Business phone disaster recovery: keep the lines ringing when things go wrong

A phone disaster isn't a hurricane. It's a roof leak in the comms cupboard on a Sunday night, a digger through the broadband at the end of your road, a power cut that lasts the morning, or a ransomware attack that takes your PBX offline for 48 hours. The realistic failure modes for a UK SME phone system are mundane — and they happen regularly.

What good looks like, what most businesses miss, and the routing patterns that keep customers reaching you even when something's on fire. Most of this is included with Membership without you having to ask.

The realistic failure modes

• Broadband outage. Openreach engineer digs through your circuit. Outage can be hours or days. Hits every IP-based service in your office.
• Power cut. Local substation, planned works, a tree on a line. Office is dark; phones are too (unless they're on UPS or PoE+battery).
• Hardware failure. Your on-prem PBX, your router, your switch. Anything mechanical eventually breaks.
• Building inaccessible. Fire, flood, burst pipe, security incident next door. Your staff can't get to the desks.
• Ransomware / cyber incident. The PBX is now offline by IT's order, and might be for days.
• Snowed-in staff. The classic UK problem. Two inches of snow, half the team can't get in.

Every one of these has a routing answer that means customers still get through.

Layer 1: failover routing on the inbound number

The single highest-leverage DR setup for an SME phone system: configure every inbound number with a failover destination. If the primary destination (your SIP trunk, your desk phones, your office PBX) doesn't answer within 10 seconds — or doesn't register at all because your broadband is down — the call automatically routes to plan B.

Plan B can be:

• A mobile or hunt group of mobiles
• A second office on a different broadband connection
• Voicemail-to-email with a clear message about the outage
• A virtual receptionist / IVR menu that triages the call

With NumberGroup Membership, this is configured in the portal — set primary and failover per number, per extension, or per hunt group.

Layer 2: softphone everywhere

When your office is inaccessible, your team needs to take calls from wherever they are. With softphones on every laptop and mobile, a snowed-in salesperson takes their extension's calls from their kitchen table. A flooded office sees zero customer impact because the calls just land on home-WiFi softphones.

With Membership, branded Acrobits softphones are supplied at our distributor cost (~75p per install). Mix-and-match with desk phones — same extension, multiple devices.

Layer 3: time-of-day + holiday routing

Most outages happen out of hours. Time-of-day routing rules let you define:

• Business hours — ring desk phones first, mobiles as backup
• Lunchtime — extended ring time, then voicemail
• Out of hours — straight to voicemail-to-email, or to an on-call mobile, or to an answering service
• Bank holidays / scheduled closures — recorded message + voicemail
• Emergency mode— a manually-triggered routing rule that overrides everything ("everyone's evacuated to the car park, ring mobiles only")

All of this is included with Membership. No bolt-on.

Layer 4: physical redundancy

For businesses where phones are mission-critical (sales floors, support call centres, anyone taking inbound orders):

• Dual broadband. Two circuits from different carriers (e.g. one fibre, one 4G/5G backup). Auto-failover on the router.
• UPS on the comms cupboard. 30 minutes of battery covers most outages and gives you time to transition gracefully.
• PoE+ switches with battery backup. Keeps desk phones up when the rest of the office is dark.

Layer 5: documented incident plan

A 1-page runbook that the duty manager can find at 3am. Keep it short, specific, and actionable. Include:

• How to switch the inbound numbers to emergency routing (and who has the login)
• Carrier support number (us: 0800 310 1010)
• List of staff mobiles to use as failover destinations
• How to record + deploy an "experiencing technical issues" voicemail message
• Who to call internally (IT, facilities, the director on duty)

Update once a year. Test it once a year (low-stakes — a half-hour drill on a quiet afternoon).

Why hosted PBX makes DR cheaper

A traditional on-prem PBX is the single point of failure that makes most of this expensive. Once you move to a hosted cloud PBX, the PBX itself lives in a Tier-3 data centre with redundant power and network. Most of your DR concerns become carrier concerns — and routing failover, softphone, time-of-day rules become configuration choices in a portal instead of hardware projects.