A pilot is designed to answer questions. Does the scanner see the right assets? Do findings reach the people who can fix them? Is the workflow tolerable for the teams involved? When the answers are positive, the pilot has done its job, and the next step is to extend it across the estate. That step is where many programmes falter. The pilot team was small and committed. The production environment is larger, less familiar and owned by many people with competing demands.
This guide describes how to move from a successful pilot to a production service that continues to work. It focuses on the operational changes that a pilot rarely tests, including ownership, scale and the routines that keep findings moving.
What a pilot does not prove
Before expanding, be clear about what the pilot has shown and what it has not. A pilot usually covers a representative segment, a limited number of teams and a controlled set of systems. It often receives more attention than the normal service will. Its results may therefore overstate how the programme will behave at scale.
It is useful to list the assumptions the pilot relied on. Examples include the availability of a particular team, the presence of accurate asset records for the pilot segment and the willingness of system owners to respond quickly. Each assumption should be checked against the wider estate before the rollout is planned.
Define what production means
Production should have a written definition. It should state which assets are in scope, who owns the service, what the service level is for scanning, triage and remediation tracking, and how the service will be measured. Without this definition, the programme can drift into a state where some teams participate and others do not, and nobody can say whether the programme is complete.
The definition should also state what is out of scope, at least for the first phase. Trying to cover every asset type, every environment and every tool at once is a common cause of failure. A deliberate, phased scope is more likely to succeed.
Plan the rollout in waves
Rollout works best in waves, each covering a group of systems with similar characteristics. A first wave might include well-understood systems owned by a responsive team. Later waves can bring in more complex estates, legacy systems and areas where the relationship with the owner is less established.
Each wave should have entry and exit criteria. Entry criteria might require an accurate asset list, named owners and an agreed scan schedule. Exit criteria might require that findings for the wave are being triaged within the agreed period and that the owners have confirmed the workflow works for them. Moving to the next wave only when the criteria are met prevents problems from accumulating silently.
Transfer ownership deliberately
A pilot is often run by the team that designed it. Production must be run by people who will be there in a year. Plan the transfer of ownership early. Identify the operational owner of the service, the people who will triage findings and the system owners who will respond. Give them time to learn the workflow before the pilot team steps back.
Documentation matters here, but it is not enough on its own. Runbooks should describe how to respond to common situations, such as a scan that fails, a finding that appears to be a false positive or a request for an exception. Practice those situations during the handover so that the new owners have experienced them before they occur in earnest.
Make the workflow hold under load
A workflow that works for twenty findings a week may fail at two thousand. Test the workflow with realistic volumes before the rollout reaches full scale. Look at how findings are grouped, how tickets are created and closed, and how exceptions are recorded. Check whether the people who must act on findings can see what they need without searching through lists.
Load also affects the scanner. Confirm that scan schedules do not overlap in ways that create congestion, and that scan windows agreed with system owners are respected. Keep an eye on the time taken to complete a full cycle, because that time tends to grow as the estate grows.
Keep the discipline the pilot had
Pilots often succeed because of close attention. Production risks losing it. The routines that made the pilot work should be written down and kept. These typically include a regular review of new findings, a fixed point each week for escalating overdue items, and a clear route for disputes between the security and operations teams.
Verification is a routine that is particularly easy to drop. In a pilot, every fix is usually checked. In production, the volume can make checking feel burdensome, and fixes are closed on trust. Keep verification in the process, and sample it if full checking is impractical.
Expect and plan for friction
Friction is normal. Some system owners will object to scanning. Some will dispute findings. Some will ask for exceptions that the programme cannot justify. The rollout plan should anticipate these situations and assign responsibility for resolving them. A simple escalation path, agreed in advance, prevents disagreements from stalling the whole wave.
Friction can also be a useful signal. Repeated disputes about a particular type of finding may indicate a problem with the scanner's accuracy or the clarity of the reports. Treat recurring objections as feedback on the service.
Measuring the transition
Track the rollout as well as the service. Measure how many assets have been brought into scope, how many have named owners and how many findings have been resolved within the agreed period. Compare these figures against the plan for each wave. Where progress is slower than planned, investigate the cause before adding resources.
Review the lessons from each wave before starting the next. What worked, what caused delay and what assumptions proved wrong? Recording those answers turns the rollout into a learning exercise rather than a series of separate efforts.
Common mistakes
The most common mistake is declaring success at the end of the pilot and expanding the scope all at once. A second is assuming that the pilot team will continue to run the service without a planned handover. A third is neglecting the people who will receive the findings. Their capacity and their understanding of the process determine whether the service is useful.
Next steps
Write the production definition and identify the first wave. Confirm the entry criteria for that wave, name the operational owner and agree the routines that will continue after the pilot team steps back. Schedule a review at the end of the wave to decide whether the next wave can begin.
Frequently asked questions
How long should a pilot run before moving to production?
Long enough to cover normal operational cycles, including a full round of triage, remediation and verification. The length depends on the size of the segment and the complexity of the workflow, rather than on a fixed period.
Should every system be brought into the service at once?
Usually not. A phased rollout by wave allows problems to be found and fixed while the scope is still manageable.
What is the most common reason production rollouts stall?
Unclear ownership. When nobody is accountable for responding to findings on a system, the findings accumulate and the service loses credibility.
Related reading
If you are working through this topic, these related articles cover neighbouring questions:
