This is the first chapter of the AdsPulse playbook. It covers the repetitive work that today drags you into the interface account by account: here it becomes a single request, the same for every client. If this is your first time with AdsPulse, start here: 01, 02 and 04 take two minutes to try and save time every day.
For each use case you will find how often we run it, the prompt to copy (change the parts in square brackets), what to expect in reply, and the mistake we see most often. Three of the four are read-only; the second writes to the account, always after your confirmation.
Multi-account morning report
Before opening Google Ads, to know where to look. It is the request that opens the day: one, for every account, and the answer says which ones are worth going into.
For every account in my MCC, compare yesterday and the last 7 days against the 30-day average on spend, conversions and CPA. Show me only the accounts with deviations above 20% and the most likely cause.
What you get. One line per account, the anomalies at the top, a probable cause for each. The accounts running as usual do not appear: that is the point.
Watch out. If you import offline conversions, the last 3-4 days are incomplete: judge only the settled days. A CPA that “exploded” yesterday on an account with offline imports is almost always an upload delay, not a problem.
Search terms and negatives
When spend grows faster than conversions. It is the work we describe in the article on governing negative keywords, compressed from an afternoon to one request.
On [account], find the search terms from the last 14 days with more than €10 of spend and zero conversions. Group them by theme, propose the negatives with the right match type and wait for my confirmation.
What you get. Wasted terms grouped by theme, the proposed negatives with their match type, and a preview of what would be added to the shared list. Until you confirm, nothing changes.
Watch out. Ask for a cross-check against the terms that convert: a negative that is too broad blocks those too. “Before proposing, check that no negative excludes terms with conversions in the last 90 days” is the sentence that avoids the most common damage.
Pacing and budget
At the start of the week and in the last ten days of the month, when you need to know whether the monthly budget will last to month end or whether some campaign is leaving conversions on the table.
On [account], which campaigns are limited by budget and which are under-spending this month? For the limited ones, show Google’s budget simulation and the impression share lost.
What you get. The month’s spending pace, the limited campaigns, and Google’s estimate of clicks and conversions at different budgets.
Watch out. Raise the budget only where impression share is lost to budget. If it is lost to rank, the problem is the bid or the quality, and more budget will not fix it. The distinction is in the answer: read it before deciding.
What changed?
A KPI moves suddenly and nobody knows why. In an account with several hands on it, between the interface, automated rules, Editor and the API, the right question is not “what is happening” but “what was done”.
The CPA of [campaign] has gone up since Monday. Cross-reference the change history with daily performance: which change coincides with the deterioration, who made it and from where?
What you get. The list of changes with author, origin (interface, API, rules, Editor) and value before and after, next to the day’s numbers. The comparison between the two columns is almost always the answer.
Watch out. The full detail only exists for the last 30 days. Beyond that, what remains is the list of what changed, without the before and after values: if the problem is older, ask for a period comparison instead of a change-by-change one.
Where they sit in the week
In the playbook’s typical week, 01 and 02 open every morning, 03 sits on Monday and 04 is used on request. If you manage many accounts, write the request once and ask it for every account in the MCC: each call names its own account, so the analyses run in parallel without mixing the data.
Next chapter
Once the routine is in place, the next step is understanding what is wrong before touching anything: chapter 2, Audit and diagnosis, with the full account audit, the tracking check, the landing page audit and the PMax and Shopping X-ray. The full map of the sixteen use cases is in the playbook overview.