glamoxyn

1. Reviewing only the part that went wrong

When a programme has a bad night, the instinct is to inspect the single link that failed: the replay server, the talkback, the graphics feed. A review that stops there tends to fix a symptom and leave the cause in place. Most faults I am asked about began upstream, in a decision made during rehearsal or in a handover that nobody wrote down.

A proper review follows the whole chain from the first rehearsal to the final feed leaving the building. That means sitting in on a run-through, reading the running order, and asking who touches each signal and in what order. It is slower than chasing one fault, but it finds the three small habits that together produced the failure.

Start from the fault by all means, then widen the circle until you reach points where nothing is going wrong. Those quiet points often show what the crew has quietly learned to work around.

  • Trace the signal from source to transmission, not just the failed link
  • Sit in on at least one full rehearsal before judging anything
  • Ask who touches each stage and what they hand on

2. Treating rehearsal as a formality

Rehearsal is where a workflow shows its real shape, yet it is often squeezed or run with stand-ins who behave differently from the live team. A review built on a tidy rehearsal will describe a tidy workflow that does not exist on the night.

Watch for the shortcuts people take when time is short: skipped line checks, cue calls given from memory, a switch of presenter without a mic test. These are not failings of character. They are signs that the schedule leaves no room for the checks the chain depends on.

I ask for the rehearsal plan as well as the transmission plan, and I note where the two diverge. Gaps between them are where the surprises live. If rehearsal never exercises a fallback route, then nobody on the crew truly knows how long the fallback takes.

3. Accepting verbal findings

A review delivered over a cup of tea in the gallery feels productive, and then it evaporates. Within a fortnight the crew remembers a different version, and the manager remembers a third. Spoken advice cannot be filed, compared or handed to a new starter.

Every review we carry out ends in written recommendations, ordered by effect on the programme and by effort to put right. Each item says what was observed, why it matters and what we suggest. If a suggestion cannot be written in a couple of plain sentences, we have not understood the problem well enough yet.

Insist on that standard from whoever reviews your chain, including your own staff. A document the whole crew can read settles arguments that memory never will, and it gives next season's team a starting point rather than a blank page.

  • Rank findings by effect and by effort
  • Keep each recommendation to a few plain sentences

4. Recommendations with no owner

The most polite way for a report to fail is to be agreed with and then ignored. A list of sensible changes, none of which has a name beside it, will sit in a shared folder until the next fault arrives.

Give every recommendation a person and a rough point in the production calendar. Not a deadline imposed from outside, simply an honest statement of who will look at it and when. Where a change needs money or equipment, say so, and record who decides.

I would rather see five modest changes made than twenty ideal ones admired. Small corrections, such as a labelled patch bay or a single agreed cue phrase, are cheap and tend to hold. They also build the crew's confidence that a review leads somewhere.

5. Ignoring the people who run the chain

Diagrams show equipment; they rarely show fatigue, habit or the unwritten rule that only one person knows how to restart the router. A review that interviews nobody misses the knowledge that actually keeps transmission steady.

Talk to the vision mixer, the sound assistant, the floor manager and whoever loads the graphics. Ask each what they would change if they could change one thing, and what they dread most on a live day. Their answers rarely match the manager's, and the differences are useful.

Be careful how those conversations are used. People speak openly only when they trust that a remark will not be repeated back to them as a criticism. We report patterns, not individuals, and we say so before the first question is asked.

  • Interview every role in the chain, not only the leads
  • Report patterns, never single names

6. Reviewing once and filing it away

A production chain changes with every new presenter, venue, camera and software update. A review carried out two autumns ago describes a workflow that has partly gone. Treating it as permanent is a quiet mistake with a long tail.

Plan a lighter check after any significant change, and a fuller look on a sensible cycle. Even an hour spent walking the chain with the original recommendations in hand will show which have held and which have slipped back into old habits.

The aim is not a perfect system but one the crew understands well enough to repair themselves at short notice. A review that leaves behind that understanding has done its work, whatever the paperwork says.

Most mistakes in reviewing a production chain come down to looking too narrowly, recording too loosely or following up too little. Follow the whole chain, take rehearsal seriously, write findings down, give each one an owner, listen to the crew and return to it when things change. None of this needs special equipment, only patience and a willingness to be told where the real friction sits.