A test report is the product. The testing is what makes the report true, but the report is the only part the client, the regulator, the insurer and eventually the court ever sees, and it either stands on its own or it does not. ISO/IEC 17025:2017 sets out what standing on its own means, in clause 7.8, and it is more specific than most people expect.
TCR Engineering Services Pvt. Ltd. has reported under that standard since accreditation and holds NABL certificate NABLT0726MH18640 across 1,483 scope items. What follows is the standard's requirement, not a house style, so it is the checklist to hold any laboratory's report against.
What is a materials testing laboratory report?
A materials testing laboratory report is the authorised record of what was tested, by what method, under what conditions, and what the result was. ISO/IEC 17025:2017 requires that results are reviewed and authorised before release, and that they are "provided accurately, clearly, unambiguously and objectively".
Two words in that sentence do the work. Authorised means a named person carries the result, not a laboratory in the abstract. Objectively means the report separates what was measured from what anybody thinks it means.
What must every accredited test report contain?
Clause 7.8.2 lists the items common to every report, and a report missing one of them is incomplete against the standard rather than merely thin. They are worth reading as a checklist, because most report disputes turn on an item from this list being absent rather than on the number being wrong.
| Clause 7.8.2 item | What it means in practice |
|---|---|
| a) A title | "Test Report", so the document's status is not in doubt |
| b) Name and address of the laboratory | The legal entity that stands behind the result |
| c) Location of performance | Where the work was done, which matters for site testing |
| d) Unique identification of every component | So no page can be separated from the report |
| e) Name and contact information of the customer | Who the report was issued to |
| f) Identification of the method used | The standard, with its part, section and year |
| g) Description and unambiguous identification of the item | And its condition where that matters |
| h) Date of receipt of the item, and the date of sampling | The chain starts here |
| i) Date or dates the laboratory activity was performed | Not the same as the issue date |
| j) Date of issue of the report | |
| k) Reference to the sampling plan and sampling method | Where the laboratory sampled |
| l) A statement that the results relate only to the items tested | The scope limit, stated |
| m) The results, with units of measurement where appropriate | |
| n) Additions to, deviations from, or exclusions from the method | The honesty clause |
| o) Identification of the persons authorising the report | A name, not a department |
| p) Clear identification of results from external providers | Subcontracted work, marked as such |
Item n) is the one that separates a defensible report from a comfortable one. A method deviation that is recorded is a professional judgement; the same deviation unrecorded is a finding waiting to be made.
When must measurement uncertainty be reported?
Under three conditions, and clause 7.8.3.1 c) names them: when uncertainty "is relevant to the validity or application of the test results", when "a customer's instruction so requires", or when "the measurement uncertainty affects conformity to a specification limit". Where it is reported it is presented in the same unit as the measurand.
Buyers get this wrong in both directions. Some expect an uncertainty figure on every line of every report, which the standard does not require. Others treat any mention of uncertainty as the laboratory hedging, when it is the opposite: a number without a stated uncertainty is a number whose fitness for the decision cannot be judged.
The third condition is the one that bites. When a result sits near a specification limit, the uncertainty decides whether the item passes, and a report that gives the number without the uncertainty has not answered the question it was asked. That is also why the next section exists.
What is a decision rule, and why must the report state it?
A decision rule is the documented rule for how measurement uncertainty is taken into account when the laboratory says an item conforms or does not. Clause 7.8.6.1 requires that the laboratory "shall document the decision rule employed, taking into account the level of risk associated with the decision rule", and that the statement of conformity identifies which results it applies to, which specifications or parts of them are met or not met, and the decision rule applied, unless that rule is already inherent in the specification requested.
Put plainly: "Pass" is not a result. "Pass, against clause X of specification Y, under decision rule Z" is a result. The difference is whether a reader can reconstruct the judgement or has to trust it.
- Which results the conformity statement covers, since a report may carry results it does not sentence.
- Which specification, to the clause, and which parts are met and not met.
- Which decision rule, unless the requested specification already fixes it.
This is the clause that most often turns a report into an argument. Two laboratories can measure the same value and reach opposite conformity statements without either being wrong, because they applied different decision rules. If the rule is not on the report, that disagreement cannot be resolved on paper.
Who may write an opinion in a report?
Only personnel authorised to express them, and the opinion has to be identifiable as an opinion. Clause 7.8.7 requires that the laboratory ensures "only personnel authorized for the expression of opinions and interpretations release the respective statement", and that opinions "shall be based on the results obtained from the tested or calibrated item and shall be clearly identified as such".
That second requirement is the useful one for a reader. In a properly written report the measured results and the interpretation of them are visibly different things, and the interpretation is traceable to the results in the same document. An interpretation that rests on something outside the report, on experience or on what the client expects, does not belong in it.
What makes a report accepted outside the country it was issued in?
Accreditation to ISO/IEC 17025 by a signatory accreditation body, and the mutual recognition arrangement that sits above it. NABL is a signatory to the ILAC Mutual Recognition Arrangement, under which an accredited report is accepted across 90+ economies, which is what allows an Indian test certificate to be relied on by an overseas owner without retesting.
Two things are worth being exact about. Acceptance follows the accredited scope, not the laboratory: a test outside the scope on the certificate travels no better than an unaccredited one, whatever the letterhead says. And the certificate is a public record, so it can be checked rather than taken on trust. TCR Engineering's is NABL certificate NABLT0726MH18640.
What to check when a report lands on your desk
Run the same five checks on every report, whoever issued it. They take a minute and they catch almost everything that later becomes a dispute.
- Is the method identified to its part, section and year? "As per ASTM" is not a method.
- Is every result inside the laboratory's accredited scope, and does the report say which are not? The scope is on the certificate, and the certificate is public.
- If there is a pass or fail, is the decision rule stated? If not, you have a judgement without its basis.
- If a result sits near a limit, is the uncertainty given? Clause 7.8.3.1 c) says it should be.
- Are deviations from the method recorded? Silence here is the item that ages worst.
A report that survives all five is one you can hand to a regulator without a covering explanation. That is the only test of a report that matters.