Overage is one of those pharmaceutical concepts that sounds simple in a textbook and gets genuinely tricky the moment you're standing at a dispensing booth with a real batch record in front of you. The idea is straightforward enough: add a little extra active ingredient to compensate for expected losses during manufacturing, so the finished product still meets its labeled potency at release. The hard part is deciding how much extra is justified — because too little overage risks a sub-potent batch, and too much can push a batch out of specification on the high side, or simply waste an expensive raw material for no real benefit.
There's no single universal number here, and any article that gives you one flat percentage across all products is oversimplifying. What follows is a practical breakdown of how overage tends to vary by dosage form and by the reason it's being added — not as a substitute for your own validated formula and regulatory filing, but as a way to sanity-check whether a figure you're looking at is in a reasonable range.
The two different reasons overage exists
It's worth separating these clearly, because they're often lumped together and they justify very different amounts:
Process-loss overage compensates for API that's genuinely lost during manufacturing — dust generated during blending, granulation, or compression that never makes it into a finished unit. This kind of overage is usually small, often in the 1-3% range for a well-controlled solid dosage process, because a properly validated process shouldn't be losing much API to begin with. If your process-loss overage is regularly running higher than that, it's often a signal to investigate the process itself rather than just compensating for it with more raw material.
Stability/degradation overage compensates for API that degrades over the product's shelf life, so that the tablet still meets its labeled potency at the end of shelf life, not just at release. This is a completely different calculation, driven by real-time and accelerated stability data specific to that API and formulation, and it can range from negligible (a very stable molecule) to substantial (an API known to degrade meaningfully over 24 months). This kind of overage should be backed by actual stability study data, not a rule of thumb.
Conflating these two is one of the more common sources of confusion — a formulator asking "what overage should I use" without specifying which of these they mean is really asking two different questions.
Rough ranges by dosage form (as a sanity check, not a rule)
Tablets and capsules (solid oral dosage): Process-loss overage is typically low, often 1-3%, since compression and encapsulation are relatively low-loss processes once validated. Stability overage, where applicable, is layered on top and depends entirely on the specific API's degradation profile.
Oral liquids and suspensions: Process overage tends to run a little higher than solids, since liquid manufacturing often involves more transfer steps, line loss, and filling variability. It's also common to see a small fixed-volume overage per container to guarantee the full labeled dose can actually be withdrawn, separate from the potency overage itself.
Parenterals (injectables): Overage practices here are typically the most conservative and the most heavily justified by data, since the margin for error is much smaller and regulatory scrutiny is higher. Both a fill-volume overage (to ensure the full labeled dose is withdrawable, accounting for container and needle holdup) and a potency overage may apply, and both are usually product-specific rather than following a general rule.
Topicals and semisolids: Process loss can vary more widely depending on the manufacturing equipment (mixing, homogenizing, filling), and overage decisions here are often more empirical — based on that specific product's validation batches — than derived from a general formula.
Two things overage is *not* a substitute for
It's not a fix for a process that's genuinely out of control. If your granulation reconciliation keeps showing 6-8% loss, the answer isn't to keep raising the overage to compensate — it's to figure out why the process is losing that much material and fix it. Overage should cover normal, expected variation, not mask a process problem.
It's not something to apply twice by accident. A mistake worth watching for: applying overage once in the target batch quantity, and then applying it again separately in the API requirement calculation, effectively double-counting it. The overage should be built into the theoretical requirement calculation once — before the assay correction is applied, not folded in twice.
Where the assay correction fits in
Once overage is accounted for, there's a second, unrelated correction that also affects how much API you actually weigh out: assay. Raw API rarely arrives at exactly 100% purity, and the certificate of analysis for that specific lot states its actual assay value. The gross weight you dispense needs to be corrected for that purity:
Theoretical (net) API required = Label claim × Batch size × (1 + Overage %)
Gross API to weigh = Theoretical API ÷ (Assay % ÷ 100)It's worth double-checking the assay figure against the current certificate of analysis for the lot actually being dispensed, since assay varies lot to lot, and using an outdated figure changes how much API ends up in the batch.
If you're working through this calculation regularly, the API Requirement calculator on PharmainformCalc applies both corrections — overage and assay — in the correct order automatically, so you can go straight from label claim to gross weigh-out without re-deriving the formula each time.