Why a launch goes wrong in a spreadsheet
Tagging one link is easy, and every UTM tool does it. A launch is not one link. Five destinations across six channels is thirty links, and the way that gets done today is a spreadsheet with a CONCATENATE formula, copied down, edited by two people, over three weeks.
That spreadsheet is where tagging drift is manufactured. One person types Facebook, the other types facebook, and analytics platforms group campaign values by exact string match with no fuzzy matching at all. You now have two rows where you meant one, each carrying half the sessions and each looking too small to defend in a budget conversation.
Building the whole matrix in one pass removes the opportunity. There is one campaign name, applied everywhere. Every source and medium is normalised to one spelling before it is written, so the same channel typed two ways collapses into one instead of quietly becoming two rows three months later.
What it does with what you paste
Destinations go in one box, one URL per line. A bare domain is fine and gets https assumed, because that is the overwhelmingly common paste and rejecting it teaches nobody anything. A line with a scheme that cannot carry a campaign tag, like a mailto: address, is rejected rather than coerced into something that looks valid.
Channels go in the other box, one source / medium pair per line. Slashes, commas, pipes and plain spaces all work as separators, since people type all four. A source with no medium is refused rather than guessed at, because a source that cannot be grouped against a medium is not a channel.
If a destination already carries UTM parameters, it says so. The existing values are replaced rather than merged, and that is worth knowing before you paste over tagging somebody else set up.
Consistency is the whole feature
Every value is written in one form: lowercase, with spaces turned into underscores. That is not a style preference, it is the only way the grouping works. The audit tool exists to find drift after it has happened; this one exists so it does not happen.
The output is also byte-stable. The same inputs produce the same URLs in the same parameter order every time, so two exports can be diffed and a rebuild does not look like a change. If you need to understand what any individual parameter is doing once the links are live, the link decoder breaks one down, and UTM naming conventions covers how to choose the values in the first place.
Where this stops
It builds links. It cannot tell you whether they worked, because a tagged URL only records anything once something is counting on the other end. That is the part Raydar does, on every click, including which campaign brought someone back a second time.