ADAM PULSE Knowledge Base
WAN · Circuit Monitoring · Checklist

What should you monitor on every internet circuit? The essential WAN monitoring checklist

By Michael Obraitis, Managing Partner, and Michael Dale, CEO·Published August 12, 2026·Last reviewed August 2026·ADAM Pulse Knowledge Base

Most businesses monitor whether their internet connection is:

Up

or

Down

That is useful.

But it is not enough.

An internet circuit can remain technically online while experiencing:

A better question is:

How Healthy Is the Internet Connection?

To answer that question, businesses should monitor not only availability but also the quality, stability, and historical performance of every important internet circuit.

This guide explains what IT teams should monitor on business internet connections and why those measurements matter.

What Is Internet Circuit Monitoring?

Internet circuit monitoring continuously measures the availability and performance of a business internet connection.

Depending on the environment, that may include:

The objective is to understand whether each connection is functioning normally and whether performance is deteriorating.

What Are the Most Important Internet Circuit Metrics?

At minimum, consider monitoring:

  1. Availability
  2. Latency
  3. Packet loss
  4. Jitter
  5. Gateway availability
  6. Firewall availability
  7. Circuit failover
  8. Historical outages
  9. SLA performance
  10. Performance trends

Depending on the environment, additional metrics may also be important.

1. Internet Circuit Availability

The first question is simple:

Is the circuit reachable?

Availability monitoring identifies complete connectivity failures.

Track:

Do not measure only long outages.

A circuit that fails for 20 seconds several times each day can create significant application problems even if monthly uptime appears high.

2. Latency

Latency measures the time required for network traffic to travel between points.

Monitor latency continuously rather than relying only on occasional ping tests.

Why?

Because latency changes throughout the day.

Imagine a circuit normally operates around:

20 ms

but begins reaching:

120 ms

every afternoon.

The circuit remains online.

Users may still experience poor performance.

Cisco's current IP SLA documentation identifies delay as an important network performance metric because increased delay can affect application experience.

3. Packet Loss

Packet loss occurs when network packets fail to reach their destination.

Monitor packet loss because even modest amounts can affect:

A circuit can show:

UP

while simultaneously losing packets.

That connection is available.

It is not necessarily healthy.

4. Jitter

Jitter measures variation in packet timing.

It is particularly important for:

Cisco specifically includes directional jitter among its IP SLA measurements because inconsistent packet timing can create problems for voice and video applications.

5. Local Gateway Availability

Monitoring only an internet destination leaves a major diagnostic gap.

If the external destination stops responding, what failed?

Possibilities include:

Monitoring the local gateway adds another diagnostic point.

If the local gateway remains available while external connectivity disappears, the troubleshooting investigation shifts outward.

6. Firewall Availability

The firewall should also be monitored independently.

This helps distinguish:

Internet circuit failure

from:

Firewall failure

A site being unreachable does not automatically mean the carrier is down.

The firewall may have:

Separate measurements create better evidence.

7. Carrier Gateway Availability

Where appropriate, monitor the carrier gateway or logical upstream handoff.

A useful diagnostic chain becomes:

Local Gateway → Firewall → Carrier Gateway → External Destination

That structure helps answer:

Where does connectivity stop?

8. Multiple External Destinations

Do not monitor only one public destination.

That creates a risk of confusing a remote destination outage with an ISP outage.

Monitor multiple independent destinations where appropriate.

For example:

Destination A healthy

Destination B healthy

Destination C unavailable

suggests something very different from:

A unavailable

B unavailable

C unavailable

while local infrastructure remains healthy.

9. Historical Outage Frequency

Do not evaluate incidents independently.

Track:

How often does this happen?

A circuit may experience:

or

Those represent very different operational experiences.

Historical reporting helps identify chronic instability.

10. Outage Duration

Track the duration of each event.

For example:

Circuit A:

3 outages totaling 18 minutes

Circuit B:

22 outages totaling 20 minutes

Total downtime looks similar.

