the first experiment.
• Days 31 to 60. Roll out what worked. Start the larger change, such as a pull signal or a smaller batch size.
• Days 61 to 90. Compare lead time, delivery and inventory with the baseline. Decide whether to extend the pilot.

Mike Orzen, writing for LEI about the value stream manager, also suggests a 90-day trial with results compared to baseline measures. Together these give you a ready review point for management.

Phase 4: how does the stream become managed?

A managed flow has three habits:

• A weekly review with the owner and the function leads, using the same measures every week.
• A map that stays current. When the stream reaches the future state, draw a new future-state map. The workbook says so.
• Two kinds of improvement. The workbook separates flow kaizen, which improves the whole stream and is led by management, from process kaizen, which removes waste inside a process. Both belong on the plan, and the owner keeps them aligned.

Software teams that practise DevOps know the same idea as a continuous cycle of measure and improve. In a plant the data come from orders and machines. The routine matters more than the tool.

Phase 5: how do you scale to the whole plant?

Scale by revenue share, not by convenience. Rank the families by revenue and, if you can, margin. Add one stream when the previous one holds its results for a quarter. Hunt's milestones are 50 percent of revenue within a year and 80 percent after that, as a reference point.

Build the annual improvement programme from the streams. For each stream, list the main flow changes, the owner, the quarter and the expected effect on the measures. Review it each quarter against the actual trend of your VSM measures. This is also the point where the roadmap connects to continuous improvement as a company habit, not as a project.

Example: a four-stream roadmap with revenue shares

This example is invented. The shares and dates are not from a real plant.

A plant has four product families. Revenue shares are 40 percent for A, 30 percent for B, 20 percent for C and 10 percent for D. The roadmap:

• Months 1 to 3. Stream A: owner named, baseline set, map drawn in two days, 90-day schedule, weekly review. A covers 40 percent of revenue.
• Months 4 to 6. Stream A moves into steady review. Stream B starts. Cumulative coverage is 40 plus 30, or 70 percent.
• Months 7 to 12. Stream C starts. Cumulative coverage is 70 plus 20, or 90 percent. The plant passes the 50 percent milestone in month 4 to 6, well inside one year.
• Year 2. Stream D gets a light review. A and B draw a new future-state map. The annual programme runs for all four streams.

The owner of A trains the owner of B. That keeps skills inside the plant and avoids waiting for a consultant. At each quarter, the plant manager looks at three measures per stream: lead time, on-time delivery and inventory days.

Common mistakes

• Starting with the whole plant. One stream teaches the method.
• No baseline. Without numbers from before, no result is believable.
• Actions with no owner. Every action needs a name and a date.
• Jumping to point kaizen. Improvements without a shared future state fix one process while the stream stays stuck.
• Forgetting to redraw the map. An old map loses trust.
• Scaling too fast. A second stream before the first one holds creates two half-managed flows.

How software can help

A roadmap with ten streams needs a shared model. Maps drawn by hand in different formats are hard to compare, and the data behind them goes stale. Software that builds the map from ERP data, and lets several people edit it, keeps the current state close to reality and saves the redraw work.

iFAKT's Polaris VSM is a module of iFAKT Polaris for data-based value stream mapping. It models material and information flows collaboratively and makes waste and bottlenecks visible. Test any tool on the pilot stream first.

Next step

A value stream management roadmap starts small: one stream, one owner, one baseline and a 90-day schedule. This week, pick the family with the most revenue, write down its baseline measures and book the two days for the first map.