GRC Implementation Roadmap: The First 90 Days
A week-by-week GRC implementation roadmap for the first 90 days, with owners, deliverables and done-when tests mapped to NIS2, DORA and NIST CSF 2.0.
A GRC implementation succeeds in 90 days when one complete loop runs: leadership has approved scope and risk appetite, the top risks have owners and scores, the controls that treat them are mapped to your obligations, evidence exists for the most important ones, and the first board report has produced a decision. It does not mean every framework is covered or every control tested.
This roadmap splits those 90 days into two-week phases with owners and a "done when" test for each. It assumes you are building cyber GRC under NIS2, DORA or both. For the full picture of what GRC in cybersecurity covers, start with our hub guide to GRC in cybersecurity.
What should be true on day 90?
Write the finish line before you start. A realistic day-90 state:
- Scope, regulations and frameworks in play are approved by leadership
- A risk appetite statement with numeric thresholds is approved
- One scoring method is in use, and the top risks are in a register with owners
- Existing controls are written once and mapped to every obligation in scope
- The highest-priority controls have defined evidence and a first test result
- A first board report has been presented and its decisions are minuted
If those six are true, you have a program to improve. If they are not, a longer plan would not have helped; the blocker is usually sponsorship or scope.
What do you need before day 1?
Three decisions save a month later.
A sponsor with authority. Someone at executive level who can approve appetite and settle disputes between teams. Under NIS2 Article 20(1), management bodies of entities in scope must approve and oversee the cybersecurity risk-management measures, so the sponsor role is partly a legal one.
A starting obligation. Pick the regulation or standard that binds you first and build the control library around it. Others get mapped onto it later.
A home for records. A spreadsheet, a GRC platform, or a document repository, but one place. Splitting the register across tools in week one guarantees reconciliation work in week ten.
Who should be on the implementation team?
Keep the core team small and the extended team named from the start:
- Sponsor: approves scope and appetite, and takes the week-13 report to the board.
- CISO or security lead: owns the plan, the timeline and the board report.
- GRC lead: runs the register, the control library and the evidence calendar.
- Risk owners: one per business area in scope, needed for the workshops in weeks 5 and 6.
- Control owners: IT, HR, procurement and facilities, needed in weeks 7 to 10.
- Legal or the DPO: confirms which obligations apply and reviews the appetite wording.
A weekly working session for the core team and a short fortnightly check-in with the sponsor are usually enough. Missing two sponsor check-ins in a row is the earliest warning sign that the plan will slip.
The 90-day GRC implementation roadmap
| Weeks | Focus | Deliverables | Owner | Done when |
|---|---|---|---|---|
| 1–2 | Scope and sponsor | Scope statement; list of regulations, frameworks and business units; named sponsor | CISO with sponsor | Sponsor signs the scope statement |
| 3–4 | Charter and appetite | GRC charter with decision rights; draft risk appetite with numeric thresholds | CISO, legal, risk | Management body approves the appetite statement |
| 5–6 | Risk method and top risks | One likelihood and impact scale with definitions; register of top risks with owners | Risk owners, coordinated by GRC lead | Every top risk has an owner, an inherent and a residual score |
| 7–8 | Control library | Controls written once, each mapped to every obligation in scope | GRC lead, control owners | No obligation in scope is left without at least one mapped control |
| 9–10 | Evidence and exceptions | Evidence definition and test frequency for priority controls; exception register | Control owners | Priority controls have a first test result; open exceptions have approvers and expiry dates |
| 11–12 | Suppliers and incidents | Critical supplier list with tiers; incident escalation and reporting path | Procurement, security operations | Critical suppliers are tiered; the reporting path is documented |
| 13 | First board report | Two-page report: risks above appetite, control health, decisions needed | CISO | The board records a decision on each item |
The order matters. Appetite comes before the register because a score means nothing without a threshold to compare it with. The control library comes after the register so that you map controls to risks you actually have.
How should each phase run?
Weeks 1–4: decisions before documents
Keep the scope statement to one page. List the regulations that apply, the frameworks you will use as structure, and the business units in and out of scope.
The appetite statement is the hardest deliverable of the whole plan, because it forces leadership to put numbers on tolerance. NIST CSF 2.0 subcategory GV.RM-02 expects risk appetite and tolerance statements to be established, communicated and maintained; DORA Article 5(2)(d) makes the management body of a financial entity approve the strategy including its ICT risk tolerance. Start with three thresholds, for example the residual score above which a risk needs board acceptance.
