Sello runs Soetdoring Cold Stores in Rosslyn, Pretoria, a third-party logistics provider (3PL) with frozen and chilled chambers and its own reefers. Its recertification audit in March ended with one major non-conformity and twelve minors. The major concerned the alarm call-out: for six weeks the chamber alarms had been dialling the phone of an engineer who had resigned, and nobody had tested the list. The grade was C. Nandi, the technical manager, wanted to start on root causes. Sello’s first questions were about the business. What will our customers see? Does a C mean two audits a year, and what does that do to our managers’ time? Can we appeal if we think the auditor got it wrong? And Jaco, the transport manager, asked the question nobody had asked before: who decides what happens to the stock the next time a chamber fails at night? A BRCGS Storage and Distribution audit grade answers some of these questions, and the decisions behind it answer the rest.
In short
- A BRCGS Storage and Distribution audit grade is a count: critical, major and minor findings produce AA, A, B, C or D, with a + after an unannounced audit, or no certificate at all.
- AA, A and B sites are audited yearly; C and D sites every six months, under a BRCGS position statement that is worth re-checking before you plan.
- Evidence for every finding is due within 28 calendar days (90 after a first certification audit). Miss it and no certificate is issued.
- At least one audit in every three years is unannounced. You can plan the year, the deputies and the non-audit days, never the day.
- After a temperature failure, the safety and quality of the stock is decided with the product owner before release (clause 7.3.5).
- The Standard asks for continuity plans for losing energy, not for a generator. A recall affecting the site goes to the certification body immediately.
What a BRCGS Storage and Distribution audit grade actually measures
The grade is arithmetic, not an impression of how the site looked. Every finding lands on one of three levels. Critical means a serious breakdown on product safety or the law. Major means the site falls well short of a clause or of a statement of intent, or the auditor’s evidence casts real doubt on whether the product or service is right. Minor means a requirement is only partly met, while the product itself is not in question. The same minor written against the same clause time after time may be upgraded to a major, and an auditor may not fold several minors into one. The certification body’s (CB’s) management confirms the grade after a technical review; the auditor does not award it on the day, and may not tell you how to fix anything.
With no majors, 5 or fewer minors is AA, 6 to 10 is A, 11 to 16 is B, 17 to 24 is C and 25 to 30 is D. Each major costs roughly one band: one major with up to 10 minors is a B, two majors with up to 10 minors a C. A single critical, a third major, or a minor count beyond the D band leaves the site uncertificated. Our guide to BRCGS Storage and Distribution certification covers the grading rules and audit options; here the question is what each outcome means for the business. The commitments that keep findings away in the first place, from the culture plan to management review, are covered in our guide to BRCGS Storage and Distribution management responsibilities.
| Outcome | Next audit | Closing the findings | What it means for the business |
|---|---|---|---|
| AA, A or B (AA+ to B+ if unannounced) | In 12 months | Evidence within 28 calendar days | An annual rhythm; customers reading the report see each finding and its root cause |
| C (C+) | In 6 months | Evidence within 28 days; a revisit within 28 days where two majors were raised | Two audits, two sets of fees and two close-outs a year until the grade recovers |
| D (D+) | In 6 months | A revisit within 28 calendar days before a certificate is issued | As for C, plus a revisit to prepare for within four weeks |
| Not certificated | A full re-audit, at least 28 days after the failed one | No certificate until the re-audit succeeds | Any existing certificate is withdrawn straight away; customers who require notice must hear at once |
How your customers read the grade
Besides the letter, the certificate carries the scope, the type of audit programme, when the audit happened, when the next one is due and when the certificate expires. Customers can check it on the BRCGS Directory at brcgsdirectory.com using your seven-digit site code, and the public search shows only sites that are currently certificated, so a lapse is visible to anyone who looks.
The report tells a fuller story. It is yours, not the CB’s, and you can give customers access through the Directory, where you cannot alter it. A customer’s technical manager who opens it sees each non-conformity with its correction, root cause and preventive action plan. Anyone with shared access also gets automatic emails if the certificate is suspended, withdrawn or expires without a replacement. In practice the letter starts the conversation and the root cause analysis finishes it: a C with convincing preventive actions can read better than a B whose root causes all say “human error”. Each customer still makes its own decision about using you, and some may accept your report instead of auditing you themselves.
Sello rang his two largest customers before the report was uploaded, explained the major and offered the evidence pack. The grade still had a cost: two audits a year until the next result comes in at B or better.
