The underwater inspection report is the real deliverable of the job: what the client is buying isn't the dive, it's organised, traceable evidence of what's underwater. A clear structure, every data point referenced to a specific position, findings assessed against an agreed criterion, and a strict separation between fact and opinion.
A client who commissions an underwater inspection never sees the structure they've paid to have checked again. Everything they'll know about it comes through a document. That's why, in this line of work, the report isn't the final formality — it's the product. A flawless inspection that's poorly documented is worth very little, and a sound inspection that's well documented can support years of maintenance decisions, investment decisions and even liability cases.
Anyone starting out in inspection tends to find this out late. The visible part of the trade is the water; the part that decides whether you get hired again is the desk.
What the report is actually used for
Before deciding how to write it, it's worth being clear on who will read it and why. The same document can end up in very different hands:
- El the maintenance manager, who needs to know what has to be repaired, how urgently, and with what resources.
- El the structural engineer, who will compare the data against previous campaigns and against design criteria.
- La management, who wants a conclusion and a priority order without reading sixty pages.
- La the insurer or the regulator, who requires proof that the inspection was carried out as required.
- Un tercero futuro, in the event of litigation, an incident, or the sale of the asset.
Writing with only the first of these readers in mind is the classic mistake. A good report serves all of them at once because it's layered: a summary for whoever decides, detail for whoever calculates, and evidence for whoever has to prove something.
Standard structure of a report
Formats vary between companies and clients, but the underlying structure is fairly stable.
1. Cover page and identification
Client, asset inspected, job reference number, dates of execution, contracting company, personnel responsible and the document's revision control. It sounds bureaucratic, and it is, but it's exactly what keeps the report traceable and citable ten years later.
2. Resumen ejecutivo
One or two pages answering three questions: what was inspected, what was found, and what needs to be done. Findings ordered by severity, with no unnecessary jargon. It's the most-read part, and the one written last.
3. Scope and methodology
What was asked to be inspected, what was actually inspected, and with what means. This is where the technique used goes —general or detailed visual inspection, thickness measurement, prior cleaning — the equipment used with its calibration data, the environmental conditions and, very importantly, any deviations from the planned scope. If an area couldn't be checked because of visibility, current or access, it's stated, the reason is explained, and it's recorded as not covered.
4. Reference system
The section that separates a professional report from a photo album. Before presenting a single finding, you need to define how every point on the structure is named: element naming, axes, levels, numbering direction, coordinate origin. If the client already has their own system, you adopt it rather than inventing a parallel one.
5. Hallazgos
The body of the report. Each finding treated as a stand-alone unit with its own identifier, exact location, objective description, dimensions, visual evidence and assessment.
6. Conclusions and recommendations
Kept separate from the findings and clearly flagged as interpretation. This is where the inspector applies judgement: what's urgent, what can wait for the next cycle, what's worth keeping an eye on, and what information is missing to make a decision.
7. Anexos
Personnel and equipment certificates, dive log, full measurement tables, photographic catalogue and video. The logic is the same as governs document management and as-built records in underwater works: the appendix holds the raw evidence so anyone can retrace the reasoning.
Referencing: where quality is won or lost
An underwater data point without a position isn't a data point. "Significant corrosion on the pile" is a useless sentence; "localised thickness reduction on pile P-14, north face, between levels −3.0 and −3.6 m relative to the quay reference" is information you can act on.
The practical rules of referencing are few, and they're not negotiable:
- A single coordinate system for the whole report, explicitly defined and consistent with the client's own system.
- Origin and direction stated. Where measurement starts from, which way the numbers increase, and what level elevations are given against.
- A unique identifier per finding, repeated identically in the text, the tables, the photos and the sketches.
- Traceability back to the source. Every item in the report should be traceable to the video or data file it came from, with its timestamp.
Photography and video
Underwater images are deceptive on scale. A photo of a defect with no dimensional reference doesn't let you assess anything, so standard practice is to include a physical scale in the shot or, failing that, an object of known size. It's also worth combining a wide shot — to locate — with a close-up — to assess — and keeping a caption that repeats the identifier, location, view orientation and date.
Continuous video, when it's recorded, works as the master record: it's what lets you go back if a question comes up months later. That's why timestamping and the diver's running commentary during the inspection have real documentary value and aren't just decoration.
Measurements
Every measurement is presented with its method, its equipment, its calibration date and its uncertainty. This matters especially in ultrasonic thickness measurement: the result depends on surface preparation and calibration, and a figure without context can lead to the wrong conclusions about a structure's condition.
