When a batch changes unexpectedly, memory rarely tells you which milk, culture or processing step caused the difference. Dairy batch records make those connections visible. They help you investigate quality, account for ingredients, identify affected product and improve the next production decision.
For a small Kenyan processor, the system does not need to begin with expensive software. A clear paper form or simple spreadsheet can be useful when it records the right information at the right time. This guide provides a practical structure, examples and ways to keep records connected from receiving to dispatch.
Dairy batch records: define the purpose first
A batch record describes what actually happened, rather than only what the recipe intended. It connects materials, people, equipment, measurements and product decisions. Keep the standard method separately so that you can compare the planned process with the completed record.
The Codex dairy hygiene code emphasises controls across the production chain. Recordkeeping supports that approach by making relevant actions and observations available for review. A completed form, however, does not replace effective control or prove safety by itself.
Decide who uses the record. The operator needs a manageable form during production. The supervisor needs enough information to review deviations and release decisions. The person handling complaints needs to connect the finished pack to the batch and dispatch information.
Avoid adding fields merely because another business uses them. Each field should support a control, investigation, stock decision or relevant requirement. A short form that staff complete accurately is more useful than a complicated template filled in from memory several days later.
Create an identity that follows the product
Choose a batch identification method that is unique and understandable within your operation. For example, a date-based code plus a sequence number can distinguish two batches made on the same day. Document the meaning so that everyone uses the code consistently.
Connect the code to the finished packs and the supporting records. Ingredient receiving, processing, packaging, storage and dispatch should not sit in unrelated notebooks with no shared reference. The identity is useful only when you can follow it through those stages.
Record the product and formula version alongside the code. If a flavour or ingredient changes, the record should show which version was made. Avoid silently overwriting a recipe file and losing the instructions that applied to an earlier batch.
Keep corrections readable. Use a defined method that preserves the original entry, the correction, date and responsible person as appropriate. Do not remove an inconvenient observation to make a record look compliant. Accurate deviations provide information needed for a sound decision.
Dairy batch records: capture ingredients and receiving
For each ingredient, record the exact product, supplier, lot identity where available and quantity used. Link the purchase or receiving reference. Include the relevant checks required by your system, such as condition and storage information for the actual material.
| Field | Example of the information needed |
|---|---|
| Batch code | Your unique production reference |
| Product and version | Named food and current formula version |
| Ingredient identity | Exact milk, culture and other materials |
| Supplier and lot | Source and traceability reference |
| Quantity used | Measured amount with the correct unit |
| Receiving status | Accepted, held or rejected with a reason |
Do not let a generic word such as culture replace the actual ingredient identity. If you use several products, that omission can prevent a useful investigation. Keep supplier instructions accessible and record the version or product reference relevant to the batch.
Where an ingredient is held because of a concern, make its status clear in both stock handling and records. Another worker should not be able to use it accidentally because it appears on a shelf beside accepted materials.
Record processing as the work happens
Use fields that match your defined process and the actual equipment. Record relevant measurements, times, observations and responsible people at the required points. Do not copy a generic temperature or dose from this article; your method must specify the appropriate controls.
Include equipment identity when it helps investigation. A change in vessel, thermometer, scale or filling tool can matter when a result differs. Keep relevant calibration, accuracy or maintenance references connected to the tools used.
Record interruptions and departures from the method promptly. State what happened, when it happened and which product may be affected. A note such as “delay” is less useful than a clear description of the stage and available measurements.
Define who reviews the deviation and what authority they have. The operator should know whether to stop, hold product or contact a supervisor. A record that identifies a missed control but leaves the decision unexplained is incomplete for practical management.
Dairy batch records: connect packs, storage and dispatch
Record the packaging type, relevant lot or purchase reference, quantity filled and rejected units. Include suitable filling and closure checks. Reconcile the pack count with the product available so that unexplained losses or overfilling become visible.
Next, link storage observations and product status. Keep released, held and rejected product clearly identified. The physical stock and the record should tell the same story; a spreadsheet status is not enough if the packs remain mixed together in a refrigerator.
For dispatch, record the batch, quantity, customer or destination reference and relevant handover details. Follow applicable privacy and recordkeeping requirements while keeping enough information for traceability. Decide who can access the records and how they are protected.
Test the connection with a simple exercise. Choose a finished pack and see whether you can identify its ingredients and dispatch. Then choose an ingredient lot and see which batches used it. Missing links show where the system needs improvement before a real concern arises.
Use records to understand cost and yield
Batch information supports more than investigations. It can reveal ingredient use, filling losses, labour time and acceptable output. Connect these measurements to your cost sheet rather than estimating every new batch from the same optimistic assumptions.
Suppose an illustrative batch costs KSh 4,800 and yields 60 acceptable packs. The allocated cost is KSh 80 per pack before any omitted expenses. If records reveal that only 55 packs are acceptable, the same batch cost is about KSh 87.27 per saleable pack.
These figures are teaching examples, not Kenyan supplier quotations. The important point is the connection between recorded output and cost. If your records count every filled pack as a sale, they can hide the financial effect of damaged closures, samples or rejected product.
Review patterns over several batches. Repeated overfilling or a recurring delay may justify a targeted improvement. Use evidence to choose the change, then record whether it improves the outcome. A useful system supports decisions rather than simply collecting paperwork.
Dairy batch records: start small and train the team
Introduce a first form using one product and a normal production day. Ask staff which fields are unclear or difficult to complete during the work. Improve the form without removing information needed for controls and traceability.
Sajayis lists Three Products Training at KSh 6,500, checked on 12 November 2026. Its published outline includes production and product safety. Ask whether the current session can discuss the measurements and notes relevant to your intended process.
For individual learning, Personalised Training is also listed at KSh 6,500. Confirm the scope rather than assuming every recordkeeping or software topic is included. Bring a draft form and one specific question to make the discussion practical.
Sajayis lists 55 Business Centre, next to Star Mall as its contact point. Call or WhatsApp +254 722 984117 for directions and an agreed venue. Learners from Nairobi, Kiambu, Thika or elsewhere should confirm arrangements before travelling. Use the dairy process enquiry to share your needs.
Set a record storage routine from the beginning. Decide where completed forms are kept, how digital copies are backed up and who can change the files. Follow the retention requirements relevant to your operation. A useful record must remain readable and retrievable after the production day, rather than disappearing into a personal phone or an unlabelled folder.
Frequently asked questions
Do I need expensive software immediately?
Not necessarily. A clear paper form or simple spreadsheet can support a small operation when it is accurate, controlled and accessible. Choose the system according to your process and requirements. Software cannot repair missing measurements or unclear responsibilities by itself.
Should I record only successful batches?
No. Deviations and rejected product provide important information. Record what happened and how the decision was made. Removing unsuccessful results can hide recurring problems and make both technical investigation and cost planning less reliable.
What is the most important link to preserve?
Keep the connection between the finished batch, its ingredients, processing and dispatch. The exact fields depend on your operation. Test whether you can trace in both directions and improve any gap before you need the information during a complaint or recall.
Can a completed form prove that food is safe?
No. Records support a system of effective controls and appropriate verification. A form can be complete while the underlying method is unsuitable or the entries inaccurate. Review the actual process and seek competent advice for product-specific controls.
How often should I review the system?
Review records routinely and reassess the form when products, ingredients, equipment or workflow change. Investigate recurring omissions and unclear entries. A practical system evolves with the operation while preserving the evidence needed for traceability and sound decisions.
Prepared with AI assistance using the linked sources and Sajayis catalogue checked on 12 November 2026. This article provides general information; illustrative calculations are identified in the text.