Weeks 5–8: risks first, then controls
Run short workshops with each risk owner instead of one large session. Score inherent risk, list existing controls, score residual risk. Our worked risk register entry shows the mechanics with real numbers.
Then write the control library. Write each control once in plain language, give it an owner, and map it to every obligation it serves. This is where a mature program saves most of its effort later, because one test can satisfy several frameworks.
Weeks 9–12: evidence, exceptions and the edges
Define, for each priority control, what evidence proves it ran and how often it is tested. Do not try to test everything. Pick the controls that treat risks above appetite.
Suppliers and incidents come last because they depend on the earlier work. NIS2 Article 21(2)(d) requires supply chain security, so tier critical suppliers and decide which assessment each tier gets, often a standard questionnaire such as the SIG questionnaire. For incidents, document the escalation path; under NIS2 Article 23(4), a significant incident needs an early warning within 24 hours and a notification within 72 hours.
Week 13: the first board report
Keep it to two pages: risks above appetite, control health, exceptions, and the decisions needed. The minutes of that meeting are the first real evidence that leadership approves and oversees the measures.
What should you leave out of the first 90 days?
- Every framework at once. Map one well; add others in the next quarter.
- Full control testing. Test the controls that treat your top risks first.
- Perfect taxonomies. A simple category list that people use beats a detailed one they ignore.
- Dashboards before data. A dashboard over an incomplete register misleads leadership.
- Integrations. Connect tools once the records are clean and owned.
How do you turn the roadmap into a GRC strategy?
Day 90 starts the operating rhythm. A workable year-one GRC strategy repeats four cycles:
- Quarterly: board report, review of risks above appetite, exception renewals.
- Per control frequency: tests on the calendar, results feeding residual scores.
- On change: new suppliers, systems and incidents enter the register as they happen.
- Annually: a minuted review of framework, appetite and policies. DORA Article 6(5) requires a financial entity's ICT risk management framework to be reviewed at least once a year, and NIS2 Article 21(2)(f) asks for procedures to assess the effectiveness of the measures.
Add a second framework in the second quarter, once the first mapping is stable.
Which GRC maturity model should you use?
Use the one you can evidence. The NIST Cybersecurity Framework 2.0 defines four Tiers for the rigor of risk governance and management: Partial (Tier 1), Risk Informed (Tier 2), Repeatable (Tier 3) and Adaptive (Tier 4).
A completed 90-day roadmap typically moves an organization from Tier 1 to Tier 2: the register exists and leadership sees it. Tier 3 needs the repeating cycles above to run for a year, because "repeatable" can only be shown by repetition. Our hub guide includes a table of observable signs for each Tier.
Do you need a GRC tool from day 1?
No. A spreadsheet is fine for the first register and control list. The case for a platform comes when the links matter: a failed control should change a risk score, an approval should leave a record, and evidence should not live in email.
ResilientX GRC software imports existing Excel registers, asset inventories, vendor lists and custom frameworks, and ships cross-mapped NIS2, DORA, ISO 27001 and GDPR libraries. Its workflow module assigns owners and deadlines, sends reminders, routes policies, exceptions and risk treatments through approvals with a recorded decision, and keeps the history for auditors. See ResilientX GRC in action if you want the roadmap's records in one place from week one.
Frequently asked questions
How long does GRC implementation take?
A first working loop takes about 90 days if a sponsor is engaged and the scope is limited to one starting obligation. A repeatable program takes at least a year, because the annual review and four quarterly board cycles are part of what makes it repeatable. Expanding to more frameworks or business units adds time in proportion to the controls involved.
Why do GRC implementations fail?
Most fail on decisions, not tooling. The common pattern is a project that builds registers and dashboards without an approved risk appetite or an executive sponsor, so nothing it produces changes a decision. The second pattern is scope that covers every framework at once and never finishes the first mapping. Both are avoided by the first four weeks of this roadmap.
Should we implement ISO 27001 or NIS2 first?
Implement against whichever binds you, and use the other as structure. Many European companies use ISO/IEC 27001 as the control structure and map NIS2 Article 21 onto it, because the directive lists measures while the standard provides a management system. Mapping once means a single control test can serve both. For the basics of the three letters, see what GRC is.
The short version
Ninety days is enough for one full loop: approved scope and appetite, owned risks, mapped controls, first evidence, and a board decision. Leave breadth for later, keep the order, and measure progress with "done when" tests instead of documents produced.
See ResilientX GRC in action — book a demo and bring your current register to the first session.