How to read this list
Two kinds of prompt appear below, and the difference decides whether they are worth anything.
Connected prompts need an assistant that can reach your Google Ads account. They ask about your data, so in a plain chat they produce a polite guess. If you have not set that up, the step-by-step is here for ChatGPT and here for Claude, or watch it done in two minutes on Google Ads and ChatGPT, connected.
Standalone prompts work in any chat window, because they are about writing, structuring, or thinking rather than about your numbers.
Each prompt is followed by one line on what a good answer looks like. That line matters more than the prompt: it is how you tell a real answer from a fluent one.
Diagnosis: what is actually happening
1. "Show me last week's spend, clicks, conversions and cost per conversion by campaign, with the change versus the previous week." *(Connected)* A good answer is a table with both weeks side by side, not just last week's totals. Comparison is the whole point.
2. "Which of my campaigns are limited by budget right now, and how much spend am I leaving on the table?" *(Connected)* Should name specific campaigns and quantify the gap. "Some campaigns may be budget-limited" is a non-answer.
3. "What changed in this account in the last 14 days, and did each change do what it was supposed to?" *(Connected)* Should list changes with dates and the metric that moved after each. This is the prompt that catches your own mistakes.
4. "Compare my conversion rate this month against the same month last year, by campaign, and flag anything that dropped by more than 20%." *(Connected)* A good answer respects seasonality by using the year-over-year window you asked for rather than quietly substituting month-over-month.
5. "My cost per conversion went up 30% this month. Walk through the possible causes in order of how likely they are for this specific account, using the data." *(Connected)* The phrase "for this specific account" is doing the work. A good answer cites your numbers; a bad one recites a checklist.
6. "Which campaigns are spending the most and converting the least? Rank by wasted spend." *(Connected)* Should define what it counted as wasted and show the ranking, so you can disagree with the definition.
Search terms and negatives: the money leak
7. "List search terms from the last 30 days with more than €20 of spend and zero conversions." *(Connected)* A plain list with spend per term. Set the threshold to something meaningful for your budget.
8. "Group last month's search terms by intent and show spend and conversions per group." *(Connected)* Good answers produce a handful of legible groups — research, comparison, brand, competitor, irrelevant — not fifty micro-buckets.
9. "Which search terms are triggering my ads but are clearly a different product or audience? Explain why for each." *(Connected)* The explanation is what makes this reviewable. You are checking its judgement, not accepting it.
10. "Add these as negative keywords at the campaign level: [paste your reviewed list]." *(Connected, writes)* Should confirm exactly what it added and where. Never approve this from a list you have not read.
11. "Suggest 30 negative keywords for a [your business] campaign, grouped by theme, before I have any search-term data." *(Standalone)* Useful for a launch. Expect obvious themes — jobs, free, DIY, competitors, wrong geography.
12. "Are any of my match types working against me — broad terms picking up traffic that a phrase or exact version should be catching?" *(Connected)* Should name the keyword pairs in conflict, not lecture about match types generally.
Campaign builds
13. "Build a search campaign for [product] targeting [location], budget [amount] a day. Create it paused." *(Connected, writes)* Should come back with full structure — ad groups, keywords with match types, responsive search ads — and confirm it is paused. If it does not say paused, check the account before doing anything else.
14. "Give me a keyword plan for [product] with 4 to 6 tightly themed ad groups, and mark the match type for each keyword." *(Standalone)* Tight themes are the test. One ad group with forty loosely related keywords is a bad plan wearing a good format.
15. "Write 15 headlines and 4 descriptions for a responsive search ad for [product]. Each headline must be genuinely different in angle, not a rewording." *(Standalone)* Count the distinct angles, not the headlines. Five rewordings of "best price" is one headline.
16. "Review this campaign structure and tell me what a Google Ads specialist would criticise." *(Connected or standalone)* Adversarial framing produces far better feedback than "is this good?".
17. "What sitelinks, callouts and structured snippets should this campaign have, based on what my landing pages actually offer?" *(Standalone)* Should reference real page content, so paste the page or give the URL.
18. "Set up a Performance Max campaign for [product] and explain what each asset group is for. Create it paused." *(Connected, writes)* The explanation is the deliverable. PMax hides a lot; make it narrate.
Ad copy and landing pages
19. "Here is my landing page: [URL]. Write ad copy that promises exactly what the page delivers — no claims the page does not support." *(Standalone)* The constraint is the point. Message-match failures are a top cause of high cost per conversion.
20. "Rewrite these three headlines to lead with the outcome instead of the feature." *(Standalone)* Should change the sentence structure, not swap adjectives.
21. "My click-through rate is below 2% on this ad group. Here are the ads and the keywords — what is the mismatch?" *(Connected)* Should point at specific keyword-to-headline gaps.
22. "Read my landing page and list the objections a buyer would have that the page never answers." *(Standalone)* The best of these prompts. Every unanswered objection is a conversion you paid for and lost.
Budgets and bidding
23. "Which campaigns should get more budget based on the last 60 days, and which should get less? Show the reasoning." *(Connected)* Should account for whether a campaign was budget-limited. Reallocating away from something that never had room to prove itself is a classic error.
24. "Cut the daily budget on any campaign with a cost per conversion above [threshold], and tell me what you would change before changing it." *(Connected, writes)* "Before changing it" is the safety clause. Read the proposal, then approve.
25. "Is my bidding strategy right for this campaign's conversion volume? Be specific about how many conversions a month I actually have." *(Connected)* Smart Bidding needs volume. A good answer names your number and says plainly whether it is enough.
26. "If I increase this campaign's budget by 50%, what does the data suggest happens to cost per conversion? Say what you cannot know." *(Connected)* The "say what you cannot know" clause separates useful forecasting from confident nonsense.
The weekly routine
27. "Every Monday, give me: last week versus the week before by campaign, any campaign pacing unusually, search terms above €20 with no conversions, and the single thing you would fix first." *(Connected)* One prompt, the whole review. Save it and send it once a week; consistency beats sophistication.
Three habits that make any of these better
Ask for the comparison, not the number. A number is trivia. A number next to last week's is information.
Ask it to say what it cannot know. Assistants are fluent by default and uncertain only when asked. Adding that clause turns a confident guess into a labelled one.
Never approve a write you have not read. Every prompt above that changes the account is safe for exactly one reason: you looked at the list first. Campaigns created paused and budget changes proposed rather than applied are what make that possible — see the documentation for how the approval model works.