In this article
What the new report is meant to answer
The useful question is no longer just “Which query spent money?” It is “What promise did this query lead the visitor to see, and which page tried to fulfil it?” A relevant query can still produce a poor lead when the ad overstates the offer or the destination serves the wrong service. Separating these layers keeps a marketer from solving every problem with a negative keyword.
This article is a diagnostic worksheet for that journey. It does not replace the general AI Max setup guide, and it does not assume that a new interface automatically improves performance. Your objective is to identify a mismatch you can verify, then change the smallest part of the journey necessary to address it.
Build a query-to-offer worksheet
Start with one campaign and one business goal. Record the conversion action you actually optimize toward, the account timezone and the comparison dates. “Conversions” might mean enquiries, purchases or several different actions. A search-term ranking is misleading if one row represents a phone click and another represents a completed purchase.
Use four fields for each review row: query theme, message shown, destination and outcome. When an account does not expose the announced journey view, use the native reports and a manual review. Do not join separate aggregate reports and claim you reconstructed individual visitor paths. Put “not available” in the message field when you cannot establish which asset was served.
| Field | Record | Why it changes the decision |
|---|---|---|
| Query theme | Commercial intent, location, service | Determines whether the traffic is relevant |
| Offer | Price, eligibility, service promise | Reveals whether the ad creates the wrong expectation |
| Destination | Final page and mobile experience | Shows whether the user can complete the promised task |
| Outcome | Named conversion and observed quality | Prevents cheap enquiries being mistaken for useful customers |
Diagnose the mismatch before changing targeting
Imagine a hypothetical plumbing advertiser that serves “emergency repair” ads but sends visitors to a general services page. The query is relevant. The failure could be that the page does not show emergency availability or a visible way to call. Excluding the query would discard useful demand rather than fix the destination.
A different row might show “plumbing courses” leading to a repair booking page. That is a query-intent problem. A third might show a correctly matched query and page, but an asset promising an unsupported same-day service. That is an offer problem. These diagnoses require different remedies even if all three rows have a high cost per lead.
Sort by meaningful exposure, not the most dramatic percentage. An apparent 100% conversion decline on two clicks cannot justify the same urgency as sustained wasted spend. Include the denominator, conversion delay and amount at risk beside every proposed change.
Run a controlled correction
Write the hypothesis in one sentence: “Visitors seeking urgent service cannot find the urgent booking action on the destination.” Keep targeting and the offer stable while fixing the relevant page element. Record when it changed, what stayed constant and when enough conversion data will have arrived to evaluate it.
If the problem is the offer, change the claim and verify the page agrees. If it is query relevance, review exclusions in the native account before approving them. Avoid changing the page, bid target and query controls on the same day. You would otherwise struggle to explain which change caused the result.
The review should include lost opportunity. A narrower query set can improve average CPA while reducing the number of useful leads. Report both volume and efficiency. A better average is not a successful intervention when it excludes the customers the business needs.
Use 1ClickReport to organize the evidence
1ClickReport is an AI marketing analyst for connected marketing data. Use it to review campaign performance and search terms, then supply any native asset or destination exports needed for this particular question. Do not assume a newly announced Google interface is already available through a third-party connector.
For this connected Google Ads account, review the last 28 complete days of search terms and campaign performance. State the conversion action and timezone. Group query themes by intent and spend. Separate query mismatch, offer mismatch and landing-page mismatch. Use the native exports I attach for asset or destination details; mark anything missing. Recommend one change to review, not an automatic edit.Pair the ad-side review with GA4 landing-page engagement when that source is connected. An engaged-session metric cannot identify the exact creative a visitor saw, but it can help locate a destination worth inspecting. Ask the analyst to keep those evidence scopes separate.
Questions to ask before scaling
Has the current offer changed? Are locations or service restrictions clear? Is the named conversion a real business outcome? Could late conversions change the conclusion? Does the report expose the field being used to justify the recommendation? These questions are more valuable than a blanket instruction to “let AI find more traffic.”
Keep an evidence note with the query family, observed problem, proposed fix and approval owner. Revisit it after a mature comparison period. If the result contradicts the hypothesis, retain that finding rather than quietly replacing the narrative.
Frequently asked questions
Is the unified AI Max report available everywhere?
The September announcement said more details and availability would be shared later. Check your own Google Ads account before planning a workflow that depends on it.
Can separate exports recreate a complete individual journey?
Not reliably. Aggregated query, asset and landing-page reports do not establish a visitor-level join unless a supported report explicitly supplies that relationship.
Should a poor destination become a negative keyword?
Only if the query is irrelevant to the business. A relevant query with a weak landing experience usually deserves a destination or offer review first.
Sources and research notes
Platform statements are linked below. The checklists, worked examples and analyst prompts are original editorial frameworks. Examples labeled hypothetical are not customer results. Research checked October 13, 2026; account availability may differ.