Scoring and ranking audit findings

An audit that ends with a flat list of findings has done half its job. It has told you what is wrong, but not what to do first, and on a building of any size those are very different questions. A list treats a blocked final exit and a faded notice as equivalent items, because a list has no opinion about severity. The manager reading it is left to impose an order the audit should have provided, usually under time pressure, often inconsistently, and rarely in a way that survives scrutiny later. Scoring and ranking findings is how an audit stops being a description and becomes a plan.

The purpose is not to add bureaucracy. It is to make the most important thing the first thing, reliably, every time, across every building and every auditor. A scoring scheme is simply an agreement, made in advance, about what matters more, so that the decision is not reinvented finding by finding.

Why a flat list fails

A list of findings fails for a predictable set of reasons. Without an agreed severity, the order in which things get fixed tends to follow the order in which they were noticed, or how loudly a particular finding was written up, rather than how much risk it carries. Across a portfolio, two auditors will rank the same issue differently, so the data cannot be compared building to building. And when a regulator or insurer later asks why a serious item sat open while a trivial one was closed, a flat list offers no defensible answer, because no priority was ever recorded.

Scoring fixes all three. It gives every finding a defensible rank, it makes findings comparable across auditors and buildings, and it produces an ordered queue that tells the maintenance and remediation effort exactly where to start.

Severity times likelihood

The most workable schemes rate two things and combine them. Severity is how bad the consequence would be if the issue led to harm. Likelihood is how probable it is that it does, given the current state of the building and its controls. A high-severity, high-likelihood finding, a blocked escape route in an occupied building, sits at the top. A low-severity, low-likelihood finding, a cosmetic defect with no safety bearing, sits at the bottom. The interesting cases are the corners: high severity but low likelihood, and low severity but high likelihood, where judgement is genuinely required and where a scoring discipline stops gut feel from quietly deciding.

A simple matrix makes the combination explicit and consistent.

The labels matter less than the agreement behind them. What turns this from a colourful grid into a working tool is attaching a meaning to each band: what an Urgent finding requires by way of response time, who an High finding must be escalated to, and what a Low finding can reasonably wait for.

Tie the band to an action and a clock

A score is only useful if it changes behaviour, and the way it changes behaviour is by attaching a required response to each band. An urgent finding might demand action that day and immediate escalation to the responsible person; a high finding a defined remediation window; a medium finding inclusion in planned work; a low finding a note for the next cycle. The point is that the band is not a description of how worried to feel. It is an instruction about what happens next and by when.

This is also where scoring connects to the statutory picture. Some findings map onto duties with their own timescales, and the scoring should respect them rather than override them. A damp and mould hazard in social housing, for instance, sits inside the Awaab's Law framework, whose first phase came into force on 27 October 2025 with timescales reported as investigating within 10 working days, a written summary within 3 working days of the investigation concluding, beginning to act within 5 working days of finding a significant hazard, and making safe emergency hazards within 24 hours. Where a duty sets a clock, the scheme should never produce a slower one; the score sits on top of the legal minimum, not in place of it.

Make the ranking the spine of the action plan

A scored audit naturally produces an action plan, because the ranking is the plan. The highest-scored findings become the first corrective actions, each with an owner and a target date drawn from its band, and the queue works its way down as capacity allows. That ordering is what a regulator wants to see: not that every finding was closed instantly, which is rarely possible, but that the building tackled its most serious risks first and can show why. Holding scored findings, their owners and their target dates in one audit record turns the scheme from a one-off grid into a living queue, and closing the loop on each item becomes a matter of working the list rather than remembering it. Our note on closing the loop on corrective actions covers that follow-through.

Consistency is the real prize

The single biggest benefit of scoring is not on any one audit. It is across all of them. When every auditor applies the same matrix, findings become comparable, trends become visible, and a portfolio can be managed by exception, attention going to the buildings and the bands that warrant it. A flat list gives you a hundred separate judgement calls. A scoring scheme gives you one agreed judgement, applied a hundred times. None of this is legal advice, and where a statutory timescale applies it takes precedence, but as a way to turn an audit into a plan that holds up, an honest severity-and-likelihood score is hard to better.