A business continuity plan for connectivity writes down what depends on your internet and phones, how long each function can be without them, what fails in each likely scenario, who does what in the first hour, and how customers still reach you. A backup circuit is one piece of it. The plan is what makes that circuit, and everything else, actually work on a bad day.
We have written separately about whether you need backup internet. This guide is the wider exercise: the plan that sits around the equipment. It is mostly paper, phone numbers and a few decisions made calmly in advance, rather than in a hurry with the office offline.
Step 1: List what depends on the connection
Start with functions, not equipment. Walk through a normal day and note everything that stops when the internet or phones stop. A simple table is enough:
| Function | Depends on | Longest tolerable outage | Fallback today |
|---|---|---|---|
| Incoming customer calls | Hosted phone service, office internet | Minutes | None written down |
| Card payments | Terminals over the office network | Under an hour | Terminals with cellular backup |
| Email and files | Cloud services, any internet | Half a day | Staff phones and home connections |
| Line-of-business application | Server in the office, VPN for remote staff | Two hours | None |
| Door access and cameras | Office network and cloud platform | A day, if doors fail safe | Physical keys |
The rows are only examples. Yours will differ, and the gaps in the last column are the point of the exercise.
Step 2: Set a recovery target for each function
Continuity planners use two terms worth borrowing. The recovery time objective is how quickly a function must be working again. The recovery point objective is how much recent data you can afford to lose, which matters for systems that store records. A business that cannot miss calls for more than five minutes needs automatic call routing. A business that can live without its file server for a morning needs a documented restore, not a second data center. Setting the targets first stops you overspending on functions that can wait and underspending on the ones that cannot.
Step 3: Walk the failure scenarios
Different failures need different responses. A backup circuit helps with some of these and does nothing for others.
| Scenario | What fails | First response | What helps |
|---|---|---|---|
| Fiber cut outside the building | Primary circuit | Confirm failover, open carrier ticket | Backup on a physically diverse route |
| Carrier-wide outage | Every circuit from that carrier | Same as above; check the backup is a different network | Backup from a different carrier |
| Power cut in the building | Firewall, switches, phones, possibly carrier equipment | Check battery runtime; move staff if long | Battery backup on network gear, cloud phones |
| Firewall or switch failure | Everything behind it, both circuits | Swap or restore configuration | Saved configuration, spare unit or fast replacement |
| Office inaccessible | The building, not the network | Staff work from home; phones follow them | Cloud phones, cloud apps, remote access |
| Cloud service outage | A provider, not your connection | Check the provider's status page; communicate | Offline procedures for critical tasks |
| Cyber incident | Systems may need to be taken offline deliberately | Follow the incident plan; isolate, do not improvise | Segmentation, protected backups, response contacts |
The firewall row catches many businesses out. Two circuits plugged into one firewall still share a single point of failure. A saved configuration file and a known replacement path turn a day-long outage into an hour.
Step 4: Keep the phones answering
For most businesses, calls are the function with the shortest tolerance. A few decisions made in advance cover most situations:
- Cloud phone service. Hosted phones keep the numbers in the provider's platform, so an office outage does not take the numbers down. See hosted phones.
- Pre-built failover routing. If the desk phones are unreachable, calls should automatically ring mobiles, a second site or an answering service. Set it up and test it; do not plan to configure it during an outage.
- An answer for every call. An AI receptionist or answering service can take messages and handle simple requests when staff are busy dealing with the outage. FiberX's own line, 478-758-8091, is answered by our AI receptionist, so we see this in practice.
- An outage message. A short recorded or website notice saying you are aware and how to reach you.
Step 5: Roles, contacts and a copy on paper
When the network is down, the shared drive with the plan on it is down too. Keep a printed copy and a copy on a phone, with:
- Who decides, and who deputises when that person is away.
- Each circuit's carrier, account number, circuit ID and support number.
- Your agent or IT provider's direct contact.
- Phone system admin login location and the steps to reroute calls.
- Building management and electrician contacts.
- A short message template for customers and staff.
Having one agent who knows every circuit on your account makes this list shorter. One call starts the tickets.
Step 6: Equipment and power
- Battery backup sized to keep the firewall, core switch, carrier equipment and a few phones up for a known time. Write the runtime down.
- A current configuration backup for the firewall and switches, stored off the devices.
- A spare or a confirmed replacement time for critical hardware.
- A cellular hotspot or router for the few people who must stay online if everything else fails.
- Failover managed automatically where possible; SD-WAN is the usual tool for switching between circuits.
Step 7: Test it
A plan nobody has tested is a hope. Two kinds of test are worth doing:
- A tabletop walk-through, at least once a year: sit the people named in the plan round a table, describe a scenario, and have them say what they would do. Gaps appear quickly.
- A live failover test, at a quiet time: unplug the primary circuit, confirm the backup carries traffic and the phones still ring, then reconnect. Repeat after any significant change to the network.
Keep the tests safe. Agree a time with the people affected, have someone ready to reconnect the primary if the backup does not take over, and do not test failover for the first time on the day of a payroll run or a product launch.
Keep the plan current
Plans go stale quietly. A carrier changes, an account number is replaced at renewal, the person who knew how to reroute the phones leaves, a new application moves into the office server room. Tie a review to events rather than only to the calendar: whenever a circuit is added or replaced, a phone system changes, a key person leaves, or the business moves or opens a site. A plan that is fifteen minutes out of date is useful. One that is two years out of date sends people to call a support number that no longer exists.
Record what happened and what you changed. Monitoring helps between tests, by telling you when a backup has quietly failed; see network monitoring for small business.
A worked example
A hypothetical illustration. The business, timings and outcomes are invented.
A 30-person insurance brokerage has fiber, a cable backup and hosted phones. On paper it is resilient. The tabletop exercise finds three gaps. The phones fail over to mobiles only if someone logs in and changes a setting, and the one person who knows how is often out on client visits. The firewall has no saved configuration, so a failure would mean rebuilding from memory. And the battery backup covers the firewall but not the carrier's equipment, so in a power cut both circuits go down within minutes anyway.
The fixes take an afternoon and modest spending: automatic call routing to a ring group of mobiles, a configuration backup scheduled weekly, and the carrier equipment moved onto the battery. The live test a month later works. Say the brokerage estimated that each hour of missed calls costs it 10 lost enquiries; the plan's value is avoiding that during the next outage, not the equipment on the shelf.
A one-page plan template
- Critical functions, each with a recovery target.
- Circuits, carriers, IDs and support numbers.
- Failure scenarios with the first three steps for each.
- Phone rerouting steps and who can perform them.
- Decision-maker and deputy.
- Battery runtime and spare equipment.
- Customer and staff message templates.
- Date of last tabletop and last live test.
Questions to ask your provider
- If my primary circuit fails, how will I know, and how do I open a ticket outside business hours?
- Is my backup on a different network and a different physical route from the primary?
- Does your equipment at my site need power from my battery backup, and how much?
- Can my phone numbers ring mobiles automatically if the office is unreachable?
- What is the replacement time if my managed firewall or router fails?
- Will you take part in a failover test?
Building the plan with the circuits under it
A continuity plan is only as strong as the circuits and phone services it relies on. We price primary and backup circuits from different carriers, check the routes, and set up phones that follow your staff, with managed network support if you want someone else watching. Several Tier 1 carriers bid on your address, the carrier pays us so there is no markup, and if your current setup already holds up, we will say so. The quote is free with a same-day response. Send us your address, call 478-758-8091 or text (347) 870-0965.
