Six Months of Identical pH Readings
At a food processing plant discharging to sewer under a consent of pH 6.0 to 9.0, the auditor is shown the daily effluent pH log. Every entry for six months falls between 6.8 and 7.2, including days when the inline monitor logged 5.2 and the regulator's spot sample recorded 5.4. The operator who keeps the log says the handheld pH meter has not been calibrated since the buffer solutions ran out, and that she 'sometimes writes the figures in at the end of the week'. The EHS manager says the meter is checked daily against the inline monitor. The calibration procedure requires weekly calibration and retention of the calibration record. The plant reports quarterly compliance returns to the sewerage undertaker.
1. Assess the evidence. Do you raise a nonconformity, and against which requirement or requirements?
Read the worked answer
Evidence base: the log, the inline monitor data, the regulator's sample, the operator's statement, the EHS manager's statement, the calibration procedure, and the absence of calibration records. These are mutually corroborating and the records contradict the instrument data.
Requirements engaged:
- 9.1.1: the organisation must determine methods for monitoring and measurement, and ensure calibrated or verified monitoring and measurement equipment is used and maintained. Uncalibrated equipment feeding a compliance-critical parameter is a direct failure.
- 7.5.3: documented information must be controlled, including protection from loss of integrity. Records created retrospectively from memory for a compliance parameter are not protected against inaccuracy; they are inaccurate.
- 9.1.2: evaluation of compliance is supposed to rest on reliable data. If the quarterly returns were based on this log, the organisation's evaluation of compliance is unreliable and may have misreported to a third party.
Conclusion: raise a nonconformity. The calibration and data-integrity failures are the same causal chain, so cite 9.1.1 as the headline requirement and reference 7.5.3 and 9.1.2 in the description. This is systemic and touches legal compliance, so it is material rather than minor. Record the specific dates and readings you verified, not a general statement that records are unreliable.
2. How should the auditor handle the operator's admission and the conflict with the EHS manager's account?
Read the worked answer
During the interview:
- Stay with the process, not the person. Ask how the reading is taken, what happens when the buffer runs out, what the procedure says, and what is done with a deviating result. The operator's account emerged unprompted; do not lead or press for names or guilt.
- Do not adjudicate credibility. You are testing whether the process and its records are reliable, not whether an individual is dishonest. The EHS manager's 'checked daily against the inline monitor' claim is testable: ask for the verification record. Its absence is objective evidence; the contradiction is then a matter of records, not of personalities.
- Close the loop with the process owner later, face to face, away from the shop floor, presenting the specific data points. Fair presentation requires the auditee to see and understand the evidence before the finding is written.
Escalation: where evidence suggests records were created to show compliance, and a regulatory return may be affected, the lead auditor should inform the audit client/auditee management at the closing meeting and let the organisation decide its legal reporting obligations. The auditor does not advise on the organisation's regulatory disclosure, does not communicate with the regulator, and maintains confidentiality (audit principles of integrity, fair presentation and confidentiality). Note the matter in the report factually.
3. How would you determine the extent of the problem, and how do you record the sampling?
Read the worked answer
Extent testing:
- Trace backwards: how long has the buffer been out of stock? Purchase records and stock records give an objective start date.
- Trace sideways: which other parameters use the same instrument or the same practice (for example boiler or cooling-tower chemistry), and which other shifts keep the same log?
- Trace forwards: which quarterly returns to the sewerage undertaker were built on these data, and has the organisation since evaluated compliance on that basis?
- Compare the inline monitor's calibration status too; do not assume the electronic record is the reliable witness without checking it.
Recording: state the sample explicitly in the finding and working papers, for example: reviewed the pH log for the period 1 April to 30 September; compared all 183 entries against the inline monitor; visited two of three shifts; interviewed the log keeper and the EHS manager on given dates. This lets a reader judge whether the finding is an isolated lapse or, as here, a practice.
Beware the trap: the six-month consistency is itself the clue. Uniform readings for a variable parameter are a data-integrity indicator in any audit, and the auditor who samples only three days will miss it.