What do includes and excludes lists do?

Includes and excludes lists in ISIC Rev. 5, UN Statistics Division, work as border controls for a category: the includes list names activities that belong inside, and the excludes list names nearby activities that belong elsewhere, with a pointer to their proper place. Together they turn a general definition into a practical sorting rule for borderline cases. When two categories both look plausible, the excludes list is often the deciding evidence.

Why does a definition alone need lists beside it?

A general definition states a principle, but principles leave borderline cases open: an activity can resemble the principle closely while belonging, by convention or by a finer distinction, in a neighbouring category. The includes list closes one side of that gap by naming activities that belong inside even when a reader might hesitate over them, and the excludes list closes the other side by naming the tempting neighbours that belong elsewhere. Without the lists, two careful readers could apply the same definition and reach different placements, and neither could show the other why. With the lists, the placement becomes checkable: the reader can point to the line that includes the activity or to the line that excludes it and names its proper place. That checkability is the real work the lists do. They do not add decoration to a definition; they make the definition usable by two readers independently reaching the same result.

How do you read an excludes pointer properly?

Follow it. An excludes entry that merely says an activity belongs elsewhere, without a pointer, leaves the reader to search; an entry with a pointer gives the destination, and the destination note should be read as well, because a border is confirmed from both sides. Reading only the first note can mislead in a subtle way: the activity you are checking may be excluded from category A and pointed to category B, while category B itself excludes a variant of it and points to category C. The chain ends only when a category includes the activity and does not exclude it onward. Short chains are common, and following them takes a minute, but stopping halfway produces a placement that the classification itself contradicts. Note also that an excludes list is not a list of unrelated activities: every entry on it is there because someone could plausibly have placed it here, so each entry is a warning about a specific, tempting mistake.

What happens when an activity appears on no list at all?

Absence from both lists is not a verdict. The lists name the doubtful and tempting cases; they do not attempt to enumerate every activity a category covers, which would be endless. An activity that appears on neither list must be tested against the general definition itself, and against the notes of neighbouring categories, to see whether it falls inside the definition or is drawn elsewhere by a neighbour’s excludes list. This is where the hierarchy helps: a section or division note may state a principle that places the activity even when no detailed list names it. If the definition and the neighbouring notes still leave the case open, report at the broader level that is clearly supported and record the question, rather than treating silence in the lists as permission to place the activity wherever it looks most convenient. Silence is information about the lists, not about the activity.

A borderline activity, sorted by the lists alone

Imagine a reader with an activity that two categories both seem to claim. She writes the activity in plain words, opens the first candidate, and reads its definition: it fits, broadly. She reads its includes list: the activity is not named. She reads its excludes list: there it is, with a pointer to the second candidate. She opens the second candidate, reads its definition, and finds a closer fit. Its includes list names a near neighbour of the activity, and its excludes list does not send the activity onward. The case is settled, not by the titles, which both looked plausible, and not by the general definitions alone, which both fit broadly, but by the excludes line in the first note and the includes line in the second. That is the everyday work of these lists: they decide the cases that definitions leave open and titles cannot decide at all.

Keep that sequence in mind whenever two categories compete for one activity. Definitions frame the question, and the lists answer it.

This guide concerns statistical classification only, and it does not resolve any reader’s own case. It assigns no code, no principal activity, and no registration to any reader, and a code for a specific record should be confirmed with the national statistical institute or business register that holds the record.