Sampling is where a capsule project either de-risks or quietly stores up trouble. Everything else can be recovered later - tooling can be recut, artwork reprinted, counts changed. A wrong approval cannot, because by the time it shows up there are several thousand units in a warehouse.
There is one structural trap specific to this format, and it catches experienced brands: the first sample you are shown is not made on the machine that will make your product.
Sampling is where the project de-risks or stores up trouble
The purpose of sampling is not to get a nice-feeling liquid. It is to answer specific questions in a defined order, and to do it while changes are still free.
Once production is scheduled, every change has a cost attached and a delay behind it. The whole value of the sampling phase is that it is the last point at which changing your mind is cheap.
Lab sample and production sample are not the same thing
A laboratory sample is usually made in small quantities with different equipment from the production line. It has a different thermal history, different shear, different drying conditions if it is encapsulated at all. For many products that is fine. For some it is not.
Where it matters: anything whose sensory character depends on structure. An emulsion fill made in a beaker does not have the same droplet size as one made on a homogeniser, which means it will not feel the same. A shell dried in a lab tray does not dry like one tumbled in a production dryer, which means it may not behave the same when the customer opens it.
So: develop on lab samples, but do not sign off on one. Ask for a pilot or first-off run made on production equipment before you approve, particularly if texture or shell behaviour is doing any work in the product. Our sampling process article covers how that stage is structured.
Define the target before the first sample lands
Three things, written down, ideally before you ask for anything:
- Sensory target. Spread, absorption, after-feel at five minutes, behaviour under makeup or on dry hair. In plain language if necessary - see our texture article for how these are specified.
- Non-negotiables. The ingredient exclusions, the claim to support, the markets, the regulatory constraints.
- Who decides. One person, ideally. Component decisions by committee produce either rounds or compromises nobody wanted.
Without those three, the process becomes taste-driven, and taste-driven sampling does not converge.
How many variants, and how many rounds
Two or three variants per round. One is not a comparison - approving the only option is a coin toss. More than three and nobody can articulate the differences, which is worse than not looking.
Two rounds is normal, three is fine, five means something upstream was not decided. Each round should change one variable at a time, or you will not know which change produced the effect you liked.
Ask explicitly what was changed between rounds. A cover note listing the deltas costs the manufacturer five minutes and tells you more than the samples do.
What to assess at the sample stage
- Appearance of both fill and shell, including through the shell as the customer will see it.
- Odour at opening, then again after a minute.
- Spread and absorption on the target skin or hair type - not on the back of a hand in a meeting room if you can avoid it.
- After-feel at a stated interval, recorded.
- How the capsule opens - push-through force, whether it tears cleanly, whether the customer can get the whole dose out.
- Fit in the intended pack, because this is the last chance to catch a mismatch before tooling.
- Dose - whether the amount delivered is what you intend; see our fill weight article.
What sampling cannot tell you
Two things cannot be assessed from a sample, and it saves time to accept that:
Stability. Nothing you can do to a fresh sample predicts what it looks like in eight months. That requires a study on the finished packed product - see our shelf life article.
Microbial behaviour. If the fill contains water, preservation must be challenged in the lab, not judged from a sample. See our preservative article.
Launching on sample impressions plus a plan to watch those two later is normal and fine. Assuming a fresh sample answers them is not.
Sampling stage by stage
| Stage | Made on | Use it for | Do not sign off on |
|---|---|---|---|
| Laboratory bench sample | Small-scale equipment | Direction, feasibility, rough sensory | Final texture; shell behaviour |
| Pilot or first-off run | Production equipment | True texture, true shell, packaging fit | Anything requiring time |
| Golden sample | Production, released specification | The reference everything else is compared to | Nothing - this is the sign-off |
| Retained batch samples | Production, kept in storage | Future comparison and complaint investigation | N/A |
The golden sample, and keeping it
A golden sample is the approved reference: the physical standard against which future production is judged. It needs to be sealed, labelled, stored sensibly, and duplicated if possible - one for you, one at the factory.
Storage conditions matter, because a golden sample that has degraded is worse than none: it makes good batches look wrong. Keep it in the conditions the product is meant to experience, note the date, and replace it periodically.
Ask for retained samples from every production batch too. When a question arises months later, having a stored unit from the same run is worth more than any amount of correspondence.
Six mistakes that cost the most
- Approving the first sample that is not actively unpleasant. That is how brands end up with a product they cannot praise and cannot criticise.
- Changing several variables at once, so nobody knows what worked.
- Evaluating on the wrong skin or hair type.
- Signing off without seeing it in the intended packaging.
- Skipping the pilot run because the lab sample was good enough.
- Not keeping a golden sample.
The fifth is the one that produces expensive surprises, and the sixth is the one you regret later rather than now.
Testing beyond sensory
Worth commissioning at the sampling stage even though it costs money early:
- Dose uniformity across capsules - especially if the fill contains any suspended material.
- Seal integrity on pilot capsules, before production volumes hide the problem. See our seam integrity article.
- Packaging compatibility on the actual pack, as covered in our compatibility article.
- Colour measurement where any tendency to discolour exists - see our discolouration article.
Each of these is cheap before launch and expensive after it.
Cost reality
Sampling rounds cost money and time, and there is a temptation to skip one to save both. The comparison worth making is not against the cost of the round - it is against the cost of a production run nobody wants.
An extra round costs weeks. Being wrong costs the run, the tooling adjustment, the delay and the credibility. Our cost breakdown article shows where those lines sit in a quotation.
How to brief it
Say how many variants you want and what target each one is testing. Say which variable changes per round. Say who approves. Ask for a written note of what changed each time. And say up front that you expect a pilot run on production equipment before final approval - it changes how the project is scheduled.
Where our sampling fits
We run capsule sampling as lab direction followed by a pilot run on production equipment, with a written note of what changed each round. Our custom softgel face serum capsules development path includes both stages, and we supply retained samples from every production batch.
We work with brands on private label (your brand on a proven base), OEM/ODM contract manufacturing, and full custom formula development - whichever matches where your brand is today.
Tell us what you want the product to feel like and who decides, and we will structure the rounds around that rather than around our convenience. Start at our inquiry page, email capsulebiotech@163.com, or message the right team on WhatsApp: Capsule Series (capsules only) +86 158-7655-5319, Cosmetics Series (serum, hair care, body care, fragrance) +86 158-7655-5319.
General manufacturing information, not regulatory or legal advice. Claim limits vary by market.