What happened to division 63 in Rev. 5?

Division 63 in Rev. 5 must be checked in the UNSD structure for ISIC Rev. 5, UN Statistics Division, because the split of old Section J reorganised the information service categories that Rev. 4 grouped nearby. This guide deliberately does not assert a Rev. 5 home for division 63 from memory: the reorganisation touched the surrounding categories, and a guessed mapping would risk a wrong code statement. The correct step is to look up division 63 in the Rev. 4 structure, follow its correspondence rows to Rev. 5, and read the Rev. 5 explanatory notes for the target categories.

Why does this guide refuse to guess a Rev. 5 home for division 63?

Refusing to guess is not a gap in this guide. It is the point of it. The split of old Section J reorganised the information service categories that Rev. 4 grouped nearby, so the surroundings of division 63 changed even before any question of its own new home arises. A mapping asserted from memory in those circumstances would look helpful and could be precisely wrong, sending a reader to a category whose explanatory note does not match the activity they described, and that wrong turn would then travel into whatever table or register entry the reader produced. The guide therefore treats division 63 as a lookup task, not a memory task. Its job is to show the reader the reliable path to an answer, and to explain why the slower path is the only one that can be trusted when a number will be published or quoted.

What exactly should a reader look up, in what order?

The lookup runs in a fixed order, and the order matters because each step narrows the next one. First, find division 63 in the Rev. 4 structure and read what it covered in its own revision, in words, rather than relying on a remembered label. Second, follow its rows in the Rev. 4 to Rev. 5 correspondence table, and note every Rev. 5 category those rows point to, including cases where one old row opens onto several new categories. Third, open the Rev. 5 explanatory notes for each target category and read the includes and excludes lists, comparing them with the activity in words from the first step. Fourth, record the result with its version label, ISIC Rev. 5, UN Statistics Division, and its level, so a later reader can retrace the same path. A reader who follows those four steps in order will reach an answer grounded in the official structure. A reader who jumps to the third step without the first two may reach an answer that sounds plausible and cannot be checked.

How do nearby reorganised categories make a guessed mapping risky?

Nearby categories are the reason a guess is dangerous here, rather than merely imprecise. When a whole block is reorganised, titles that once sat beside each other can end up in different sections, and a title that survives the revision can cover a narrower or broader scope than it did before. A reader working from memory will tend to carry the old neighbourhood along with the old number, assuming that what sat nearby in Rev. 4 still sits nearby in Rev. 5, and that assumption is exactly what a reorganisation breaks. The same risk appears inside a single category: an organisation may combine several activities, an activity may change character over time, and a new method of production may blur an old boundary, so even a correctly remembered title can mislead if its new note is not read. Checking the correspondence rows and the target notes, rather than trusting the old map, is the only correction for that risk.

What record should a reader keep after checking division 63?

Keep a short note that a checker can follow without asking the original reader what they meant. The note should name the Rev. 4 starting point in words, list the correspondence rows followed, name the Rev. 5 category or categories reached, state the level at which the result is reported, and carry the version label beside it. It should also flag any break in series if figures from different revisions are compared, and note the reference period, because national adoption dates can fall inside a series. Where the detail available does not justify a fine category, the note should say that the result was reported at a broader level, and why. That record turns a private lookup into a checkable result, which is the standard a published or quoted number should meet.

As with every guide in this cluster, the scope here is statistical classification only: no reader’s code, principal activity, or registration is determined by this guide, and a code for a specific record should be confirmed with the national statistical institute or business register that holds the record.