The Difference Between Monitoring a System and Actually Being On Call for It · Midcore Operations
Midcore Operations
Cloud & Infrastructure

The Difference Between Monitoring a System and Actually Being On Call for It

A dashboard that turns red is monitoring. Visa's own network exists partly because that is not the same as someone actually responding in time.

Published
21 October 2026
Reading time
9 min read
Author
Cloud & Infrastructure practice
Topic
Cloud & Infrastructure
Key takeaways
  • NIST's current incident response guidance, SP 800-61r3, separates Detect and Respond into distinct functions within its CSF 2.0-based model. Possible incidents being found and analyzed is a different activity from actions being taken regarding a detected incident, and a system can do the first well while nobody is positioned to do the second at all.
  • NIST's guidance on incident response roles describes incident handlers as on staff, on contract, or available when needed. Each of these is a real, specific commitment. Having a monitoring dashboard is none of them, because a dashboard does not obtain, contract, or commit anyone to respond.
  • Visa Core Rules Section 7.3.3.1 requires a Member to provide authorization services 24 hours a day, 7 days a week, with an issuer or its authorizing processor responding to requests in an average time not exceeding 5 seconds per calendar month in the US region.
  • Visa's Stand-In Processing exists specifically for the gap between monitoring and response: it activates when an issuer, its processor, or a scheme processor is unavailable or exceeds the maximum response time, which is exactly the scenario where a system was being watched but nobody was actually there to answer in time.

The difference between monitoring a system and actually being on call for it is the difference between knowing something is wrong and being the specific person committed to fixing it within a bounded time. NIST’s own current guidance on incident response separates these into distinct functions for exactly this reason. In payments specifically, the gap between the two has a name and a real consequence: Visa built an entire network component, Stand-In Processing, around the fact that a monitored system and a system someone is actually responding to are not the same thing.

What NIST’s current model actually separates

NIST Special Publication 800-61 Revision 3, the current version of the federal government’s incident response guidance, organizes incident response around the six functions of the Cybersecurity Framework 2.0. Two of those functions map directly onto the distinction this article is about. Detect covers possible cybersecurity attacks and compromises being found and analyzed. Respond covers actions regarding a detected incident actually being taken.

These are explicitly described as separate functions, not two names for the same activity. A system can be monitored well, generating accurate alerts the moment something goes wrong, while the organization running it has no equivalent strength on the Respond side. NIST’s guidance treats this as a real failure mode worth naming specifically, not an edge case. Detection without a committed response path is a system that knows it is broken and has no mechanism guaranteed to do anything about that fact within any particular window of time.

Why “we have monitoring” and “we have on-call” are different claims

NIST’s guidance on incident response roles makes the distinction concrete by describing what a real response commitment actually looks like. Incident handlers, the people who verify an incident has occurred and act to limit damage, can be on staff, on contract, such as a managed security services provider, or available when needed, such as through a parent organization or an external services provider. Each of those is a specific, defined arrangement. Someone has agreed to be reachable, within some agreed window, with some agreed scope of authority to act.

A monitoring dashboard does none of that. It can be perfectly accurate and still sit unattended at 3 a.m., not because the organization lacks visibility into the problem, but because nobody specific has committed to seeing the alert and acting on it within any particular amount of time. The dashboard satisfies Detect. It says nothing about who owns Respond, when they are reachable, or what they are actually authorized to do once they see the alert. Confusing the two is confusing “we would eventually notice” with “someone is specifically on the hook to fix this now.”

What this gap looks like in a payments system specifically

Visa’s own rules describe exactly how seriously this distinction is treated in a live payments network, because the consequences of getting it wrong are not abstract. Visa Core Rules Section 7.3.3.1 requires a Member to provide authorization services 24 hours a day, 7 days a week. In the US region, it requires an issuer or its authorizing processor to respond to authorization requests in an average time not exceeding 5 seconds during each calendar month. That is not a requirement to have a system someone can check on. It is a requirement that someone, or something, actually answers within a defined time, continuously, without exception for nights, weekends, or a quiet holiday when nobody happens to be watching the dashboard.

Visa’s network has a specific mechanism built for exactly the moment that commitment fails. Stand-In Processing is defined in Visa’s own rules as the component that provides authorization services on behalf of an issuer when the issuer, its processor, or a Visa scheme processor is unavailable, or when issuer responses exceed the maximum response time. Read that definition against the monitoring-versus-on-call distinction directly. It activates precisely when a system that was presumably being watched failed to produce an actual, timely response, for whatever reason. Visa did not build Stand-In Processing because issuers lack monitoring. It built it because monitoring alone does not guarantee the response the network actually needs within the window the network actually requires.

A concrete example: the alert that fired and waited

Picture a payments platform with a monitoring dashboard that is genuinely good. It tracks authorization success rates in real time, and at 2:14 a.m. it correctly flags a drop in approval rates on one processing route. The alert fires exactly as designed. Nothing about the detection failed.

What happens next depends entirely on whether anyone is actually on call, not on anything the monitoring system itself does differently. If the alert pages a specific engineer with a committed response window, that engineer is woken up, confirms the issue, and starts working it within minutes. If the alert instead lands in a channel that gets checked whenever someone next looks at it, the drop in approval rates continues for hours, not because detection failed but because detection was never connected to a real, time-bound commitment to respond. The business impact in those two scenarios is completely different, even though the monitoring system behaved identically in both. The gap is not in what was observed. It is in whether anyone was actually obligated to act on it.

