Average is not enough
Imagine a batch where half the tablets carry 130% of the dose and half carry 70%. The average is a perfect 100% — yet every single tablet is wrong, and a patient gets either too much or too little. That is why a separate family of tests, uniformity of dosage units, checks that individual units cluster tightly around the label, not just that the mean is right.
Uniformity is judged two ways. Weight variation is the quick proxy: weigh individual units and check their spread, which works when the drug is a large, evenly distributed fraction of the unit. Content uniformity is the rigorous version: actually assay individual units for drug content, required for low-dose or potent drugs where weight alone could hide a bad mix. The pharmacopoeial method usually tests 10 units, computes an acceptance value from the mean and spread, and may extend to 30 if the first stage is borderline.
Assay: how much active is really there
The assay answers the headline question: how much active pharmaceutical ingredient (API) does the product contain, expressed as a percent of label claim? A typical limit is 95.0–105.0%, tighter for some drugs. The workhorse method is high-performance liquid chromatography (HPLC), which separates the API from everything else and measures it against a reference standard. Crucially, an assay is only as good as the method behind it — which is why analytical method validation proves the method is accurate, precise, specific and robust before any result is believed.
Tablet assay (% of label claim)
Label claim = 250 mg per tablet
20 tablets weighed, total = 6.40 g -> average tablet = 320 mg
Powder taken for assay = 320 mg (equiv. to 1 tablet)
HPLC result, drug found = 248.0 mg
% of label claim = (found / claim) x 100
= (248.0 / 250) x 100 = 99.2%
Specification: 95.0 - 105.0% of label claim
99.2% lies inside the range -> assay PASSES.Impurities: what should not be there
Strength is only half the story; purity is the other half. An impurity is any component that is not the API or a deliberate excipient — a starting material, a by-product of synthesis, or a degradation product that grows over time. In a finished-product spec these usually appear as related substances: each named impurity has a limit, unspecified impurities share a lower limit, and the total is capped. Because some degradants are far more harmful than others, limits are set from safety, not convenience.
Not every quality test is chemical. Mechanical robustness matters too: the friability test tumbles tablets and checks they do not chip or crumble in handling and shipping. Together — uniformity, assay, impurities, friability — these tests guard the individual dose. The final guide steps back to the package and the dossier that bind them into a marketable product.