Twelve months or six: what a C or D costs in time
Since 1 February 2023 a BRCGS position statement has applied one rule to every site, whatever it handles: an audit every 12 months after an AA, A or B, and every 6 months after a C or D (the book version allowed longer gaps for certain consumer-products-only sites). Position statements are binding and collected in the SD404 document on brcgs.com, so check the current version before you plan.
A six-month cycle doubles more than the audit days. Every audit brings the CB’s fees plus a BRCGS fee, and neither report nor certificate counts until both are settled; every audit also opens another 28-day close-out. The one-in-three unannounced rule still applies, with announced audits in between, and in the unannounced year a site on a six-month cycle may nominate only 5 non-audit days instead of 10. The way back to an annual cycle is the next grade.
Due dates follow fixed rules. Each new one is set by stepping forward from the last due date, never from the day of the audit, and an announced audit must fall in the four weeks that end on the due date. Bringing an audit forward, say to line it up with another scheme, moves the due date to 12 months after that early audit. Holding it late without an acceptable excuse earns a major next time (clause 1.1.10), and building work or a staff shortage is not an acceptable excuse. Soetdoring’s calendar now looks like this:
| Step | Soetdoring date | Rule behind it |
|---|---|---|
| Recertification audit (announced) | 3 March | Inside the 28 days up to and including the 15 March due date |
| Evidence for all 13 findings | By 31 March | 28 calendar days from the audit date |
| Report and certificate | Typically by mid-April | Typically within 42 calendar days of the audit |
| Next due date | 15 September | 6 months on from the previous due date, because the grade was C |
| Next announced audit window | 19 August to 15 September | The 28 days up to and including the due date |
| Due date after that, if the grade is B or better | 15 September next year | 12 months on from the previous due date |
The 28-day clock is a management project
The closing meeting starts the clock, with a written summary of findings at the meeting or within 1 working day. Each finding needs three answers: the correction (what you fixed straight away), the root cause, and a preventive action plan with dates, which is printed in your report. The proof, such as revised procedures, records, photographs and invoices, has 28 calendar days from the audit to reach the CB, or 90 days after a site’s very first certification audit. A temporary fix can be accepted where the permanent one needs longer, and unconvincing evidence can lead to a revisit. A revisit is required for any D, and for a C with two majors, within 28 calendar days; anything new found that day must be resolved too. Miss the deadline and no certificate is issued: the site starts again with a full audit.
Close-outs rarely stall on the quality work; they stall on decisions that need a manager with authority. Decide those in advance:
| Where close-out stalls | The decision that unblocks it | Reference |
|---|---|---|
| The fix needs a capital repair to a door, panel or rack | Approve a recorded, time-limited temporary repair now and date the permanent one | Part III 2.4; 6.2.6 |
| The root cause is staffing or shift cover | Decide the resource; retraining alone rarely convinces a CB | 1.1.3; 3.3.3 |
| The evidence depends on a contractor or haulier | Escalate through the contract and the supplier review | 3.5.1.4; 3.5.2.3 |
| The finding owner is on leave | Reassign the finding on day one to the named deputy | 1.3.3 |
| You believe a finding was graded wrongly | Appeal in writing within 7 calendar days of the decision; the CB decides within 30 and may charge if the appeal fails | Part III 6.4 |
At Soetdoring the correction took an hour: the call-out list was fixed and tested. The root cause was that the list belonged to nobody once the engineer left. The preventive action gave it to the maintenance manager, added a monthly test call at an awkward hour, and put it on the leaver checklist. If that cure has not held by the next audit, clause 1.1.11 turns it into a new finding. Our comparison of root cause analysis tools and our guide to closing an audit finding so it stays closed show what a CB tends to accept.
Planning for the unannounced year
On the announced programme, no certificated site goes three years without an unannounced audit, which takes the place of that year’s announced one. Within 3 months of the previous audit the CB names the year, never the date, and the visit comes at any time in the last 4 months before the due date, during normal operations. Failing it means a full re-audit, which may be announced, but the three-year rule still stands, and a site that lets three years pass without one risks BRCGS rejecting its last audit.
The date is not yours to plan, but almost everything around it is:
| Decision | When | Why it matters |
|---|---|---|
| Confirm which year the CB has named | Within 3 months of the previous audit | Leave and peak plans for that year assume an audit on any working day in the window |
| Nominate non-audit days, with reasons | At least 4 weeks before each date | Up to 10 days, or 5 on a six-month cycle; a customer’s own audit is a sound reason, a manager’s leave usually is not |
| Notify days you will not be operating | As soon as they are planned | They do not count against the allowance; an auditor who finds nothing in scope running cannot audit, and must come back |
| Brief deputies for every key role | Before the window opens | The site inspection starts within 30 minutes of arrival; an absent manager does not stop it |
| Keep the CB contract and site details current | All year | A notified unannounced audit survives a change of CB, and the new CB must see your last report |
Refusing entry to an unannounced auditor costs more than any finding: the certificate is suspended until another surprise audit has taken place, possibly as much as 4 months on, and the bill for the wasted trip usually lands with the site. Thinking of changing CB close to an audit? Check the current position statements first.