What actually being on call requires that a dashboard does not

Being on call for a system means a specific person or team has agreed to be reachable within a defined window, has the authority to act once reached, and has a clear path to escalate when the first response is not enough. That requires a paging system that actually interrupts someone rather than a dashboard that waits to be checked, a named owner for every category of alert rather than a shared inbox nobody is specifically accountable for, and a documented response procedure so the first action taken does not depend on which engineer happens to be awake that night.

None of this replaces good monitoring, and none of it works without good monitoring behind it. It is what monitoring has to be connected to in order to actually produce the response time a payments platform’s own commitments, and its processors’ and card network’s rules, actually require. A platform that has invested heavily in dashboards and alerting while leaving the on-call side genuinely informal has built exactly the half of the system NIST’s guidance describes as Detect, with no reliable Respond behind it.

Why the roles behind on-call matter as much as the schedule

NIST’s guidance is specific about how wide the circle of people behind a real response actually is, and this is where many platforms stop short even after they build a paging rotation. Incident handlers are the core of the response, but NIST also names technology professionals, legal, and asset owners as roles that routinely need to be involved once an incident is actually confirmed. A payments platform’s version of this list has its own specifics: the engineer paged first, the person who can authorize a failover to a backup authorization path, and whoever owns the relationship with the processor or sponsor bank who may need to be told what is happening while it is still happening, not after it is resolved.

A rotation that only covers the first of those roles is still an improvement over no rotation at all, but it recreates a smaller version of the same gap this article describes. The engineer gets paged and responds quickly, then discovers the actual fix requires an authorization only someone else holds, and that person was never part of the on-call plan at all. Building the full roster, not just the first responder, is what keeps a fast page from stalling out one step later.

What this means for a payments platform’s operational maturity

The practical test is simple to state and uncomfortable to run honestly. Pick the alert most likely to fire at the worst possible time, and ask who is specifically committed to responding to it, how fast, and with what authority to act. If the honest answer is “whoever happens to see it,” the platform has monitoring without on-call coverage, and the gap will surface exactly when Visa’s own Stand-In Processing definition describes: when a system becomes unavailable or exceeds the response time nobody was actually watching the clock on. Running that test honestly, before an actual outage forces the answer, is far cheaper than discovering the gap live.

Building the paging, escalation, and ownership structure that turns detection into a genuine, time-bound response is exactly the discipline behind Midcore’s CloudOps work. Where the underlying deployment and incident-response tooling itself needs to be built to support that structure, rather than bolted on after the fact, DevOps practices are where that foundation gets laid. And for a platform that wants the on-call commitment itself handled by a team already built around payments-specific response expectations, rather than assembled internally for the first time under pressure, managed cloud support services is exactly that function.

Frequently Asked Questions

What is the actual difference between monitoring a system and being on call for it?

NIST’s current incident response guidance, SP 800-61r3, separates these into distinct functions in its CSF 2.0-based model: Detect, where possible incidents are found and analyzed, and Respond, where actions regarding a detected incident are actually taken. Monitoring satisfies Detect. Being on call means a specific person or team has committed to performing Respond within a defined time, which a monitoring dashboard alone does not guarantee.

What does Visa’s rulebook actually require for authorization system availability?

Visa Core Rules Section 7.3.3.1 requires a Member to provide authorization services 24 hours a day, 7 days a week, and requires an issuer or its authorizing processor in the US region to respond to authorization requests in an average time not exceeding 5 seconds during each calendar month. This is a response-time commitment, not just a requirement to have a system that can be observed.

What is Visa’s Stand-In Processing, and why does it exist?

Stand-In Processing is the component of Visa’s network that provides authorization services on behalf of an issuer when the issuer, its processor, or a Visa scheme processor is unavailable, or when issuer responses exceed the maximum response time. It exists precisely for the scenario where an issuer’s system was being watched but nobody was actually positioned to respond to a problem in time, which is the same gap that separates monitoring from genuine on-call coverage.

Sources: NIST Special Publication 800-61 Revision 3, Incident Response Recommendations and Considerations for Cybersecurity Risk Management, April 2025, and Visa Core Rules and Visa Product and Service Rules, Section 7.3.3.1 and the Stand-In Processing (STIP) definition.

Cloud & Infrastructure practiceMidcore Operations · 21 October 2026
All insights
Free consultation

Get your free Assess

Tell us what is happening. We reply within one business day with a US-based practice lead, not a salesperson.

  1. 01Within 1 business dayA US-based practice lead replies and books a 30 minute call.
  2. 02On the callWe map the problem, the volume, the partners, and the constraints.
  3. 03After the callYou get a written read of one to three pages, yours to keep.
  • Handled under NDA
  • Written, not a slide deck
  • No obligation to continue

No obligation. We will tell you if you do not need us.

Before you go

Take the written read with you

Every engagement opens with an Assess: a written read of what is happening, what it costs, and what to do about it. It is quoted on its own and you can stop after it.

+1 (786) 619-0152