Approving a Backup Factory Without Losing Product Consistency

Approve a backup factory against the product and application your buyer has already accepted. Confirm the new site’s scope, review its food-safety evidence, compare the relevant raw-material and process differences, and test the proposed product under agreed conditions. Record exactly what the resulting approval permits. A matching certificate or an attractive sample is only part of that work.

The objective is a usable alternative that meets the agreed requirements when it is needed. That requires two connected reviews: whether the source is qualified for the proposed supply, and whether its product performs acceptably in the customer’s use. A positive result in one review cannot fill a gap in the other.

Define exactly what the backup must replace

Start with the interruption you are trying to cover. A second factory for a seasonal raw-material shortage may still depend on the same crop region. A site with a separate raw-material route may lack the approved retail packing line. Write down the failure the alternative should address, the delivery window, and the product volume or SKU range that matters.

Then identify the comparison baseline. Use the current approved specification, relevant customer requirements, pack information and application reference. A sample from an old development round may be useful for discussion but should not silently replace the current written requirements. If the original approval is unclear, resolve that before asking another factory to reproduce it.

For an illustrative strawberry programme, the baseline might be a defined whole-fruit ingredient used in a customer’s cooked filling. The review needs the fruit description and defect criteria, the pack, the preparation method and the filling performance the customer values. “Equivalent strawberries” leaves too much undecided.

Separate essential conditions from negotiable differences. A legally required declaration, an agreed identity claim and a customer’s maximum particle size do not have the same flexibility as carton dimensions that the warehouse can accommodate. Give each proposed difference an owner and a decision. A lower price should enter the comparison only after the team understands what product and approval work it buys.

Check the site and certification scope

Identify the proposed production establishment by legal name and address, then connect it to the activities used for your item. Record where preparation, freezing, sorting, packing and storage occur. If a subcontractor performs a relevant step, include that arrangement in the review. A group company name or a seller’s office address is an incomplete production-site reference.

Check certificates against that identified site, the relevant product and process scope, current status and the buyer’s requirements. Our food-safety certificate guidance explains the wider qualification context. Two sites carrying the same scheme name can still have different equipment, raw-material controls, process limits and experience with your product.

Check the edition as well as the scheme. During a transition, the latest published code is not necessarily the edition used for the site’s audit. SQFI’s edition-selection guidance makes that distinction and directs sites to confirm the applicable audit cycle. Keep the certificate and audit basis together rather than combining requirements from different editions without explanation.

Request evidence proportionate to the intended supply. Useful material can include the relevant audit scope and findings, corrective-action closure, product experience and the facility’s response to previous problems. SQFI’s approved-supplier guidance connects assessment with risk, specification and performance. Treat the review as a decision about this source and item, not a general endorsement of everything the factory sells.

Compare the raw material and process route

Ask both sources to describe the route in terms that can be compared. A shared flow-chart heading such as “sorting” does not establish the same sorting criteria or control. A shared word such as “blanching” does not establish the same treatment. Identify the differences that could affect the agreed food-safety, physical, sensory or packing requirements before selecting tests.

For whole strawberries, relevant questions can concern variety where specified, maturity, size distribution, damaged fruit, foreign-material controls, preparation, freezing and handling before packing. The photograph supplies a visible whole-fruit reference. The actual comparison requires identified samples and records from the proposed sources, with their production and storage histories.

The table is a review aid, not a fixed test schedule. Use the differences to decide what evidence is needed. If one source measures defects after thawing and another reports a frozen inspection, resolve the method before comparing percentages. A pair of numbers can appear close while describing different conditions or denominators.

Review food-safety evidence for the proposed source

Have the responsible food-safety team assess the proposed source against the food, process and intended use. Identify who controls the relevant hazards and what evidence supports those controls. Do not use an acceptable colour, texture or filling trial to answer a microbiological or chemical-control question. Equally, a satisfactory laboratory result does not demonstrate all the functions the customer needs.

As one jurisdiction-specific example, the U.S. preventive-controls framework includes a risk-based supply-chain programme for raw materials and ingredients with hazards requiring a supply-chain-applied control. The applicable requirements depend on the operation and food. They should not be copied as a universal checklist for every international purchase.

For relevant U.S. verification work, FDA’s third-party audit resources explain why an audit’s standards and scope need consideration against applicable requirements. Receiving facilities and importers retain their responsibilities. Ask the qualified reviewer what the proposed audit establishes and whether additional information or activities are needed for this source.

Preserve the difference between receiving documents and evaluating them. Identify the report, site, food, period, findings and the reviewer who assessed it. Record unresolved questions and their effect on the sourcing decision. A commercial team can coordinate access to factory records, but it should not turn “documents received” into “technical approval complete” without the responsible person’s conclusion.

Design a fair product comparison

Write the comparison plan before the samples arrive. State the question, the samples required, preparation conditions, measurements and the criteria for a decision. Ask how the sampling covers expected production variation. Several measurements from one small sample are not the same evidence as independently produced lots, and neither should be described more broadly than the design supports.

Record each sample’s source site, product, lot, production date where available, pack and storage history. Compare suitable material at comparable condition and age, or document why a difference cannot be avoided. If the primary-source reference has deteriorated in storage, matching it may be the wrong target. Use the approved requirements to decide whether that reference remains suitable.

In the illustrative filling comparison, keep the preparation protocol the same for A and B while assessing the source difference. Use the same recipe, equipment, input basis, handling sequence and defined assessment point. Record any departure. If the team changes the starch, fruit level and mixing method only for B, it has tested an adapted formulation rather than a direct replacement under the original conditions.