Some sites opt into the voluntary unannounced programme, telling the CB within 3 months of the last audit. Customers can take that as evidence of a mature system and may cut back their own audits, so it is a commercial decision as much as a technical one.
Temperature failures: decisions you share with the product owner
The Standard sets no temperature limits; they come from the owner’s specification and the law where the goods are sold. What it does set is who decides after something fails. After an equipment failure, nothing is released until the site and the product owner have jointly settled whether the stock is still safe and what the failure did to its quality, with the outcome on record (clause 7.3.5). Meanwhile the stock stays on hold until the investigation ends (clause 3.8.2), the owner is told (clause 3.8.3), and held stock is identified and secured so it cannot ship by mistake or deteriorate further (clauses 3.9.1, 3.9.2). Because product was put at risk, root cause analysis is compulsory (clause 3.3.3). If a probe was reading wrong, the out-of-tolerance procedure covers product that relied on it (clause 6.3.5).
The site never signs off affected stock alone, however confident the engineer sounds at 03:00. Sello’s team settled five questions in daylight:
- Who may put stock on hold at night without asking anyone? At Soetdoring, any shift supervisor. Releasing stock needs the owner’s decision.
- Who rings each product owner, and is there a tested after-hours number for every one of them?
- What will the owner need to decide? Logger downloads, alarm and response times, product temperatures and how long the stock was exposed.
- Where does held stock go so that it cannot be picked, and who blocks it in the warehouse system?
- Who decides whether the event is a significant product safety incident that the CB must hear about within 3 working days (clause 3.8.6)?
After a failed compressor or evaporator is replaced, the chamber’s ability to hold temperature is re-established where appropriate, for example by mapping it again (clause 7.3.8). Our article on what auditors check first in a warehouse audit covers the alarms, mapping and records an auditor walks through on the floor.
Power cuts and the generator: a management choice, not a clause
For a South African cold store, long power cuts are a foreseeable risk. The continuity clauses expect a plan for serious incidents, including the loss of electricity or water (clause 3.8.4), naming an incident team with after-hours numbers, the backup arrangements that let you keep serving customers, and how you will inform customers and, if necessary, the authorities (clause 3.8.5). Nowhere does the Standard demand a generator. Standby power, diesel supply and battery backup are ways of meeting the plan, and every option you pick creates duties of its own:
| Option | What it protects | What it commits you to | Clauses |
|---|---|---|---|
| Standby generator sized for the refrigeration load | Chamber temperatures through long outages | Planned maintenance and recorded test runs under load | 6.2.1; 3.8.4 |
| Diesel delivery agreement that also works after hours and on holidays | A generator that keeps running when outages run long | Supplier approval, a written specification and performance monitoring | 3.5.1 |
| Battery or UPS backup for loggers, alarm diallers and the warehouse system | Records and call-outs when mains power drops | Alarms that still reach a person; backed-up records | 7.3.2; 3.1.3.2 |
| Standby agreement with another cold store | Customers’ stock if a chamber cannot be recovered | Treating it as subcontracted storage: contract, approval, register | 3.5.2; 3.8.5 |
| Manual temperature rounds while systems are down | Evidence that product stayed within its limits | Recorded checks, typically every 4 hours or often enough to act in time | 7.3.2 |
Record the choice and its reasoning in the business continuity plan, revisit it at management review, and test it: run the generator under full refrigeration load and ring the alarm call-out with the mains switched off. Soetdoring’s major came from the same kind of gap, a control that existed on paper and had never been tried at 02:00.
Recalls and notifications: who decides and who must be told
A 3PL seldom runs a recall. The brand owner usually leads anything that reaches consumers, while the warehouse tracks down the stock, separates it, holds it and then sends it back or destroys it as the owner directs. (A withdrawal recovers goods from trade customers before shoppers buy them; a recall recovers them from consumers.) If your own people spot a safety or quality defect in an owner’s goods, you notify that owner formally and file the action you agreed (clause 3.7.2). The procedure has to work at any hour of any day and is tested at least once a year (clauses 3.7.3, 3.7.4). The management decisions are where recovered stock is held, who signs for its destruction and who speaks to customers.
