What an NDT program gap assessment samples, finds and sequences

What a gap assessment actually tests

A gap assessment is not an audit and not a certification review. It is a controlled comparison between three things that drift apart in every NDT program: the written practice as approved, the procedures as issued, and the work as actually performed and recorded. Auditors find gaps by sampling. A gap assessment finds them the same way, ahead of time, and adds the one thing an audit never provides — a remediation path with a realistic duration attached to each finding.

The distinction is commercial as much as technical. An audit produces findings the employer must answer under a clock set by someone else. A gap assessment produces findings the employer still controls, at a point where the sequencing decision remains open. The value is not the list; any competent Level III generates a list. The value is knowing which items close this week, which require re-examination, and which require elapsed calendar time that no amount of effort compresses.

Scope is set by the standard the employer is audited against rather than by preference. SNT-TC-1A programs, ANSI/ASNT CP-189 programs, NAS 410 and EN 4179 aerospace programs, and ASME Section V written practices each place different obligations on the same records. A gap assessment run against the wrong criteria produces clean findings and a failed audit. The first hour establishes which document governs, which edition the written practice invokes, and which customer specifications flow down on top of it.

Written practice versus actual practice: the core comparison

The written practice is a promise about behavior. The comparison asks whether the promise is kept, clause by clause. Methods listed in the document against methods actually performed on the floor. Training hours and experience months stated in the tables against what personnel files evidence. Examination structure described against examinations actually administered. Recertification interval stated against certification dates in force. Each mismatch is a finding regardless of which side of it is the more demanding.

The commonest direction of drift is a written practice more demanding than the shop. Employers adopt a template, sign it, and inherit obligations they never intended — a three-year recertification interval when SNT-TC-1A suggests five years as a maximum, annual color contrast testing when the standard calls for it at initial certification and five-year intervals, or a method still listed in scope that the company stopped offering years ago. Every one of those is a self-inflicted nonconformance against the employer's own document.

The opposite drift is more serious and less common: practice exceeding what the document authorizes. Technicians performing a technique the written practice does not describe, certifications issued at levels the document does not define, or limitations applied informally rather than recorded on the certification itself. SNT-TC-1A permits the employer to modify its program but not to eliminate training, experience, testing or recertification, and undocumented practice sits outside the program rather than inside a relaxed version of it.

How the records sample is drawn

A sample that only reads well-organized files proves nothing. The assessment draws across every method in scope, every certification level in use, and every location where NDT is performed, then adds targeted pulls: the newest certification, the oldest live certification, anyone certified during a Level III vacancy, anyone with a limitation on their certification, and anyone whose recertification date falls inside the audit window. Coverage of the population's edges beats depth in its center every time.

For each file, the assessment checks the content SNT-TC-1A calls for: the individual's name, level of certification, NDT method and technique with any limitations, educational background and experience, a statement of satisfactory training completion, current examination copies or evidence of successful completion, composite grades or suitable evidence of grades, and results of the vision examinations prescribed for the current certification period. Absent items are recorded as absent rather than inferred from surrounding context.

Inspection records receive a parallel sample. Reports are traced back to the procedure revision in force on the date of the examination, to the technician's certification scope on that same date, and to the calibration status of the equipment used. Traceability breaks surface here rather than in the personnel files: a report signed by a technician whose certification had lapsed four days earlier, or written to a procedure revision superseded three months before the examination took place.

Examination papers, grading evidence and retention

SNT-TC-1A requires the certification record to hold current examination copies or evidence of successful completion, along with composite grades or suitable evidence of grades. It fixes no retention period; the retention period is whatever the employer's own written practice states. That is a trap of the employer's own construction. Companies discard graded papers on a schedule nobody wrote down, then discover at audit that their written practice commits them to holding those papers for the full duration of certification.

The assessment therefore reads the retention clause first and samples against it. Where papers are missing, the question is whether the written practice required them and whether alternative evidence of grades exists. Where papers exist, the check moves to whether the general, specific and practical components are all represented, whether the composite grade calculation matches the written practice's own weighting, and whether the administrator's signature belongs to someone other than the candidate or the candidate's subordinate.

Specific examinations attract the most attention because they are the set most often copied. A specific examination is meant to test the employer's own procedures, equipment, acceptance criteria and product forms. One that asks generic method questions signals that the examination program was purchased rather than built, and it invites the auditor to test whether the Level III who approved the questions understood what they were approving. That line of questioning rarely ends at the examination paper.

Vision currency and why it dominates the findings list

Vision findings are the most frequent because vision is the only requirement with an annual clock attached to individuals rather than to documents. SNT-TC-1A calls for annual near-vision acuity examination — Jaeger Number 2 or equivalent at a distance of not less than 12 inches, or an equivalent such as an Ortho-Rater minimum of 8 — together with color contrast differentiation demonstrating the capability to distinguish the colors or shades of gray used in the method, at initial certification and at five-year intervals thereafter.

