Value Stream Mapping: the exercise that decides your automation budget
Before you choose a tool, you need a defensible number for what the current process costs. That is what a value stream map produces.
Most automation programmes are approved on a hunch. Someone senior knows the close takes too long, or that three people spend their mornings rekeying data, and a budget gets attached to that feeling. The build then delivers exactly what was asked for, and nobody can say afterwards whether it paid for itself.
Value Stream Mapping (VSM) exists to replace the hunch with a number.
What a value stream map actually is
It is a diagram of every step a unit of work passes through — an invoice, a claim, an order — from the moment it arrives to the moment it is done, annotated with two figures per step:
- Process time: how long the step takes when someone is actually working on it.
- Lead time: how long the unit sits before that happens.
The gap between the two is where the money is. In most mid-market back-office processes, process time is a small fraction of lead time. A five-day invoice cycle usually contains under an hour of actual work.
Why it comes before tool selection
Three things fall out of the map that no vendor demo can give you:
A cost per transaction. Total loaded labour cost across the steps, divided by volume. This is the number your CFO will hold you to, and it is the denominator for every ROI claim you make later.
The location of the constraint. Automating a step that is not the constraint produces no throughput gain at all. It produces a faster queue in front of the same bottleneck. The map shows which step is actually setting the pace.
A rework rate. Every loop-back on the map is a defect being corrected. Rework is the most commonly automated-over problem in operations: teams build a robot to do the broken step faster instead of fixing the input that caused it.
How to run one without stopping the business
You do not need a month. A workable map for a single process takes two working sessions:
- Walk the process with the people who run it, not with their managers. Managers describe the documented process. Operators describe the real one, and the difference between the two is usually where the waste lives.
- Count, don’t estimate. Pull 30 real transactions and time them. Estimates cluster around what the process is supposed to take.
- Mark every system boundary. Each hand-off between systems is a candidate for integration, and each one is a place where data is retyped.
- Put a dollar figure on each step before you leave the room. A map without costs is a diagram; a map with costs is a business case.
What good output looks like
A finished map should let a COO answer three questions in one sentence each: what does this process cost us a year, which step is holding it back, and what would it be worth to fix. If the map cannot answer those, it is decoration.
That is also the test we apply to our own assessments. The deliverable is not the diagram — it is the ranked list of what to fix, with a projected return attached to each item.
Working through this on your own process? The $4,500 Process Health Snapshot runs the mapping for you in five business days and ends with the ranked list.