Who Holds the Pen? The Approval Audit That Decides Whether Your UK Brand Makes the Story

There is a familiar post-mortem in UK communications. A journalist asks for a line by 4pm. The line arrives at 6.40pm. The story runs without it. The spokesperson is described as slow, or unprepared, or insufficiently briefed. In most cases, none of those things is true. The spokesperson was ready. The sentence was written. What failed was the route between the two.

Approval chains are the quiet structural risk in media relations. They rarely appear on a risk register, they are seldom owned by a single director, and they tend to expand by accretion rather than design. Each additional signatory feels prudent in isolation. Collectively, they convert a responsive communications function into a queue.

The argument here is not that approvals should be abandoned. Legal, regulatory and reputational review exist for good reason. The argument is that most UK organisations have never audited their approval chain as a media asset — and that the weeks before December are the right time to start.

The myth of the untrained spokesperson

When a response misses its window, the instinct is to send the spokesperson on another media-training day. This treats a systems problem as a skills problem. A well-drilled spokesperson cannot outrun a review process that requires four signatories and a legal read before a two-sentence comment can leave the building.

The tell is in the language used internally. Phrases such as 'just needs a quick legal check' or 'brand will want to see it' are not descriptions of process. They are descriptions of latency. If those phrases appear in every request, the chain is not a safeguard. It is the default.

Counting the layers

The first step is deliberately mechanical. Take a representative sample of ten recent media requests — a mix of routine corporate comment, product or service questions, and at least one sensitive issue. For each, list every person who touched the draft before it was sent. Not the people copied in. The people whose approval was required.

Most teams discover two things. First, the number of signatories is higher than anyone in leadership assumed. Second, the chain differs by request type, which means no one can describe it accurately from memory.

Then time it. Record the moment the journalist's email arrived and the moment the approved line was sent. Do this for ten requests and calculate the median and the worst case. A 'quick comment' that routinely takes four hours is not a quick comment. It is a workflow, and it should be described as one.

Finding the bottleneck

The bottleneck is rarely the person who appears slowest in the inbox. It is usually the point where a narrow review becomes a broad one. Legal review is a common example: a request that needs a check on one factual claim is routed into a full legal read because there is no triage step. Brand review is another: a tonal preference becomes a blocking approval because no one has defined which tonal questions are material and which are not.

Ask a more useful question than 'who is delaying this?' Ask 'where does the request wait, and what is it waiting for?' That distinction separates capacity problems from authority problems. Adding more reviewers to a queue does not shorten it. Clarifying who can decide does.

Delegated authority thresholds

The remedy is a tiered model with named decision-makers rather than committees. The precise thresholds will vary by sector, but the structure is portable.

Tier one: pre-approved lines. A small set of holding statements and factual positions that any trained spokesperson can use without further sign-off. These cover the predictable questions — service disruption, pricing enquiries, routine appointments — and they are reviewed quarterly, not per request.

Tier two: routine reactive comment. The communications lead plus one subject expert. A target turnaround of thirty minutes during working hours. No legal read unless a defined trigger is met, such as an unproven allegation or a regulated claim.

Tier three: sensitive or reputational matters. Communications lead, legal and the relevant executive, but with a named duty approver empowered to sign off when others are unavailable. A target of ninety minutes, with an agreed fallback line if the target is missed.

Tier four: crisis. A pre-convened group with a standing rota, agreed escalation triggers and a single chair who can make the final call.

The critical design choice is one approver per tier, supported by named deputies. Approval by consensus is approval by nobody. A committee that cannot decide by 5pm is not a control. It is an outage waiting for a news cycle to expose it.

The November stress test

The festive period is the annual proof of whether the chain works. Between mid-December and early January, newsrooms continue to publish, and many of them are working with reduced rotas and longer lead times that make a fast, clean line more valuable, not less. Meanwhile, a significant proportion of approvers are on leave, and the people covering for them often lack the authority they need.

Run a stress test in late November. Submit three fictional requests through the real process: one at 4pm on a Friday, one on a date when the primary approver is known to be away, and one that triggers the sensitive tier. Measure the actual elapsed time and note where the request stalled. Then publish a festive rota that names a duty approver for each tier, with a mobile number and a clear instruction on what they can sign off alone.

Pre-approve the seasonal lines that will almost certainly be needed: trading updates, weather disruption, staffing changes, delivery delays. These are not unpredictable events. They are annual ones, and treating them as surprises is a choice.

Approvals as an asset

The purpose of this work is not speed for its own sake. It is to ensure that the judgement an organisation has already invested in — its positions, its legal boundaries, its tone — reaches the newsroom while the story is still being written. An approval chain that cannot deliver a sentence in time is not protecting the brand. It is removing it from the conversation.

Audit the layers. Time the comment. Name the decision-maker. Then test it in the week before December, when the newsroom is still asking and half the building is already out of office.