Choosing the integration model for Northbridge Insurance
Northbridge Insurance is a mid-sized general insurer with 2,800 staff, 14 business-critical IT services and five external providers. Network services are delivered by Provider A, the mainframe and data centre by Provider B, application development by Providers C and D, and the service desk by Provider E. Provider B has held its contract for 12 years and accounts for 40% of external IT spend. Contracts are staggered, with the first expiring in 18 months. The retained IT organization has 22 people, no formal service integration capability and no configuration management database. The board wants a 'single throat to choke' and a 12% reduction in external run cost. Provider B has offered to act as lead supplier, pricing its integration role 6% below the indicative cost of a standalone integrator. However, B caused three severity-1 outages in the last 12 months, and it competes directly with Providers C and D. The CIO must recommend a target SIAM model to the board within six weeks.
1. Evaluate whether Northbridge should adopt a lead supplier, an external service integrator or an internal service integrator. Justify your recommendation using the evidence in the setup.
Read the worked answer
The evidence does not support the lead-supplier offer as presented. The decisive issue is conflict of interest: Provider B would be accountable for end-to-end performance while also owning 40% of spend and competing with Providers C and D. That distorts prioritization, root-cause attribution and performance reporting. B's three severity-1 outages in 12 months also weaken its credibility as the party assuring service performance.
The 6% price advantage is small relative to lock-in and assurance risk. A standalone external integrator gives independence and brings capability, but adds a layer and cost while the retained organization is thin. An internal integrator would require recruiting process owners, tooling and a CMDB, which is unlikely within the first contract expiry.
A defensible recommendation is a hybrid path: appoint an external service integrator for 24-36 months, with a contractual transition option to an internal integrator once governance, data and process ownership mature. If the board prioritizes cost, a lead supplier could be considered only if B is excluded from the role, or if independent assurance, transparent reporting and strong conflict-management safeguards are contractually enforced. The choice should be driven by capability, conflict risk and total cost, not the headline 6%.
2. If the board insists on the lead supplier model, identify the governance safeguards you would require before contract signature and explain what risk each safeguard addresses.
Read the worked answer
Before signature I would require four safeguards, each tied to a specific risk.
First, independent service performance reporting: the lead supplier must publish end-to-end performance from a source the customer can audit, not only its own tooling. This addresses the risk that B reports green component SLAs while cross-provider handoffs fail.
Second, a conflict-of-interest register and recusal rules: B must declare where it competes with C and D, and must not chair prioritization or problem reviews affecting its competitors without an independent chair. This addresses biased allocation of work and blame.
Third, a right to audit and step-in: the customer can inspect B's integration records and appoint a replacement or independent assurance body if performance or transparency fails. This addresses lock-in and information asymmetry.
Fourth, multi-source data and exit rights: all providers must expose a defined minimum data set, and the customer must own or have access to service data on exit. This addresses the risk of B becoming the only party able to describe the service.
These safeguards do not remove the conflict, but they make it visible and governable.
3. Outline the first three actions the CIO should take in the 90 days after the model decision, regardless of which model is chosen.
Read the worked answer
First, establish a baseline of the current service chain: the top 10 business services, their providers, contract terms, SLA/OLA gaps, major incidents and handoff failures. Without this, any model decision remains theoretical. The baseline should include the 14 services and the three severity-1 outages, not a generic inventory.
Second, stand up transitional SIAM governance: a service integration board chaired by the CIO or delegated owner, with decision rights for priority-1 declaration, escalation and change authorization. Publish a single escalation route and require providers to use it. This begins to create 'one throat to choke' before contracts change.
Third, prepare the commercial and data foundations: a data-sharing schedule, minimum service and CI data model, and contract change notices for the first expiring contract. In parallel, start a CMDB reconciliation pilot on a bounded service. These actions are model-neutral and reduce the risk that the chosen integrator inherits an ungovernable estate.