🛡️ Business continuity: communications that stay up

If your communications go down, your business stops: no calls come in, no orders arrive, customers are not served. Continuity means designing communications so they do not go down —and if something fails, they recover on their own.

The pillars of continuity

• Redundancy: more than one path for calls and data.
• Failover: if a link or service fails, traffic is rerouted automatically.
• Monitoring: 24/7 watch that detects and alerts before it hits (see NOC Monitoring).
• Cloud: managed, distributed platforms, not a single box in your office.

Why the cloud helps

A Cloud Phone System (UCaaS) does not depend on a single site: if your office loses power or internet, your calls can keep routing to mobiles or another location. With a SIP Trunk and managed routes, the operation keeps running.

How to start

Assess your single points of failure (one internet link, one PBX) and add redundancy where it hurts most. At TelHarbor we design communications with end-to-end continuity.

Find your single points of failure

Before spending on redundancy it helps to know where it hurts. The exercise is short: list everything that, if it fails, leaves your company without communications. In most offices the list is shorter than people expect, and it almost always includes:

• A single internet provider. The most common point of failure and the easiest to fix.
• A single PBX, in your office. Lose power or flood the comms room and it is over.
• A single published number. If that number fails, the customer has no plan B.
• A single person who knows how everything is configured. This one is real and never makes it onto the diagram.

How long can you go without phones

Two questions frame the whole conversation and stop you overspending:

• How long can you be down before it really hurts. Half an hour is tolerable for many companies; for an emergency Contact Center it is not.
• How much data can you lose. Recordings, history, unanswered messages.

If your answer is "minutes", you need automatic failover. If it is "a few hours", a well-documented and tested manual procedure is enough. Paying for automatic when manual would do is just as common as the opposite.

Levels of continuity, cheapest to most complete

1. Emergency forwarding: the main number forwards to a mobile. Costs almost nothing and avoids the worst case. It has to be configured beforehand, not on the day.
2. Cloud PBX: your phone system stops depending on your office power and internet.
3. Dual internet links with automatic failover on the router.
4. Redundant outbound routes: more than one path for calls, so one carrier's outage does not mute you.
5. Distributed sites or agents: if one site goes down, another keeps answering.

An untested plan is not a plan

The part that is almost always missing: testing it. Once a year, disconnect the primary link on purpose, in a controlled window, and check that failover does what the document says. It is uncomfortable and something always turns up: a forward pointing at a mobile that no longer exists, an emergency contact who left eight months ago, a procedure referring to hardware that has since been replaced.

Write down who does what, with names and numbers, and where that document lives. Nobody improvises well in the middle of an outage.

Voice continuity: what happens to the calls

Business voice continuity is the part of the plan nobody writes down until the day the phones go quiet. Email waits; a call that never gets answered is a customer who calls someone else. Four things decide whether your calls survive an outage:

• Where the phone system lives. A PBX in your office dies with your office: one power cut, one fibre cut, one flood. In the cloud, the outage at your site does not take the system with it.
• Where the numbers point. Your published numbers must be able to reroute — to a second site, to mobiles, to voicemail with an email copy — without waiting for the carrier to do it for you.
• How the agent connects. A Softphone on a laptop or a mobile keeps the same extension from home, a café or another branch, so «the office is down» stops meaning «nobody can call us».
• The second path out. A backup internet link, or a second trunk on a different provider, so calls keep flowing while the main one is being repaired.

Write down two numbers for voice, separately from the rest of the plan: how long you can go without answering calls, and how much of the call history you can afford to lose. Those two set how much redundancy is worth paying for.