Ad updates that are about your competitors, not about our checking

What this solves

• Entries that were really about us. "Running 24 ads we had not seen before" counted what our check had newly recorded. Each check reads a limited, varying sample, so the same advertising could be announced as new over and over while nothing had actually changed. Across a month, the typical entry claimed about three quarters of a competitor's whole account was new — while their ad count was unchanged nine times out of ten.
• The same event twice. A batch of new ads produced one entry counting them and a second saying the total had risen — the first restated as arithmetic.
• Everything was "major". Any competitor adding five ads earned the label, so routine activity at a large advertiser looked as urgent as a small one doubling their output.
• A total that promised more than it knew. "Running 77 active ads" read as an exact live count.

What changed

• A launch is reported only where a launch can be shown. Meta publishes the date an ad started, so an entry now means "they started these since we last looked" — each ad counted once, not a rolling figure re-announced on every check.
• LinkedIn publishes no start dates, so it no longer claims a competitor launched anything. It reports what it can stand behind: their ad count went up.
• One entry per change, carrying how many are new, the total, and what it was before.
• "Major" is judged against that competitor's own advertising — ten ads is routine for someone running five hundred, and a big move for someone running twenty.
• The total says what it can defend: "at least 77 running" where a check reached its sampling limit.
• Entries already recorded have been re-labelled, de-duplicated, and the ones describing our own checking removed.

Where to find it

Competitors → What changed, and each competitor's window → History.

Useful for

Anyone scanning the feed for what rivals actually did. The entries that stand out are now the ones that earned it.