A few weeks before go-live, dozens of change requests arrive, and each one looks urgent. Every approved change request takes time away from the critical path, and the project cannot afford a slow approval process. What you need is a short cycle, with declared turnaround times and few categories, that also prepares the handover to application management.
What a change request is and why it must be governed
In a long project the functional scope is defined by approved documents. A change request (CR) arises when something falls outside that scope or changes after approval: a requirement that emerged during testing, a new regulation, a business request. Without governance the number of CRs grows, estimates are inconsistent and dates slip. Governance exists to decide what goes in, when and at what cost, with risks and opportunities in plain view.
The types of CR
| Type | Origin | Particularities |
|---|---|---|
| Out of scope | requests outside the scope, or that emerged after the functional documents were approved, including during testing | handled with ad hoc documentation or added to documents that are still open |
| Adaptation to be applied on several systems | regulations or business needs that must be implemented in parallel on the new system and on the legacy one (or on third-party systems) | needs a cut-off date for opening, with an exception for regulations whose deadline falls before go-live |
| Service enhancement | arising from production of the previous system or from the management service | assessed case by case whether and when to handle it on the new system |
A simple taxonomy has one advantage: every CR has a known path and a known decision-maker.
The life cycle
A lean process has six steps, with indicative turnaround times declared in working days:
- Logging: the requester opens the CR, stating explicitly whether it is necessary for go-live.
- Qualification (a few days): the supplier checks that the information is complete; if not, it asks for clarification and possibly organises a meeting with the functional contact.
- Estimate (about ten days): an outline solution, an estimate in person-days with its rationale, whether it can be taken on before go-live, and an update of the register.
- Approval: the client assesses the estimate and the real need, the risk, and the impact on the business and on the plan.
- Planning: if approved, the CR enters the plan (analysis, development, test); for adaptations, the decision to proceed is communicated to the other suppliers involved.
- Execution and tracking against requirements monitoring.
A CR committee meeting frequently (for example every two weeks during the project, weekly in AMS, the application management service, for each area) decides and updates the online register. Few owners, a single register, short response times.
Statuses and the "necessary for go-live" criterion
A useful register uses macro-statuses: to be estimated, awaiting approval, approved, included in live / live, backlog (to be handled after go-live, awaiting clarification), cancelled / rejected. For the necessary for live field there is one rule: yes only if a business process is missing, there is a legal obligation with a deadline before live, or its absence creates an economic or compliance risk that cannot be handled with a workaround. A "nice to have" CR goes to the backlog.
The catalogue of interventions, with estimate, status and classification, is the basis for priority decisions: it shows how much effort is already approved and how much is still waiting. Be careful: estimates of CRs not yet analysed carry wide uncertainty and must be marked as such.
Handover to application management (AMS)
AMS does not start on go-live day: it has to be prepared with a set-up plan that defines scope, resources, tools, access and rules.
Operating model. The language of ITIL is used: incident (restoring the service), service request (standard request), problem (root cause), change and release. Levels of competence: L1 intake and triage, L2 configuration and analysis, L3 development.
Backlog and reports. A single backlog per area, clustered (dependencies, problems, processes affected, workarounds). A weekly report (for example on Mondays) with the trend of the total backlog and per area, tickets closed against resolved, critical points and initiatives.
Release calendar. A fixed cadence of deployments, with testing of fixes before release, release notes completed before every release and communicated to the client. Manual clean-up activities are temporary: they must be listed, with a deadline, and replaced by fixes.
Incidents and service requests. For incidents: intake, analysis, assignment to configuration or development, planning, execution, UAT with outcome, release notes, verification and closure. For service requests: triage, planning, execution, request for authorisation if working in production, post-execution check.
Escalation. Define, by priority, the times for intake and resolution, the first- and second-level contacts, and when and who declares a major outage.
Periodic governance. Area meetings (operational, weekly or fortnightly), monthly progress reviews by domain (credit, billing, CRM, measures), a service review and a business review with KPIs.
Recurring regulatory changes as a process
ARERA (the Italian energy regulator) and the legislator produce changes very often: updates to charges, new billing rules, information obligations, tax and excise news. A stable process includes:
- monitoring of publications (a regulatory calendar) with an owner;
- impact analysis on processes, price components, communication templates, integrations with the SII (the central market data hub) and the SDI (the Italian e-invoicing exchange system);
- classification: configuration (parameters, tables) or development;
- effective date as a planning constraint, with regression testing;
- double application when several systems (the new and the legacy one) coexist: it must be declared who adopts what;
- traceability: reference document, ticket, release notes.
Common mistakes
- CRs opened without saying whether they are necessary for live.
- Estimates with no rationale and no owner.
- A committee that meets rarely: requests pile up.
- A separate backlog per team: the client does not see the whole picture.
- Releases outside the calendar without release notes.
- Treating a regulatory adaptation as an exception instead of as a cycle.
Frequently asked questions
Who decides whether a CR is necessary for live? The client, on a reasoned proposal from the requester and with the supplier's opinion on the planning risk.
How long should the estimate phase last? As long as it takes to produce a credible rationale; an operating reference is about ten working days.
Is a regulatory adaptation always a CR? If it is already covered by the service contract it is an ordinary enhancement; otherwise it is a CR with the constraint of the effective date.
Which AMS KPIs? Backlog and trend, intake and resolution times, compliance with service agreements, successful releases, repeated incidents.
Regulatory references
- ARERA, measures and regulatory calendar (resolutions section): https://www.arera.it
- ARERA Resolution 463/2016/R/com, TIF (Annex A), as an example of a consolidated text subject to periodic updates: https://www.arera.it/atti-e-provvedimenti/dettaglio/16/463-16
- Regulation (EU) 2016/679 (GDPR), art. 28 (processor) and art. 32 (security) for the AMS supplier's access to production data: https://www.garanteprivacy.it
- Legislative Decree 196/2003 (Italian Privacy Code), as amended by Legislative Decree 101/2018: https://www.normattiva.it
- ITIL framework: a good-practice reference, not a regulation.
This article is an OIAR skill
What you have just read is the content of a skill from the OIAR catalogue, written for people. Knowledge like this, but richer and deeper, written for the agents, drives the interviews, the analysis and the report inside the platform. Skills are used in OIAR: to try this one on a project of yours you need an account.
change-governance-and-ams-handoverMethod for governing change requests and starting application management (AMS) in a CRM and billing project for an energy retailer: request types, a short cycle with declared turnaround times, handover to the service after go-live. Use it to define or review the change process and the handover to AMS.