New Service Development Model

The New Service Development model runs the same discipline as product development but with fifteen detailed stages, from setting service objectives through design, testing, piloting and post-launch review, because a service is delivered live and can't be inspected before it's sold.

Fifteen small stages march across the page, far more granular than a typical product flow.

Scheuing and Johnson's fifteen stages, from setting objectives to reviewing the launch.

Reach for this when…

You want to add a service line but have only ever launched physical products.
A new service idea sounds good in the room and falls apart with a real customer.
You keep launching services without ever piloting them on a small group first.

How to run it

Set clear new-service objectives before generating a single idea.
Generate and screen ideas, then develop and test the concept with real customers.
Run the business case and get formal project authorisation before building anything.
Design and test the service, its delivery process and systems, and the market programme, and train the people who'll deliver it.
Pilot, test-market, then launch full-scale, and review what actually happened afterwards.

A worked example

Situation. Camila Fuentes ran ReVital Fisioterapia, a three-site physiotherapy clinic in Valparaiso, Chile, and wanted to add a telehealth follow-up service but had never designed a service from scratch before.

Applied. She blueprinted the whole telehealth journey, booking, video call, exercise plan hand-off, follow-up message, then piloted it with fifteen existing patients before opening it to everyone.

Result. The pilot caught a gap: patients wanted a printed exercise sheet, not just a video, before they'd trust the session was 'real'. That one fix, added before full launch, kept telehealth no-shows below in-clinic rates.

Vitalidad's pilot run caught the fix a full launch would have missed.

The catch

Because a service is delivered live by a person, no blueprint fully controls for a bad day, a distracted staff member, or a confused customer, so consistency is harder to engineer than in a factory. Piloting on a small, friendly group can also hide problems that only show up at scale with a less forgiving audience.

If your blueprint has no step where a human being has to use judgement, you haven't blueprinted a service, you've blueprinted a form.

Origin: Eberhard E. Scheuing and Eugene M. Johnson

connecteddale