Fill weight is the least glamorous number in a capsule specification and the one that causes the most arguments. It decides your unit cost, because you are paying for every milligram that goes into the shell. It decides complaint rate, because under-filled capsules look cheap and over-filled capsules leak. And it decides whether your declared net content on pack survives a check.
This article explains how fill weight is controlled, what tolerances are realistic rather than aspirational, and what to write into a specification so that a disagreement later has something to refer back to.
Fill weight is the number that decides cost and complaint rate
On a capsule run, the fill is usually the most expensive component per unit. A capsule filled at 300 mg and one filled at 330 mg look identical to a consumer and differ by ten percent of your ingredient cost. Multiply that across a production run and the number becomes significant.
At the same time, inconsistent fill is visible. Consumers open capsules. A capsule that is noticeably emptier than the last one in the same pack generates a review, a message to customer service, or a return. Dose consistency is a perceived-quality issue as much as a cost issue.
What "fill weight" actually refers to
Three different numbers get called fill weight, and confusing them is a common source of disagreement:
- Target fill weight - the nominal dose you designed, for example 300 mg.
- Average fill weight - the measured mean across a sample from the batch.
- Individual capsule weight - what a single capsule contains, which varies around the average.
A specification should state all three: a target, an acceptable range for the average, and a limit on individual variation. A specification that only says "300 mg" is unenforceable, because no real process hits one number.
Why the pump, not the operator, sets accuracy
In rotary die encapsulation, dosing is done by a pump that injects a volume into the die at the moment the ribbons meet. Accuracy is a function of pump condition, fill viscosity, fill temperature, and the timing of the injection relative to the die.
This means accuracy is maintained, not achieved. Pumps wear, viscosity drifts as the batch warms or cools, and timing can slip. The controls that keep accuracy are calibration at start-up, periodic in-process weighing, and scheduled replacement of wear parts. A factory that only checks weight at the end of a run is reporting history, not controlling it.
Viscosity is the hidden variable
Viscosity is the variable most often overlooked in a fill specification, and it is the one that quietly ruins dose accuracy. A pump delivers volume, but if the fill's viscosity changes during a run, the delivered mass for that volume changes too.
Viscosity moves with temperature, and a fill held in a warm room behaves differently from the same fill in a cold one. It also changes with shear and, for some formulations, with time. The practical control is to specify a viscosity range at a stated temperature and to keep the fill within that range during production.
If your formula contains suspended material - powders, exfoliants, encapsulated actives - settling is a second risk. A fill that settles in the holding tank doses lighter at the end of the run than at the start unless it is kept moving.
Realistic tolerances: what to specify
| Parameter | Typical expectation | Why it matters |
|---|---|---|
| Average fill weight | Within a few percent of target | Controls cost per unit |
| Individual variation | Tighter for small fills, looser for large | Controls perceived consistency |
| Shell weight variation | Controlled by ribbon thickness | Changes total weight and appearance |
| Fill viscosity at fill temperature | Stated range, checked per batch | The main hidden cause of drift |
| Sampling frequency | Regular intervals plus start-up check | Detects drift while it can be corrected |
The exact percentage that is achievable depends on fill viscosity, capsule size and equipment condition. Rather than fixing a number from a template, ask your manufacturer what their process capability is for your specific fill, then write that into the specification. A tolerance the process cannot hold is worse than no tolerance, because it guarantees disputes.
How in-process weighing works
The standard method is to sample capsules at defined intervals, weigh them, and subtract the expected shell weight - or to weigh the shell and fill separately for a subset. On better lines, automatic check weighing removes under- and over-weight units.
Two practical points. First, weighing a whole capsule and subtracting an assumed shell weight is only as good as the shell weight assumption; if ribbon thickness drifts, the calculation drifts. Second, sampling at the start and end only will miss a mid-run excursion. Ask for the sampling plan, not just the result.
Over-fill versus under-fill: which costs you more
Under-fill is the obvious commercial risk: a lighter capsule reads as a cheaper product and, at the extreme, may push the batch outside its declared net content.
Over-fill is the quieter risk and often the more expensive one. Excess fill makes sealing harder, increases the chance of product trapped in the seam, and raises the leakage rate in transit. It also costs you the ingredient outright. In practice, the leakage and rework cost of systematic over-fill usually exceeds the perceived-quality cost of slight under-fill.
Fill weight and declared net content on pack
Net content declarations are regulated, and the rules differ by market. If you declare a count and a net weight, the actual contents need to be consistent with the declaration across the batch, not just on average.
This is a labelling question as much as a production one, and it interacts with how you count capsules per pack. Requirements by market are summarised in our labelling guide. Raise it before artwork is finalised, because changing a declared weight after printing is expensive.
Fill weight and claims: when dose matters to efficacy copy
If your marketing refers to a dose - a percentage of an active, or a stated amount per capsule - the delivered dose has to support it throughout shelf life, not only at release. Actives can degrade, and a claim that holds at month zero may not hold at month eighteen.
This is where fill weight and stability intersect. Specify the dose at release with a small margin above your claim, and confirm through stability testing that it stays above the claim for the intended shelf life. Our stability testing guide covers the commissioning.
What to put in your specification
- Target fill weight, and the acceptable range for the batch average.
- Individual capsule limits, stated as a range rather than a single number.
- Fill viscosity range, with the temperature at which it is measured.
- Sampling plan: how many capsules, how often, and what action triggers a line stop.
- Action on out-of-specification results: rework, rejection, or concession with your written approval.
- Reference sample: a retained batch sample that defines what "acceptable" looked like at approval.
The last item is the one buyers forget. Without a retained reference sample approved by you, every later discussion about colour, weight or feel is a matter of opinion. Approve a golden sample before the commercial run; the process is described in our sampling article.
Where our capsule capability fits
We run capsule serums with defined fill weight specifications and in-process weighing, across formats including hyaluronic acid softgels, vitamin E softgels and cleansing oil capsules. If you are specifying a new product, we will tell you what tolerance our line holds for your fill rather than quoting a generic number.
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.
Send us your target dose, your market and your pack count, and we will tell you what is realistically achievable. Start at our inquiry page, email capsulebiotech@163.com, or reach us on WhatsApp at +86 158-7655-5319 (Capsule Series), +86 158-7655-5319 (Cosmetics Series).
This article is general manufacturing information, not regulatory or legal advice. Net content and labelling rules vary by market.