When previous campaigns exist, showing the trend is worth more than the absolute value. A given thickness loss figure says little on its own; that same figure compared with the measurement from four years ago says whether the problem is progressing, and how fast.
Assessment: criterion before label
Putting "critical" or "acceptable" next to a finding without explaining what it's being measured against is one of the fastest ways to lose credibility. Assessment needs an explicit criterion, agreed with the client before work starts: the project specification, the applicable standard, an internal procedure of the asset owner, or a severity scale defined in the contract itself.
The discipline underpinning all of this is a strict separation between three things:
- Observation: what's seen and measured. Verifiable by anyone who reviews the evidence.
- Assessment: what that observation means against the agreed criterion.
- Recommendation: what should be done, with the reasoning behind it.
Mixing these up is the most common flaw in inspection reports, and the one that causes the most problems when the document is reviewed years later with fresh eyes.
What the client expects and rarely asks for in writing
There are expectations that never appear in the spec but decide whether a report is considered good:
- That it arrives promptly. An excellent report delivered two months late has already lost much of its usefulness for scheduling maintenance.
- That it's comparable. If there's a history of previous campaigns, the format should allow direct comparison with them.
- That it's honest about its limitations. Clearly stating what couldn't be inspected protects the client and the inspector alike.
- That it's usable. Manageable files, traceable images, exportable tables. A report nobody can open might as well not exist.
- That it distinguishes what's urgent from everything else. Forty findings with no priority order force the client to do the work they'd paid someone else to do.
Mistakes that keep repeating
The same ones, campaign after campaign: vague descriptions with no dimensions, photos with no scale or caption, naming that changes halfway through the document, no list of what wasn't inspected, assessments with no stated criterion, generic recommendations like "monitoring is recommended" with no detail on what or when, and contradictions between the executive summary and the body of the report.
Almost all of them have the same root cause: writing the report days later, with the memory still fresh but the data disorganised. The fix isn't better writing, it's better capture during the work, with naming fixed before the first dive and notes taken on the spot.
A skill that pays off
A diver or inspector who can deliver clean documentation holds a different place on the team. They're the one who ends up handling contact with the client's representative, the one who takes on the jobs that need a formal deliverable, and the one with a natural route towards supervisory and technical-office roles, which account for a good share of the industry's skilled job market. It's also one of the reasons why qualifying as an underwater inspector is one of the specialisations with the best return within the industry, and a recognisable piece of the wider map of professional certifications.
Put another way: plenty of people can go down and look. Going down, looking and recording it so someone else can decide on the strength of it — far fewer can do that. And that's exactly what the client is buying when they commission an inspection, whatever the order form actually says. If you're building your profile, it's worth treating report writing as part of the trade from day one, the same way you'd treat any other technical skill in the commercial diver's career path.
Frequently asked questions
What's the first thing a client reads in an inspection report?
The executive summary. One or two pages stating what was inspected, what was found in order of severity, and what needs to be done. It's written last, once the findings are finalised, and it has to be consistent with the body of the document.
Why does the reference system matter so much?
Because an underwater data point without an exact position gives you nothing to act on. Before presenting findings you need to define naming, axes, levels, origin and numbering direction, and use the client's own system if one already exists. Every finding carries a unique identifier that's repeated exactly in the text, tables, photos and sketches.
How should underwater photographs be presented?
With a dimensional reference in the shot, combining a wide shot to locate with a close-up to assess, and with a caption repeating the identifier, location, view orientation and date. An underwater image with no scale doesn't let you judge the size of a defect.
Should you include what couldn't be inspected?
Yes, always. Deviations from the planned scope — inaccessible areas, insufficient visibility, current — are stated explicitly. Leaving them out lets the client believe they have coverage where they don't, and exposes the inspector if a problem later turns up in that area.
How is a finding assessed as critical or acceptable?
Only against an explicit criterion agreed before starting: the project specification, the applicable standard, a procedure from the asset owner, or a contractual severity scale. Labelling a finding without stating the comparison criterion undermines the report's credibility.
What's the difference between observation, assessment and recommendation?
Observation is what's seen and measured, verifiable by anyone with the evidence. Assessment interprets that observation against the agreed criterion. Recommendation proposes an action. Mixing them up is the most common flaw, and the one that causes the most problems when the report is reread years later.
Published 9 September 2026 · By Arthur More, professional saturation diver
Photo: U.S. Navy Seabee Museum · CC BY via Openverse