Carer applying for a parent
No doc covers applying on someone else's behalf. The form assumes the applicant is the account holder.
Service and CX design
Your blueprint shows the journey you designed. Subjourney shows the one people actually get, by sending simulated personas through every channel, handoff and policy your service depends on.
The journey map looks clean on the wall. But the real service lives in a dozen other places: the website copy, the call-centre script, the eligibility policy, the email templates. Nobody has walked all of them together.
So the first time the full journey gets tested end to end is when a customer tries it. And the first you hear of it is a complaint.
The broken handoff.
The web flow says “we'll call you within two days”. The contact-centre SOP has no step for it.
The persona you didn't design for.
The carer applying on someone else's behalf, who the form assumes doesn't exist.
The channel contradiction.
The app says one thing, the policy says another, and the customer can't tell which to trust.
The dead end.
A path where the only next step is “contact us”, and contacting us leads back to the start.
| Today | With Subjourney | |
|---|---|---|
| Journey validation | Workshops and walkthroughs on the wall | Personas walk every touchpoint against the real docs |
| Edge-case customers | Covered if someone in the room remembers them | Run as personas, every time |
| Cross-team handoffs | Discovered when a customer falls through | Each broken handoff named, with the doc on each side |
| Research personas | Presented once, then shelved | Run against every change to the service |
| After launch | Complaints tell you what broke | Re-runs tell you before customers do |
Journey validation
Edge-case customers
Cross-team handoffs
Research personas
After launch
In practice
A service design team imports a benefits application journey: the web form, the eligibility policy and the call-centre script. A persona who recently changed jobs stalls at income verification. The policy asks for “current income”, the form asks for last year's, and the script tells staff to accept either. Three docs, one fix, found before launch instead of in the complaints log.
A run, step by step
01Import.
The team brings the online application form copy, the eligibility policy, and the contact-centre script for application queries.
02Model.
Subjourney builds the journey.
It flags that Call for status has no step for applications still in verification.
03Run.
Four simulated personas set off. Choose one to see what they want and what gets in their way.
CarerAuthorised representative
Applying on behalf of a parent.
04Findings.
Carer applying for a parent
No doc covers applying on someone else's behalf. The form assumes the applicant is the account holder.
Changed jobs two months ago
The policy asks for current income (Section 3.1). The form asks for last financial year. The script tells staff to accept either.
Calls mid-way to check progress
The script covers submitted and decided applications only. In-verification callers have no next step.
05Fix and re-run.
Policy, digital and contact-centre owners each get the finding for their doc. After the fixes, all four personas reach a decision.
Three teams, three docs, one journey that finally lines up.
Personas to start with
No. It makes research go further. Run your research personas against every version of the service, and save real sessions for what only real people can tell you.
That's where it's most useful. Bring the docs from each team, and a run shows where they don't line up.
No. Start with the docs behind the service. Subjourney builds the journey from them, and the gaps show you what the blueprint is missing.
Each finding names the step, the persona and the paragraph. Send it to whoever owns that doc, with the evidence attached.