Owners commission condition assessments expecting a verdict on whether the building is in good shape. What makes an assessment useful is something narrower and more actionable: how long each significant asset has left, what it costs to replace, and what happens if it fails before it is replaced.
The distinction sounds academic until you try to use the document. A report full of condition ratings tells you the state of things. It does not tell you what to do first, and the entire reason to commission the work is to know what to do first.
What should be in it
- Asset inventory. What exists, where, of what type and capacity, installed when. This sounds trivial and is frequently the most valuable part, because many organizations do not have a current one.
- Condition, with the basis stated. Not just a rating but how it was determined: visual observation, testing, records review, or estimate from age. These carry very different confidence and the difference should be visible.
- Remaining useful life. The number the capital plan actually needs, expressed as a range rather than a point.
- Replacement cost. Current cost, with the basis noted.
- Consequence of failure. What happens operationally if this asset fails unexpectedly. This is what converts a condition report into a priority list, and it is the field most often missing.
Consequence is the field that does the work
Two assets can be in identical condition with identical remaining life and belong in completely different places on a priority list, because one of them takes the building offline when it fails and the other does not.
An assessment that ranks purely by condition will put a badly deteriorated non critical item above a moderately worn critical one. That ranking is wrong for almost any purpose an owner has. Ranking by expected consequence, meaning probability of failure multiplied by what failure costs, produces a list that survives contact with a budget process.
When commissioning an assessment, specify that criticality and consequence of failure are required fields. Many standard scopes do not include them, and adding them afterward means revisiting assets.
Common failure modes
Too much precision on things that were estimated. A remaining life figure derived from an equipment age table and one presented as though it came from testing look the same on the page. The report should distinguish them, because a capital plan built on age tables is considerably less reliable than one built on condition testing and the owner deserves to know which they have.
No treatment of interdependence. Assets are listed individually, but they fail as systems. Replacing an air handler while leaving the distribution and controls it depends on untouched produces a fraction of the expected benefit. Good assessments flag where items should be bundled.
Written to be filed rather than used. If the deliverable is a long narrative document, it will be read once. The useful form is structured data you can sort and filter, with the narrative as supporting material. Ask for the underlying data in a usable format as a contract deliverable, not as a favor.
Never updated. An assessment is a snapshot and it decays. Work gets done, conditions change, and by year four it is being cited with a confidence it no longer warrants. Build in a routine for updating it as work completes, otherwise you will be recommissioning the whole exercise.
What it costs and what it saves
Assessments are not cheap and the value is indirect, which makes them hard to justify on a spreadsheet. The argument that tends to work is not about the assessment itself but about the decisions it enables: capital planning that is not guesswork, a defensible funding case, and avoidance of the emergency replacement premium that comes from discovering an asset's condition when it fails.
The corollary is that an assessment which does not enable those decisions has not earned its cost, regardless of how thorough it looks. That is the standard to hold it to, and it is worth writing into the scope before anyone is engaged.