Something changed my Google Ads budgets and it wasn't me — how do I find out what did?

The commodity answer is "check Tools and Settings, then Change history." The version worth reading is narrower: it may not be a person. Scheduled scripts edit accounts on a timer, they can silently reverse your changes overnight, and if one is running then **no before-and- after read on those campaigns is clean.

On this page

  • How do I audit inherited negative keywords?
  • How do I tell a scheduled script apart from a person?
  • Can I disable a Google Ads script through the API?
  • How far back does Google Ads change history go?
  • Why check change history before any before-and-after test?

How do I audit inherited negative keywords?

Pull change history across the window and group by user and client type. Not just by date.

Across one book of 12 accounts over 29 days: 1,141 of 1,148 events were mine. The other 7 were scheduled scripts running on two client-organisation seats, firing at 06:00, 06:36 and 07:37, all on brand campaigns — including a budget cut that silently reversed one of my own increases, and two target loosens weeks later.

Seven events out of 1,148. Invisible in any summary view. And they were changing the dials on campaigns I was actively steering.

How do I tell a scheduled script apart from a person?

To separate several scheduled runs from one script iterating your accounts, compare the seconds.

Second-tight repeats at a fixed minute — 06:00:32, 06:00:50, 06:00:56 across different weeks — mean distinct scheduled runs. A single script iterating accounts drifts by minutes, because per-account work varies.

Then group by seat. Two different user accounts means two different owners to go and ask.

Can I disable a Google Ads script through the API?

You can see the damage. You cannot remove the cause programmatically. Scripts are interface-only — Tools and Settings, Bulk actions, Scripts. There is no way to disable one through the API.

Two consequences:

  • Check the parent manager account as well as the sub-accounts. A manager-level script with an account selector does not appear in any individual account's script list.
  • If the authorising seat belongs to the client, the durable fix is the client disabling it.

Deleting somebody else's authorised script tends to reappear. The better ask: have your campaigns excluded from the script's selector rather than the whole script switched off.

How far back does Google Ads change history go?

Lookback is 30 days, exactly. Not 31. And results cap out at roughly 300 rows per query, so on a busy account the visible window silently starts later than you asked for. "No third-party events" is unprovable on a truncated span — say so rather than certifying it.

Change history lags. Changes surface hours later. Zero rows immediately after you push something is not evidence the push failed. Confirm against the live setting instead.

Why check change history before any before-and-after test?

Before you trust any before-and-after on an account you did not build, run this. If somebody else's automation is nudging the dials, your test window is contaminated and you will attribute their changes to your work, or yours to theirs.

Related questions

How do I find out how much of my Performance Max spend is going to my own brand searches?

Performance Max does not report brand spend as a line item, so you build it from the search terms it does expose plus the network split — and accept that the figure is a floor, not a total. Read the full answer.

How do I work out what target ROAS to actually set for my store?

Derive it from contribution margin and your fixed per-order costs, not from a benchmark. The formula everyone uses — 1 ÷ gross margin — is a breakeven floor, not a target, and it ignores the per-order costs that decide whether an order is actually worth having. Read the full answer.

How do I work out what percentage to put in a Google Ads seasonality adjustment?

From your own measured conversion-rate change in the equivalent past period, not from a guess — and only for short, predictable events. If you cannot point at a comparable historical window, do not set one. Read the full answer.

Written by Usama Khan, who runs Google Ads for e-commerce brands and home-services companies at AdventurAds. Figures are from accounts under active management and are anonymised. Last reviewed 2026-08-26.

Work with me