Identifying Value Streams
This domain tests whether you can find and bound value streams correctly - the right customer, trigger, start, and end - before any mapping begins. Most mapping failures are actually identification failures.
What makes something a value stream
A value stream is the sequence of value-adding and non-value-adding activities required to deliver a product or service to a customer, from a defined trigger to a defined exit. Four elements must be nameable: the customer (external or internal), the trigger event, the start and end points, and a measurable lead time across the whole flow. If you cannot name the customer and the trigger, you have described a function, not a stream.
Operational and development streams
Karen Martin separates operational value streams, which repeatedly deliver an existing product or service, from development value streams, which bring something new into existence. A software organization usually has both: idea-to-production (development) and incident-to-resolution or order-to-cash (operational). Streams can also be classed as primary (they serve the paying customer) or supporting and enabling (they serve other streams, such as HR or finance).
How to identify them
Work backwards from the customer; ask what outcome they receive and what they pay for. Enumerate distinct triggers - an order, a claim, a feature request, an incident - because each distinct trigger usually implies a distinct stream. Follow the real work rather than the org chart. Then set the scope: a stream you can map in a day or two and act on. Enterprises typically have more primary streams than they expect, often a handful to a couple of dozen; dozens per team signals over-splitting.
Worked example
An insurer drafts candidates: quote-to-policy, claim-to-payment, premium-to-collection, and idea-to-release. A manager proposes an underwriting value stream. Applying the four elements shows it fails: underwriting is a department, it has no external customer of its own, and it has no trigger or exit that delivers value. Quote-to-policy passes on all counts, so it is kept; underwriting becomes a step inside it.
Traps that cost marks
- Naming streams after departments or systems.
- Creating one stream per team or per feature.
- Forgetting internal customers and supporting streams.
- Assuming a product and a value stream map one-to-one; one product can be served by several streams, and one stream can serve several products.