Water needs a preservative
Any formula that contains water can grow bacteria, yeast and mould. Water is the medium, and the small amounts of sugar, protein, gum and plant extract that make a product pleasant are also food. A cream that is 70 percent water and has no preservative is not a natural product; it is a growth medium with a lid. The risk is not theoretical, and it does not depend on how clean the bench looked. Pseudomonas, Staphylococcus and Candida are common in contaminated cosmetics, and a jar that a client keeps in a warm bathroom gives them everything they need. A preservative is not optional in a water product.
Water activity, pH and packaging decide how fast the problem arrives. A product with high free water, a near-neutral pH and a jar that fingers enter is the hardest case. A rinse-off product with a low pH and a pump is easier, but easier is not safe. Adding salt or glycerin lowers water activity a little, and that helps, but it does not replace a preservative system. Nor does a low pH by itself stop mould. Each factor shifts the odds; only a preserved, tested formula changes them enough to sell. A product with a preservative that has never been tested is still an untested product.
There is one honest exception, and it is narrow. An anhydrous product, with no water and no water entering in use, does not support microbial growth in the same way, so it needs an antioxidant rather than a preservative. A lip balm, a body oil and a water-free balm can fall into that group. The moment water is added, or the moment a product is used with wet hands, the exception ends. A scrub with sugar and oil that is kept in the shower is not anhydrous in use, because water reaches it every day. When in doubt, treat the product as a water product.
Hygiene buys time but does not replace the system. Clean vessels, filtered water, a closed batch, a sanitised filler and a trained operator all reduce the starting load. They do not stop the client from introducing microbes on the first use. A preservative system that is correctly dosed, compatible with the formula, and proven by a challenge test is the only part of the control that keeps working after the jar is opened. A workshop that sells without one is selling an experiment to a person who did not agree to be in it. That is why the preservative and its test appear in the planning, not in the last week before launch.
- 1. Ask whether the formula contains water, or will meet water in normal use, before anything else.
- 2. If yes, plan a preservative system from the first draft, not after the batch is stable.
- 3. Note the pH, because the system must work inside its own pH window.
- 4. Note the pack, because a jar that fingers enter needs more protection than a sealed pump.
- 5. Note the use, because a rinse-off has a shorter contact time than a leave-on.
- 6. Keep the water, vessels and filling area clean, and record the cleaning.
- 7. Do not call a water product natural or preservative-free and put it in a jar for a client.
- 8. Book a challenge test before the first sale, and keep the report with the batch record.
Bench example: the same cream in two packs
A 100 g moisturiser is made to the lotion card: water 77.20, glycerin 5.00, xanthan 0.20, disodium EDTA 0.10, panthenol 1.00, cetearyl alcohol 2.00, glyceryl stearate 1.50, PEG-100 stearate 1.50, caprylic/capric triglyceride 6.00, isopropyl myristate 3.00, dimethicone 1.00, phenoxyethanol at the supplier dose 0.80, ethylhexylglycerin 0.10, tocopherol 0.10, citric acid 10 percent solution 0.50. The pH is 5.2. Filled into an airless pump, the product is protected from fingers and from air. Filled into a wide 100 g jar, the same formula is opened, touched and left open in a steamy bathroom. The preservative system is identical, but the challenge is not. If the salon sells both packs, the jar deserves the more conservative system and the earlier re-test. The formula is only half the decision; the pack is the other half. If the salon sells both, the jar carries the earlier re-test date and the clearer use instruction, because the client's fingers are part of the system. Write the pack type on the batch record beside the batch number, and remember that a pump refilled from a bulk jar is not a pump any more.
Caution: Never sell a water-based cream or lotion without a preservative system and a challenge test. A clean bench does not make a product self-preserving.
Choosing and dosing
Do not invent a preservative percent. Use the supplier data sheet: it gives the use level, the pH window, the temperature limit and the compatible partners. The dose is the use level, not a suggestion, and the pH window is the range where the material actually works. A preservative used below its dose, outside its pH, or added to a batch that is too hot is a line on a sheet that does nothing in the jar. Read the sheet for the grade you bought, because the same INCI name can be supplied in different blends and at different recommended levels.
A legal ceiling is not a target. Phenoxyethanol is commonly limited to 1.0 percent in the European Union, and benzyl alcohol as a preservative is commonly limited to 1.0 percent there as well. Those numbers are the top of the legal range. A formula that uses phenoxyethanol at 0.80 percent does so because the supplier dose, the pH, the pack and the challenge test support that level, not because the ceiling is 1.0. Starting at the ceiling leaves no room, adds irritation risk, and usually signals that nobody read the sheet. Choose the lowest dose that passes the test in the actual formula.
Organic acids work in the acid form. Sodium benzoate and potassium sorbate generally need a product pH under about 5.5, and their effect falls as the pH rises. That is why a shampoo at pH 6.0 is a harder preservation case than a lotion at pH 5.0, and why a benzoate and sorbate pair is common in a shampoo but must still be tested. A single acid is often weak against mould or against Pseudomonas. A chelator such as disodium EDTA supports the system by binding metals, and a booster such as ethylhexylglycerin at about 0.1 to 0.5 percent can improve performance, but neither is a complete system on its own.
Temperature and phase decide where the preservative goes. Some materials tolerate the hot phase, and some must be added under 40 C. A volatile preservative added to a hot batch can flash off, leaving the product underdosed while the sheet still reads correctly. Add it at the temperature the supplier allows, stir it in fully, and record the batch temperature at that moment. If the formula also contains a fragrance or an essential oil, remember that those are allergens and fragrances, not preservatives, and they do not reduce the dose of the real system. A fragrance allergen declaration is a labelling duty, not a substitute for preservation.