Choose sensory methods for the question being asked. The public abstract for ISO 4120:2021 describes triangle testing for perceptible difference or similarity, while noting that the method does not identify the magnitude, direction or responsible attribute of a difference and has product-suitability limits. A sensory specialist should select the design and interpretation. An informal tasting with no obvious difference is not sufficient evidence of equivalence.

Keep the raw results, including variation and departures from the plan. Report the actual measurement stage and units. If a result sits near an acceptance boundary, consider the method’s uncertainty and the agreed decision rule rather than rounding it into a pass. Do not invent a universal number of samples, lots or assessors; the required confidence and product risk should shape the design.

If the factory and buyer report different results, compare the sample identities and methods before disputing the conclusion. Check whether both tested the same lot, prepared it the same way and used the same reporting basis. Retained samples can support an agreed investigation where their condition remains suitable. Keep the original results and the explanation of any resolved difference together.

Test the ingredient in the customer application

A raw-product comparison can identify differences that matter, but it cannot reproduce every customer operation. Ask where the ingredient must perform: dosing, mixing, heating, cooling, freezing, holding or serving. Select the relevant trial stage and involve the people who use the product. A laboratory preparation may establish enough for one use while another requires a controlled customer-line trial.

The SQF Quality Code, Edition 9 provides a specific scheme example: it connects product realization with process capability, facility trials, product testing and controlled specifications. This supports asking for evidence of consistent performance under the intended conditions. It is not a claim that every source follows that scheme or that certification replaces a buyer’s application decision.

For a mushroom ingredient, the relevant questions may concern the specified cut, distribution through a filling or sauce, the texture at the agreed assessment point and the effect of the customer’s handling. These are different questions from those used for whole strawberries. Keep the trial tied to the actual recipe and process; approval for one item should not automatically extend to another product from the same site.

Separate replacement and adaptation decisions. If B needs a recipe adjustment to achieve the desired result, document the proposed adjustment and obtain the appropriate approval. Review its implications for ingredients, declarations, processing and cost before treating it as routine supply. The alternative may still be useful, but the customer needs to understand which operating conditions make it acceptable.

Where storage performance matters, agree how it will be evaluated and when a conclusion can be made. An initial trial cannot establish every point in the finished product’s commercial life. Record what has been demonstrated and what remains under evaluation. Keep experimental product under the site’s applicable controls until its permitted use and disposition are resolved.

Approve a defined scope and control the handoff

Write a decision that purchasing, production and receiving can use. Identify the production site, product, specification revision, pack, application and supporting evidence. Include any restrictions, additional checks and review conditions. Avoid a one-word “approved” message that leaves the recipient to guess whether it covers the sample, one trial lot or unrestricted repeat supply.

The illustrative decision record connects Site B, product S-1, filling use and specification R3. Those identifiers show the scope a record should connect; they are not an actual approval. In practice, the authorized people should also identify the evidence reviewed, outstanding conditions, effective date and the orders or lots to which the decision applies.

Keep technical qualification, customer acceptance and commercial authorization traceable to their responsible parties. Their sequence can vary with the programme, but unresolved items must remain visible. A buyer’s acceptance of one sensory difference does not override a food-safety requirement. A factory’s production confirmation does not automatically show that the customer’s trial has passed.

Update the operational references before supply changes. The purchase order, approved supplier or site list, specification, artwork where affected, inspection instruction and receiving record should point to the intended source. Tell the receiving team how to identify the first alternate lot and where its conditions are recorded. Retain the old and new references so a later complaint can be traced to the source actually used.

Handle urgent supply through a documented exception

Urgency changes the time available; it does not supply missing evidence. Escalate the proposed alternative with a clear statement of what is known, what remains open and what the order needs. Consider whether stock, scheduling or a narrower approved product range can cover the interruption while the remaining qualification work is completed.

Where a temporary route is permitted, define its conditions explicitly. Under applicable U.S. section 117.420 provisions, temporary sourcing from an unapproved supplier must be necessary and appropriate, with adequate verification before the material is accepted for use. The official part 117 text also separates documented approval, receiving procedures and verification records. This is a conditional regulatory example, not general permission to bypass qualification.

For the commercial exception record, state the site, item, lot or quantity, permitted use, extra checks, decision owner and end condition. Make its expiry or review trigger operationally visible. Receiving a temporary lot should not silently convert the source into a permanent approved supplier, and an unresolved requirement should not disappear when a new purchase order is raised.

SQFI’s change-management guidance includes temporary, unplanned and emergency changes in its assessment approach. Use the applicable customer, site and regulatory process to decide what may proceed. When adequate evidence or required authorization cannot be obtained in time, the shipment plan needs to change.

Keep the backup ready after approval

A qualified alternative can become unsuitable while it sits unused. Agree the events that trigger a review: a changed production site, raw-material route, process, pack, certificate status, relevant complaint or extended interruption in production. The review interval should reflect the programme; a date in a spreadsheet is useful only if someone owns the underlying checks.

Use the first commercial lots to check whether the approved arrangement is being followed. Compare their identified results with the agreed specification and trial conditions, review deviations and communicate the findings to the buyer. Keep those observations separate from the original trial evidence. They add production experience rather than retroactively changing what the trial demonstrated.

Review practical availability alongside technical readiness. Confirm the relevant crop or material position, production and packing route, quantity, lead time and delivery window when an order is proposed. A technically qualified factory with no workable capacity for the required period cannot cover that particular interruption. Likewise, a free production slot does not establish current product approval.

Maintain one accessible readiness record that links the current scope, last review, supporting evidence, restrictions and responsible contacts. When the alternative is activated, carry those conditions into the order. A well-prepared backup lets the team explain both why the source is suitable and how the customer will receive the product it agreed to buy.