Some events also have to reach your certification body, and the deadlines are short:
| Event | Who must be told | When | Reference |
|---|---|---|---|
| A product recall affecting the site | The CB | Immediately | Part III 6.1 |
| Legal proceedings on safety or legality, serious site damage, a new owner, or a significant change in operations or scope | The CB | Immediately | Part III 6.1 |
| A significant product safety incident, or regulatory enforcement such as a notice | The CB that issued the certificate | Within 3 working days | 3.8.6 |
| A wholesaler’s own decision to recall | The CB | Within 3 working days of the decision | 10.1.2.3 |
| An incident that may have affected an owner’s goods | The product owner | Promptly and formally, with the agreed action recorded | 3.8.3; 3.7.2 |
| Certification suspended or withdrawn | Your customers | Immediately, with the circumstances and the plan to regain it | Part III 6.5 |
Once told, the CB can leave certification as it is, suspend it while it looks into things, ask what you are doing about it, come and see, withdraw the certificate or reissue it to a new owner, and any change in status shows on the Directory. The notification decision belongs with a named senior manager. Module 4 of our BRCGS Storage and Distribution Issue 4 for Management course works through temperature failures, continuity and recall as management decisions.
Five things to do this week
- Put your due date, the 28-day window before it and the next re-audit date in the board calendar, and confirm whether you are on a 12-month or a 6-month cycle.
- Read your last audit report as a customer would. Would its root causes and preventive actions convince you?
- Write down who may put stock on hold at night, who rings each product owner, and who decides whether to notify the CB.
- Ask your CB which year your unannounced audit falls in, and list genuine non-audit days and planned shutdowns.
- Test your backup power with the alarm call-out: run the generator under load and ring every call-out number with the mains off.
Related course: BRCGS Storage and Distribution Issue 4 for Management
A four-hour, self-paced course for the managers who carry the audit result. Module 5 covers the audit programmes, the mandatory unannounced audit, non-conformities and grading, the 28-day clock, due dates and what follows the audit; Module 4 covers temperature failures, the generator decision, incidents and recall. Case studies, knowledge checks and scenario games run throughout, with a final assessment and an ASC course certificate on passing.
See the BRCGS Storage and Distribution Issue 4 for Management course (4 hours, online, R1,195). Training a team? Choose “EFT or company invoice” at checkout, or WhatsApp us for a quote.
Common questions
How is a BRCGS Storage and Distribution audit grade worked out?
The grade comes from the number and level of non-conformities. With no majors, up to 5 minors is AA, 6 to 10 A, 11 to 16 B, 17 to 24 C and 25 to 30 D. Each major drops the result about one band, and a critical or a third major means no certificate. Grades from unannounced audits carry a plus sign.
How often is a site graded C or D audited?
Every 6 months, against 12 months for AA, A and B, under a BRCGS position statement applied since February 2023. A D, and a C with two majors, also needs a revisit within 28 calendar days. Check the current position statements on brcgs.com.
What happens if we miss the 28-day deadline for corrective action evidence?
No certificate is issued and a new full audit is needed. For every finding the certification body needs the correction, the root cause and the preventive action plan, with proof, inside 28 calendar days of the audit, or 90 days after a site’s first certification audit.
Does BRCGS Storage and Distribution require a generator?
No. Clause 3.8.4 asks for contingency plans covering the loss of key services such as energy, and clause 7.3.2 requires temperature alarms or regular recorded checks. A generator, a fuel contract and backup power for alarms are common ways to meet them, and each must then be maintained and tested.
Who decides whether stock is safe after a cold store failure?
The site and the product owner together. Under clause 7.3.5 the owner shares the judgement on safety and quality before any affected stock is released, and the decision is recorded. Until then the stock stays on hold.
Next step. If the managers who face the auditor need a shared grasp of grades, timescales and the decisions behind them, the BRCGS Storage and Distribution Issue 4 for Management course covers them with South African logistics case studies and a grading game. Supervisors can start with the Introduction to BRCGS Storage and Distribution Issue 4 course, and for the wider BRCGS family see our BRCGS certification and training hub for South Africa.
This article is general guidance for South African warehousing, cold storage, transport and distribution businesses. It is not a substitute for the Standard: a site is audited against the BRCGS Global Standard for Storage and Distribution (Issue 4) together with any position statements in force. Check brcgs.com for the current issue and position statements before you plan an audit cycle, and check which registrations, approvals and licences the law requires for your premises and vehicles. Clause numbers are given so you can navigate your own copy. ASC Food Safety Training is independent of BRCGS: this article and our courses explain the Standard in our own words, are not endorsed by BRCGS, and lead to an ASC course certificate, not a BRCGS qualification.