Fallback, Failover and Redundancy: Three Trade Words Explained Plainly

Walk into almost any gallery during a live programme and you will hear someone say they have a fallback for the vision mixer, or that the router is redundant, or that the feed will fail over if the line drops. The three words are often swapped as though they were one idea. In a written workflow review they are kept apart, because each one asks a different question of the crew and of the equipment.

Redundancy is about having a spare. Failover is about what switches to the spare, and how. Fallback is about what the people do when the technical answer runs out. A studio can be strong on one and weak on the other two without anybody noticing until a bad evening.

When I review a gallery, I start by asking the team to define each term in their own words. The answers rarely match from one person to the next, and that gap is usually where the first useful finding sits.

Redundancy simply means a duplicate exists. A second power supply in a rack, a spare camera control unit on the shelf, two independent paths for the programme feed to the transmission point. Nothing in that definition says the duplicate will be used correctly, or at all.

The common trap is assuming that a second item gives protection. Two devices fed from the same power strip, or two cables laid in the same duct, share one point of failure and are only redundant on paper. A review traces each duplicate back to what it depends on and marks where the shared dependencies sit.

It also asks whether the spare has been powered up and tested recently. A duplicate that has sat unused for a year and turns out to carry old settings is a spare in name only, and finding that out mid-programme is expensive.

• Check what each duplicate shares with the original
• Test spares on a set schedule, not only when something breaks
• Record the settings a spare must carry

Failover is the moment of changeover. It can be automatic, where equipment detects a fault and moves to the spare by itself, or manual, where someone presses a button or patches a cable. Both are legitimate. What matters is that the crew knows which one they have for each signal.

Automatic failover is quick but can surprise people. A feed that quietly moves to its backup leaves the crew unaware that they are now running without protection. A good setup raises a clear alert when failover happens, so the next fault does not arrive unannounced.

Manual failover relies on someone knowing where the control is, what it does and who is allowed to use it. In a review I time it with the crew: how many seconds from noticing a fault to the spare being live. That figure often changes the conversation about what needs automating.

Fallback is the human layer. It covers the decisions made when the spare also fails, or when the fault is not a piece of kit at all, such as a presenter taken ill or a guest link that will not connect. It is written down as a short sequence the director can follow without having to invent it under pressure.

A usable fallback is specific. It says which slate or holding graphic goes to air, who tells the presenter, who calls transmission, and at what point the programme moves to a pre-recorded item. Vague lines such as go to something else leave the gallery guessing in the one moment guessing is costly.

These documents are best kept to a page or two and rehearsed. A fallback that nobody has read aloud together tends to fall apart on first contact with a real fault, so I recommend a short walk-through with the whole gallery before it is relied on.

• Name the person who calls each decision
• Keep the document short enough to read in a minute
• Rehearse it with the full crew at least once

In a review, each signal path gets looked at through all three words in turn. For every important feed, such as the programme output, the talkback and the studio clocks, I note what the spare is, how the switch happens, and what the crew does if both fail. Gaps show up quickly when the same question is asked three ways.

The output is a written report, not a verbal debrief. It lists findings in the order they are likely to matter on air, with a practical fix beside each one and a note of how much effort the fix involves. Some are a label and a rehearsal; a few need equipment or rewiring.

Nothing in the report assumes a bigger budget than the studio has. A well-rehearsed fallback and a labelled manual changeover often protect a programme better than an expensive device nobody has practised with.

You do not need a consultant to start. Pick the one signal the programme cannot lose and ask the crew three things: what is the spare, how long does the switch take, and what do we do if that fails as well. If all three answers come quickly and agree, the gallery is in better shape than most.

If the answers differ, or someone says they would have to look it up, that is useful information. It means the knowledge sits in one head or one folder rather than in the crew as a whole, and a weeknight rehearsal is cheaper than learning it during a broadcast.

These weeks of autumn, with schedules often tightening towards the winter programming, are a sensible time to do this exercise before the busiest part of the year begins.