Google Ads Explanations: What It Covers, What It Misses (2026)

Updated July 2026.

Google Ads Explanations is a built-in diagnostic that tells you why a metric moved between two comparable time periods — and within its lane, it's genuinely useful and free. Its limits are equally real: per Google's own documentation, it only works within the last 90 days, only triggers for large changes, and only explains the metrics your campaign optimizes toward. And by design it only sees Google's side of the world: it can tell you that your budget changed, but not who changed it, why, or what it did to your Meta spend the same week. Here's what it covers in 2026, and where a real causal layer picks up.

What Google Ads Explanations does

When a compatible campaign or ad group shows a significant period-over-period change in a key metric (impressions, clicks, cost, conversions), Google surfaces an "explanation" you can open from the performance table when comparing two date ranges. It draws on factors Google can observe internally: bid and budget changes, bid strategy target changes, auction competition shifts, search interest changes, conversion-tracking status, and eligibility issues.

Coverage has expanded steadily; Google's help center lists support across most major campaign types in 2026 — Search, Performance Max, Shopping, Display, Video, App, Demand Gen, Hotel and Smart campaigns among them, with AI Max for Search campaigns included as that rollout matures. Check Google's "why you might not have explanations" page for the current compatibility list, because it changes.

The documented limitations (per Google)

1. 90-day window. Explanations are only available for periods within the last 90 days. Quarter-over-quarter or year-over-year questions are out of scope.
2. Only large changes. Explanations won't trigger for smaller fluctuations, in percentage or absolute terms. Gradual erosion — the 3%-a-week CPA creep that quietly compounds into a 40% problem — never trips the threshold.
3. Only optimized-for metrics. Explanations trigger for the metrics your bid strategy targets. On a Maximize Conversions campaign, don't expect explanations for a CPC anomaly that hasn't yet dented conversions.
4. No partial data. Date ranges including today are excluded, and comparison ranges must be equal-length. Reasonable — but it means "what's happening right now" is precisely the question it won't answer.
5. Not all campaign/strategy combinations. Some bid strategies and setups simply don't produce explanations.

A fair assessment: where Explanations genuinely helps

For a single campaign that just took a visible hit, Explanations is a good first click. It's fast, it's integrated, and it reliably catches the mechanical causes: you raised a budget, a bid target changed, auction competition intensified, your conversion tag went stale, a campaign became limited by budget. For solo advertisers running a handful of campaigns, it may be all the diagnostics they need. Use it — it's free and it's right there.

How to get the most out of Explanations

1. Compare two equal-length, completed date ranges (e.g., last full week vs. the week before). Never include today.
2. Stay inside the 90-day window, and diagnose promptly — an anomaly you investigate two weeks late has less context available.
3. Look at the metric your bid strategy optimizes toward first; that's where explanations actually trigger.
4. Treat the output as a lead, not a verdict. "Auction competition increased" tells you the category of cause — confirm it in Auction Insights before you act on it.
5. Cross-reference every explanation against Change History. Explanations tells you a budget changed; Change History tells you when and via which access path.

Where it stops: the questions Explanations can't answer

1. Who made the change, and why? Explanations can point at "budget changed." It doesn't tell you whether that was your teammate, your agency, a script, or an auto-applied recommendation — or that the same person made four other changes that day. That requires an audit trail: see who changed my Google Ads account.
2. Cross-campaign and cross-account causality. A budget shift in one campaign cannibalizes a sibling's auctions; a PMax change bleeds into brand Search. Explanations analyzes the campaign in front of you, not the interactions between them.
3. Cross-platform effects. If Meta prospecting was cut on Monday and Google brand conversions sagged Thursday, no Google-side tool will connect those dots.
4. Small, compounding changes. Below the significance threshold, nothing triggers. Slow leaks are invisible.
5. History beyond 90 days. "When did this actually start, and what did the account look like before?" needs a permanent record, not a rolling window.
6. Proactive detection. Explanations is reactive — you look at a table, notice a change, and click. Nobody is watching while you're not.

A concrete scenario that shows the gap. Monday: an agency user lowers a shared budget and loosens a tROAS target across a portfolio. Tuesday: brand Search CPCs drift up as PMax picks up queries the Search campaign lost. Thursday: conversions sag account-wide. Open Explanations on the Search campaign Friday and you may get "auction competition increased" — true, locally, and almost useless. The actual root cause is two changes made by one person in a different campaign four days earlier. No single-campaign, Google-side diagnostic can assemble that chain; a permanent change record plus cross-campaign causal analysis can.

What a cross-account causal layer adds

This is the gap The Ad Spend is built for. It monitors Google Ads, Meta, LinkedIn, TikTok, and Reddit accounts every ~6 hours with 1,900+ detection algorithms — so anomalies get flagged whether or not you happened to be comparing date ranges. It keeps a permanent, version-controlled record of every change (who, what, when) with no 90-day expiry — the piece Google's native change history makes painful at scale (see Google Ads change history alerts). And its causal inference engine traces a performance move to the exact change that caused it, across campaigns and platforms — the answer to "why did this happen" in minutes, not a day of forensics. The transparency problem this solves is bigger than one feature — see the AI bidding transparency gap and causal inference in advertising for the deeper argument.

The honest recommendation: keep using Explanations for the quick single-campaign look. Then connect your account and get the actual cause traced to the exact change — with Slack alerts when something moves, not just when you go looking. Free performance and budget pacing alerts are included on the free tier.

FAQ

Why don't I see explanations in my Google Ads account?

The usual reasons, per Google: the change wasn't large enough, your date range includes today or exceeds the 90-day lookback, the ranges aren't equal-length, the metric isn't one your campaign optimizes toward, or your campaign type/bid strategy combination isn't supported.

Does Explanations tell me who changed my account?

No. It can attribute a performance shift to a category of change (e.g., a budget or target modification) but not to a person, script, or auto-applied recommendation. For who/what/when, you need change history — natively or via a monitoring layer that records it permanently.

Does Explanations work for Performance Max in 2026?

Yes — Google lists Performance Max among supported campaign types, and coverage has expanded across most major types. Availability still varies by bid strategy and situation, so check Google's current compatibility documentation.

Is Google Ads Explanations enough for an agency managing many accounts?

As a spot-check tool, yes; as a monitoring system, no. It's reactive, single-account, 90-day-limited, and blind to who made changes. Agencies typically pair it with continuous monitoring, permanent change logs, and alerting.