- 1. Read the supplier sheet for the exact grade, and write the recommended dose on the card.
- 2. Compare that dose with the legal ceiling in the market of sale, and use the lower figure.
- 3. Check the pH window of the system against the pH the formula will actually have.
- 4. Decide the phase from the temperature limit: hot phase only if the sheet allows it.
- 5. Add a booster only as support, never as a replacement for the primary preservative.
- 6. Keep a chelator such as disodium EDTA at about 0.10 percent to support the system.
- 7. Record the actual weight, the batch temperature and the pH at the addition.
- 8. Confirm the choice with a challenge test in the finished formula and pack.
Bench example: dosing a lotion without starting at the ceiling
The lotion card carries phenoxyethanol at 0.80 percent and ethylhexylglycerin at 0.10 percent. The supplier sheet for the phenoxyethanol grade states a use level of 0.5 to 1.0 percent, a pH range from about 3 to 8, and addition below 60 C. The formula sits at pH 5.2 and the pack is a pump, so the system has an easy pH and a protected pack. A 0.80 percent dose is inside the supplier range and below the 1.0 percent EU ceiling, which is a ceiling and not a target. In a 500 g batch that is 4.00 g of phenoxyethanol and 0.50 g of ethylhexylglycerin, both added at 38 C. If the challenge test at week four shows a weak result against mould, the answer is a different system or a higher tested dose, not a jump to the ceiling. Change one variable, re-test, and record both reports. If the test fails, do not raise the dose quietly; change one variable, such as the booster or the pack, and re-test. Record the reason on the card, so the next batch inherits the lesson and not the guess.
Caution: A preservative ceiling is the top of the legal range, not a dose. Never raise a dose to the ceiling to avoid doing a challenge test.
The challenge test
A challenge test, also called a preservative efficacy test, deliberately contaminates a sample with known microbes and measures how well the formula kills them or holds them down. It is run to a standard method, commonly ISO 11930, and it uses a defined set of bacteria, yeast and mould. The lab inoculates the product, samples it at set intervals, and counts what survives. Pass criteria depend on the category of product and on the method, and the report states them. This is the proof that a preservative system works, and no bench observation replaces it. Ask the lab to state the method and the pass criteria in the report, so the result can be read by someone who was not there.
The test must use the finished formula in its final pack. A system that passes in a pH 5.5 lotion may fail in a pH 3.5 serum, and one that passes in a gel may partition into the oil phase of a cream and leave the water underprotected. The pack matters too, because a jar that fingers enter is inoculated by the client every day, while a pump is not. Send the sample the salon will actually sell, with the same preservative dose and the same pH, and tell the lab how the product is used. A report for a different formula is not a report for this one.
Timing matters. A challenge test takes weeks, so it belongs in the plan from the first trial, not at the end of the launch. Stability can run in parallel. A common pattern is a four-week screen first, then the formal test on the version that will be sold, with a retained sample kept from the same batch. If the formula changes after the test, the report no longer applies, even if the change looks small. Raising the glycerin, lowering the pH or swapping an emulsifier can all move the result. Write the test date and the formula version on the report, and file it beside the batch record so the two can be matched.
A smell test is not a microbial test, and a clear product is not a clean product. Contamination can be invisible and odourless at the start. The one-week smell check that many small workshops use is a spoilage alarm at best, and it usually goes off after the product has already been sold. If a batch smells wrong, treat it as contaminated until a lab says otherwise: quarantine it, do not sell it, and find out what happened in the process. The cost of a test is small against the cost of a client with a skin infection. That is not a cost the salon can absorb quietly, and it is not a risk the client agreed to take.
- 1. Choose the formula version that will be sold, and make a batch in the final pack.
- 2. Send the sample to a laboratory that runs a recognised method, commonly ISO 11930.
- 3. State the product category, the pH, the preservative dose and how the client will use it.
- 4. Ask for the raw counts and the log reductions at each sampling point, not only a pass or fail.
- 5. Keep a retained sample from the same batch beside the test sample.
- 6. Run stability in parallel, and record both sets of results against the same batch number.
- 7. Re-test after any change to the preservative, the pH, the pack or a major ingredient.
- 8. File the report with the batch record, and do not sell before it passes.
Bench example: planning a test before the launch
A salon plans to sell a 50 g jar of the lotion and a 200 ml pump of the same lotion. The jar is the harder case. The plan is to make a 2 kg batch at pH 5.2 with phenoxyethanol at 0.80 percent, fill six jars and six pumps, and keep two of each as retained samples. Three jars and three pumps go to the lab for a challenge test to ISO 11930, and the rest go on a four-week stability shelf at room temperature and at about 40 C. The lab report comes back with the counts for each organism at each interval, and the jar series is watched more closely because fingers enter it. If the jar passes but the margin is thin, the next version uses a different system rather than a higher dose. The batch number on the report matches the batch number on the jar. Nothing is sold until both the test and the stability screen are in the file. Keep the packing slip and the lab's sample receipt with the batch, because a report is evidence only if it can be tied to the jar.
Caution: Do not judge a preservative by smell, colour or a one-week check. Only a challenge test on the finished formula supports a sale.
This is a Nandu Nai textbook. It is a private brand book, not a government licence, and it does not give the right to give medical treatment.