Four Parties, One Value Stream, No End-to-End Owner
A retail bank's online account-opening value stream involves four parties: an internal digital team, a core banking platform team, a KYC vendor, and a document-checking BPO. Each reports its own improvement metrics: the digital team counts story points delivered, the platform team tracks change success rate, the vendor tracks SLA compliance, and the BPO tracks cases per agent. End-to-end, the journey from application to active account takes 6.2 days against a customer expectation of same day, and 31% of applications are abandoned before completion. Each party keeps its improvement register to itself. The quarterly service review examines each supplier separately, and no one is accountable for the end-to-end journey. The bank has just appointed a value stream owner reporting to the COO, with a mandate to 'make account opening same-day' but no authority over supplier contracts, which run for another 18 months. The value stream owner asks you how to structure continual improvement.
1. Explain why continual improvement at each party has not improved the end-to-end value stream, using ITIL 4 concepts.
Read the worked answer
ITIL 4 frames value as co-created through value streams in which multiple parties - including suppliers - contribute. Here, each party improves its own part of the chain and reports output measures that are, by design, blind to the handoffs between them. The digital team optimises story points, the platform team optimises change success, the vendor optimises SLA compliance, the BPO optimises cases per agent. Each can hit target while the end-to-end journey stays at 6.2 days and 31% of applications are abandoned. This is suboptimisation: local efficiency at the expense of the flow of the whole.
The 'improve' value chain activity is being performed within silos rather than across the value stream, and the four dimensions are misaligned: partners and suppliers are managed contractually, value streams and processes are not mapped end to end, and information is not shared. The quarterly service review reinforces the problem by examining suppliers separately and reviewing inputs rather than customer outcomes. Nothing in the current arrangements rewards a party for improving a handoff it does not own, and no one is accountable for the customer's 6.2 days.
2. Propose how the value stream owner should organize continual improvement across the four parties, including roles, cadence and metrics.
Read the worked answer
The value stream owner needs a federated structure rather than a new central team. Roles: the owner is accountable for the end-to-end outcome to the COO; each party nominates an improvement champion with authority to commit that party's effort; a joint improvement board meets monthly and includes operational leads, not only account managers. One shared improvement register - or a shared view of four registers using common IDs and fields (baseline, outcome, owner, handoff affected) - prevents duplication and makes cross-party items visible. The bank's CI practice defines the method; each party applies it.
Metrics should be end-to-end and shared: lead time from application to active account (median and 85th percentile), abandonment rate, percentage complete and accurate at each handoff, touch time, first-time-right rate, rework and customer effort. Each party keeps internal measures but reports its contribution to these. Cadence: fortnightly flow review of the value stream, monthly joint board, quarterly renegotiation of scorecards. Contracts need shared objectives and gain-share at renewal, with a pilot clause if possible. The owner should report the same dashboard to the COO and to suppliers, so there is one version of the truth.
3. What should the value stream owner do first, given no contractual authority over the parties?
Read the worked answer
Start where you are. Before changing contracts or governance, make the value stream visible with data the parties already hold: reconstruct 50–100 recent applications, measure time in each stage and handoff, and count where rework occurs. Presenting that map is usually more persuasive than a mandate, and it turns a blame conversation into a flow conversation. In parallel, build a coalition with the operational leads in each party - they often see the handoff problems and can act without waiting for legal change.
Then run one bounded improvement using the continual improvement model: choose a segment (one product, one branch or one application type), agree a shared outcome such as reducing abandonment by five points, set a baseline, and improve the worst handoff over six to eight weeks. Use existing forums for visibility; do not wait for the quarterly review. Escalate to the COO only for decisions the owner cannot make. At contract renewal, convert proven joint working into shared objectives and gain-share. Collaborate and promote visibility, and progress iteratively with feedback, are the principles that fit a mandate without authority.