You can't put a single molecule on a set of kitchen scales. Mass spectrometry solves the problem of weighing something far too small to weigh directly by doing something that sounds counterproductive: it destroys the sample, converting molecules into a spray of electrically charged fragments, and infers the original molecule's identity from the precise pattern of masses those fragments produce.
Ionise, accelerate, deflect, detect
A mass spectrometer runs a sample through roughly the same sequence of stages regardless of the specific instrument design. First, the sample is ionised — knocked into electrically charged fragments, often by bombarding it with a beam of electrons, which both breaks molecules into pieces and strips or adds electrons to leave each fragment carrying a charge. Those charged fragments are then accelerated through an electric field and passed through a magnetic or electric field that deflects lighter fragments more sharply than heavier ones, sorting the whole spray by their mass-to-charge ratio. A detector at the far end records exactly where each fragment lands, producing a mass spectrum: a graph of fragment abundance against mass-to-charge ratio.
Reading the wreckage backwards to identify the whole
The resulting spectrum isn't a single number — it's a distinctive pattern, because a given molecule tends to fragment in predictable, characteristic ways, breaking most readily at its weakest bonds. A trained chemist, or increasingly a computer matching against a reference library, can read that fragmentation pattern back into the structure of the original molecule, the same way you could infer a fair amount about an intact object from a controlled, repeatable way of breaking it apart and cataloguing every piece. The heaviest peak in the spectrum, the "molecular ion," is often close to the original molecule's full mass, before any fragmentation occurred.
What we're still unsure about
The physics of ionisation, acceleration and deflection is well understood and precisely engineered, but predicting exactly how a genuinely novel, complex molecule will fragment — rather than matching it against a known reference spectrum — is still partly empirical rather than something chemists can reliably calculate from first principles alone. For very large biological molecules like intact proteins, getting a clean, interpretable spectrum without excessive, uninformative fragmentation remains a significant technical challenge, and improving that is an active area of instrument design rather than a solved problem.
This sits inside Mass Spectrometry, one of seven topics in Analytical Chemistry, one of six domains in Chemistry, one of seventeen subjects the app can quiz you on.