Operational behavior does not.

Circuit B may create far more disruption because repeated flapping can interrupt sessions continuously.

11. Circuit Flapping

Monitor repeated state changes.

For example:

UP

DOWN

UP

DOWN

UP

DOWN

This is circuit flapping.

It may contribute to:

Short failures should not disappear into monthly averages.

12. WAN Failover

If a site has redundant circuits, monitor whether failover actually works.

Track:

Buying a backup connection does not guarantee redundancy.

It needs to be monitored.

13. Backup Circuit Health

A backup connection can sit unused for months.

Then the primary circuit fails.

That is the worst time to discover:

The backup circuit is also down.

Monitor backup circuits independently.

Test:

Redundancy is only valuable when the backup is actually operational.

14. SD WAN Path Health

For SD WAN environments, monitor individual underlay circuits.

Two WAN connections may both show online while one performs poorly.

Track:

The question is not only:

Which circuit is up?

It is:

Which circuit is healthy enough for the application?

15. DNS Performance

Internet connectivity may be healthy while DNS creates user problems.

Consider monitoring:

Users frequently describe DNS failures as:

“The internet is down.”

Separating DNS from connectivity improves diagnosis.

16. Application Reachability

Consider monitoring critical business services independently.

Examples:

This helps distinguish:

Internet connection is healthy

from:

Critical application is unavailable

17. Route Behavior

Network paths can change.

That may affect:

Historical route information can provide important evidence during intermittent problems.

Fortinet's current path monitoring approach, for example, stores route information so administrators can examine the path that existed during periods of degraded performance.

18. SLA Performance

If the organization pays for a business internet service with an SLA, measure performance against the applicable commitment.

Potential metrics may include:

The actual SLA terms depend on the carrier contract.

Monitoring does not replace the provider's contractual measurement process, but historical evidence can strengthen service reviews and escalations.

19. Performance Baselines

Do not evaluate every site using exactly the same assumptions.

One location may normally have:

15 ms latency

Another may normally have:

65 ms latency

The important measurement is often:

How does current performance compare with normal performance for this circuit?

Create a baseline for each site and circuit.

20. Performance Trends

A circuit does not need to fail completely before it deserves attention.

Imagine:

Monday: 20 ms average latency

Tuesday: 23 ms

Wednesday: 29 ms

Thursday: 41 ms

Friday: 62 ms

The circuit remains online.

But the trend deserves investigation.

This is the difference between monitoring for failure and monitoring for deterioration.

What Does a Healthy Internet Circuit Look Like?

There is no single universal number.

Health depends on:

A useful definition is:

A healthy circuit behaves consistently within the normal performance range required by its business applications.

That is why baselines are more useful than relying solely on generic thresholds.

How Often Should Internet Circuits Be Monitored?

Critical business connections should be monitored continuously.

If the circuit fails for two minutes between manual tests, manual testing will never reveal the event.

Persistent monitoring is particularly important for:

Should You Monitor Broadband Connections?

Yes.

Even lower cost business broadband connections can support critical operations.

If employees depend on the connection, it deserves monitoring appropriate to its business impact.

Should You Monitor Dedicated Internet Access?

Absolutely.

Dedicated Internet Access often comes with stronger performance expectations and contractual service commitments.

Historical circuit data can be especially valuable for validating performance and supporting carrier escalation.

Should You Monitor Cellular Backup Connections?

Yes.

Cellular backup circuits can fail because of:

If the cellular backup is part of the continuity plan, its health needs to be known before the primary circuit fails.

Should You Monitor ISP Performance Across Multiple Locations?

Yes.

This can uncover powerful patterns.

For example:

Carrier A operates 35 circuits.

Seven sites repeatedly experience latency increases.

All seven are in the same region.

That may indicate a provider or regional pattern rather than seven unrelated branch problems.

Centralized visibility makes this type of correlation possible.

What Should an Internet Circuit Monitoring Dashboard Show?