Aerospace programs diverge and the assessment must not blur them. NAS 410 and EN 4179 moved the near-vision criterion away from Jaeger 1 toward Snellen 20/25 or equivalent, on the basis that Jaeger notation was a printer's type size that converts inconsistently to acuity measures. Those standards also require the vision test to be administered by personnel trained and designated by the Responsible Level 3 or by qualified medical personnel — an administrator requirement, not merely a result requirement.

A vision gap has one property that makes sequencing simple: it closes in a day and cannot be closed retroactively. A test performed now evidences current acuity. It cannot evidence acuity during the eleven months it was overdue, and reports issued in that window carry the gap with them. Remediation is therefore two-part — restore currency immediately, then assess whether work performed during the lapse requires technical review or re-examination of the product.

Procedure and technique revision control

Procedures fail audits in three ways: they are not approved by a qualified Level III, they are not the revision actually in use, or they are not qualified for the technique being performed. Revision control is where the written practice meets the quality system, and it is the boundary most often left to a document controller rather than to the Level III. A procedure list that does not match the procedures in the technicians' hands is a finding on both sides of that boundary at once.

The assessment traces a specific revision through its whole life: approval signature and date, distribution record, the technique sheets derived from it, the training or briefing given when it changed, and inspection reports written under it before and after the change. Uncontrolled copies surface in this trace — a laminated sheet at a workstation, a PDF on a tablet, a technique card in a toolbox — because the report cites a revision the controlled set already retired.

Nadcap experience shows where this concentrates. The NDT task group publishes its most frequent nonconformances annually on eAuditNet, and self-audit and calibration flow-down requirements sit near the top of that list. Calibration of process instrumentation — ultrasonic and eddy current equipment, gauss meters, light meters for penetrant and magnetic particle work, and reference standards for sensitivity verification — is among the highest-rejection categories, and calibration status is what makes a procedure revision executable rather than merely approved.

The findings taxonomy and what each class costs

Findings sort into classes by remediation path, and the class matters more than the severity label attached to it. Documentable-now findings have evidence that exists but is not in the file or not in the required format: a training certificate sitting in a manager's email, an experience log in a spreadsheet, a limitation applied in practice but never written onto the certification. These close in days and cost administrative time only, which is why they belong last rather than first.

Re-examinable findings require an examination event: an invalid administrator, a missing examination component, a composite grade that cannot be reconstructed, or a certification issued without the specific examination the written practice requires. These close in weeks and need a qualified Level III, an approved question set, and — for practical components — specimens carrying known discontinuities. They are the class most consistently underestimated when a remediation plan is built backwards from the audit date.

Time-bound findings require elapsed calendar time that no effort compresses: missing training hours, insufficient documented experience, or a certification level the individual has not yet earned. The honest close is a documented restriction on scope until the time is served. Structural findings sit apart from all of these — an absent Level III designation, a written practice on the wrong basis standard, or no internal audit program at all. Structural findings invalidate the classes beneath them and close first.

Deliverable format and remediation sequencing

The deliverable is three documents rather than a report. First, a clause-by-clause matrix: every requirement of the governing standard on the left, the employer's written practice text against it, the observed evidence, and a conformity verdict. This is the document that survives the audit, because an auditor can read a single row and see exactly what was checked, what was found, and what the employer concluded.

Second, a findings register carrying, for each item, the class, the affected personnel or documents, the required action, the responsible owner, the minimum elapsed time, and a target date. Third, a sequenced remediation plan that orders those items by dependency rather than by severity, with the audit date fixed as the terminal constraint. Severity ordering is the standard mistake: it starts the loudest finding and finishes the slowest one late.

Sequencing runs in one order. Structural first, because everything else inherits from it — a written practice approved on the wrong basis makes every downstream examination arguable. Time-bound second, because those clocks start the moment they are identified and run whether or not anyone is working on them. Re-examinable third, ordered by specimen availability. Control findings fourth. Documentable-now last, because it is the only class that reliably compresses into the final two weeks.

Where the gap assessment sits on the audit calendar

Ninety days before the audit is the point where all classes remain closable. At sixty days, time-bound findings convert to documented restrictions rather than closures. At thirty days, re-examinable findings compete with specimen lead times and the assessment becomes a triage exercise against the calendar. Running the assessment immediately after the previous audit, rather than shortly before the next one, is what turns it from a fire drill into a program control that pays for itself across cycles.

Aerospace programs tie the timing to the accreditation cycle directly. Initial Nadcap accreditation runs twelve months, and eighteen- and twenty-four-month merit depend on consecutive cycles clear of Non-Sustaining Corrective Actions, verification of corrective action audits and Type C auditee advisories. A gap assessment run early in a cycle protects merit; one run in the final month protects only the accreditation itself. Sequencing against AC7114 criteria is set out in /consulting/nadcap-ac7114-audit-readiness.

Sector context changes the sample rather than the method. Refinery and petrochemical programs carry API in-service inspection records and OSHA process safety obligations alongside the personnel files, addressed in /consulting/petrochemical-ndt-consulting and /consulting/oil-gas-ndt-consulting. Atlantis runs gap assessments as fixed-scope engagements against a named audit date and hands over the clause matrix, the findings register and the sequenced plan. Affordable, accessible, fully customizable — request a scoping call or a quote.

Request a consultation