What is a COA and how do I get one for my product (also covers product specifications)
One of the questions I get asked a lot by smaller brands who begin exporting their products is ‘I’ve been asked to provide a C of A with my products and I don’t have one or know what one is, can you make me one and how much will that cost?’
This is how I answer.
So a COA or C of A is a ‘Certificate Of Analysis’.
Basically it is a piece of paper that gives actual test results for the batch of product that you are exporting. The test results are usually reported against the typical specification.
A Specification for a product is a piece of paper that gives guidelines of the physical and maybe chemical parameters of a product. The figures on a specification are generally ranges whereas a C of A gives absolute, batch specific figures or a figure that confirms the batch has been tested to and meets the requirements of the spec.
Here is an example of a C of A for Hyaluronic Acid. Finished products and cosmetic ingredients all go through the same procedure for creating a spec/ COA but the data given might differ between the two. I’ll explain that in a moment.
The point of a specification (which on this C of A would be represented by the ‘standard’ data column) is to give the ingredient (or finished product) maker quality guidelines. When a new batch of product is made it must be tested to find out if that batch is acceptable (within spec). Specification guidelines usually give some leeway of results to enable a bit of room for batch-to-batch variations. How much leeway is given will totally depend on the product and how critical each factor is to the quality of the finished goods. Medicinal products usually have a tighter spec than cosmetics due to a greater need for precise batch-to-batch consistency. Organic cosmetic products may have a much wider specification for things like colour and sometimes odour than a regular cosmetic due to the fact that the oils and waxes that go into organic cosmetics are more likely to vary in colour over time and depending on where they are sourced from.
So how do you get one?
You can’t have a COA without a spec.
The first thing is to get a spec.
In an ideal world and in the real world of cosmetic manufacturing you can’t create a spec without a batch history and stability data.
When you first make a product that product will have a ‘temporary spec’ based on the parameters of the first batch you have scaled up. You must make a couple more batches to get a good idea of the typical range for your product. This is important as if you have only one batch that you have made in your lab and you use that to make your spec there is no guarantee that a manufacturer who has scaled up your product can achieve what you have.
In the world of kitchens, markets and small-batch manufacturing this procedure may not be as rigorously followed. This is a risk, as you won’t know as much as you probably ought to know about your product before committing to a spec but sometimes it has to do. Basically the consequences of creating a spec and subsequent C of A’s on limited scale-up data is that the batch-to-batch variability may be quite high and you might find yourself either constantly fiddling with the spec (which is not ideal) or rejecting product. Then again, it may never be a problem.
So what should go onto a finished product spec?
The minimum data that usually goes onto a cosmetic spec is a description of the appearance, odour, pH (if the formula contains water), viscosity, specific gravity.
The viscosity is particularly important for most products as it gives you a numerical range for how thick or thin the product is and this will directly impact how the product performs in use. Will it come out of the final packaging, will it leak, will the customer be able to dose the product correctly, has the product been manufactured correctly?
Specific gravity is another key measure as it tells you how dense the product is in relation to water – is it lighter or heavier? This is a very useful measure for your stability also as changes in the SG over time means that the product is either losing or gaining moisture.
If you have no laboratory equipment to test SG and viscosity at home you can get these measured at a third-party lab. If you are doing that I would suggest you try to send three batches and use the data from that to set a specification for yourself. One batch is not illegal but there is always the danger that the batch you sent was not representative of what normally happens.
If you want to create a very, very basic spec then you can get away with just listing pH, appearance and odour. There is no law against doing that but the downside in not specifying viscosity for example is that you have no record of what it should be and customers may complain or feel dissatisfied with the batch that you have sent them.
Once you have created your specification you then use that to measure each batch against in order to create a batch specific C of A.
So a C of A becomes part of your quality control program and as it requires a batch number you need to have a batch number system in place. This can be as simple or complex as you like but it needs to be there so that you can track back each C of A to the manufacturing date, the batches of the materials you used to make the product and who made it (ideally) and where it was made.
What about micro?
So far the above has only gone into the basic physical test results usually present on a spec. As you can see from the spec for Hyaluronic Acid micro data is also useful. The way micro testing works in the cosmetic industry is that we have a separate micro spec for a cosmetic. Usually this is achieved in the same way as a regular spec, three batches or more sent off for micro counts then a limit is set and other batches are tested against that.
All cosmetics have to be clean and there are regulations in place to outline what ‘clean’ means in terms of Total Viable Count (TVC) for each category of cosmetic. In addition cosmetic products should be free from pathogenic organisms so that is tested as well. The last step for a cosmetic formula is a PET (preservative efficacy test). This is generally done once in a products life time but we do recommend running a PET at the beginning of the products shelf life and at the end, especially for cosmetics with a 20-30 month shelf life.
Batch specific results usually retain their name of ‘micro count results’ rather than ‘C of A’ but essentially they are the same thing. Micro counts are done on each batch in most cosmetic manufacturing companies.
Anything else to measure?
If you have a product whose efficacy results depend on the presence of a particular active and that active might oxidise or reduce in quality over time then that needs to be tested. A couple of examples are: Vitamin C serums or creams, Retinol (or vitamin A) products, Peptide creams, Sunscreens.
We might take our batch to an analytical lab and ask them to perform an ‘assay’ on it to tell us how much of the ‘active’ remains in the product after x amount of months.
This is not mandatory testing here in Australia for anything other than sunscreens but I absolutely recommend it as degradation of a key active WILL impact on your product’s efficacy – your product may no longer work. Also in some cases degradation of the active can lead to the formation of irritating by-products and that needs to be mitigated.
Heavy metal analysis is more relevant to individual ingredients than it is for a finished product but in some cases it might good to get the heavy metal analysis done – remember the lead in lipstick uproar?
So how do you get this started?
Realize Beauty can organise this data for you and complete specifications for finished products or ingredients.
In terms of how much this costs that will depend on what data you want on your spec. If you need assay results we will have to get a quote for that, if you need a micro spec we will have to factor in the costs of running that too. As a guide a very basic spec might cost a hundred dollars where as complex spec may be closer to a thousand. Do also keep in mind that a spec is only as good as the product that we are given to test so if you can only spare us a tiny amount from one batch then we have very little to go on. The best specs are produced when we can have access to at least three batches of product, we will need 500g of each batch as a rough idea and that should cover SG, viscosity, micro and assay of active.
If you have any further questions drop us an email at firstname.lastname@example.org