At minimum:

Location

Carrier

Circuit

Current status

Firewall status

Gateway status

Latency

Packet loss

Recent incidents

Historical availability

SLA performance

For dual WAN environments:

Primary status

Backup status

Current active path

The dashboard should help the operator determine what needs attention quickly.

Why Is Up or Down Monitoring Not Enough?

Consider:

Availability: UP

Latency: 190 ms

Packet loss: 7%

Jitter: High

Is the circuit technically online?

Yes.

Would users consider it healthy?

Probably not.

That is why:

Availability Does Not Equal Quality

Monitor both.

The ADAM Pulse Approach to Internet Circuit Monitoring

ADAM Pulse is designed to provide more context than:

Circuit Up

or

Circuit Down

Our goal is to help organizations understand:

Which circuit is affected?

Is the gateway responding?

Is the firewall responding?

Is the carrier reachable?

What is the current latency?

Is packet loss occurring?

Is the condition new or recurring?

Is performance deteriorating?

Is the backup connection healthy?

Does the evidence point toward the carrier or local network?

Build an Internet Circuit Health Record

Every important business circuit should have a history.

That record can show:

That history becomes enormously valuable during:

Stop Monitoring Your Internet Like a Light Switch

Your internet connection is not simply:

ON

or

OFF

It has a quality level.

It changes.

It degrades.

It recovers.

It develops patterns.

ADAM Pulse provides managed network monitoring designed to help organizations continuously understand the health of internet circuits across their locations.

Know whether the circuit is online.

Know whether it is healthy.

Know how it has performed over time.

Learn more about ADAM Pulse and talk with USA Telecom about monitoring your business internet connections.

Frequently asked questions

How Healthy Is the Internet Connection?

To answer that question, businesses should monitor not only availability but also the quality, stability, and historical performance of every important internet circuit. This guide explains what IT teams should monitor on business internet connections and why those measurements matter.

What Is Internet Circuit Monitoring?

Internet circuit monitoring continuously measures the availability and performance of a business internet connection. Depending on the environment, that may include: The objective is to understand whether each connection is functioning normally and whether performance is deteriorating.

How Often Should Internet Circuits Be Monitored?

Critical business connections should be monitored continuously. If the circuit fails for two minutes between manual tests, manual testing will never reveal the event. Persistent monitoring is particularly important for:

Should You Monitor Broadband Connections?

Yes. Even lower cost business broadband connections can support critical operations. If employees depend on the connection, it deserves monitoring appropriate to its business impact.

Should You Monitor Dedicated Internet Access?

Absolutely. Dedicated Internet Access often comes with stronger performance expectations and contractual service commitments. Historical circuit data can be especially valuable for validating performance and supporting carrier escalation.

Should You Monitor Cellular Backup Connections?

Yes. Cellular backup circuits can fail because of: If the cellular backup is part of the continuity plan, its health needs to be known before the primary circuit fails.

What Should an Internet Circuit Monitoring Dashboard Show?

At minimum: For dual WAN environments: The dashboard should help the operator determine what needs attention quickly.

Sources

Editorial note

Monitoring requirements and the controls appropriate to them vary by organization. A single test from a single location at a single moment rarely proves where a fault sits — correlate against history, test from more than one point, and preserve evidence before changing configuration.

Talk to us about your network security

USA Telecom Consulting LLC is a Service-Disabled Veteran-Owned Small Business running a 24/7 NOC. We monitor networks, circuits and firewalls for regulated and defense-supply-chain organizations.

Related articles

How to troubleshoot VoIP call quality: packet loss, jitter,… · How to monitor your ISP SLA and hold internet carriers acco… · How to monitor multiple business locations from one network… · Is it the ISP or the firewall? · Monitoring software vs managed NOC · o · u · u · u

← More from the ADAM Pulse Knowledge Base

Prepared by ADAM Pulse (USA Telecom Consulting LLC) — managed AI governance, network, and NOC services. Support: (888) 989-4872 · support@adampulse.us