Opus Growth

Strategy

Ready-made prompts for Claude, ChatGPT and other assistants connected to your ad accounts. Copy one, fill in the brackets and send.

Strategy

Shift budget between Google Ads and Meta based on 30-day CPA

A single campaign table across Google Ads and Meta and a line-by-line budget reallocation plan built on a clear CPA rule.

Compare my Google Ads and Meta campaigns over the last 30 days and tell me where my budget should move. My monthly ad budget is [monthly budget] and my target cost per conversion is [target CPA].

1. For every active campaign on both platforms, pull spend, conversions, cost per conversion, conversion rate and current daily budget. Use the same date range for both platforms.
2. Build one table sorted by cost per conversion. Include each campaign's share of total spend and its share of total conversions.
3. Mark any campaign with fewer than 10 conversions as "low confidence" and do not move its budget yet.
4. For Google Search campaigns, add impression share lost to budget.

Decision rule:
- A campaign with at least 10 conversions and a CPA more than 30% above target loses budget.
- A campaign below target CPA that loses more than 20% impression share to budget, or is a Meta ad set spending its full budget, gains budget.
- Never change any single budget by more than 20% in one step.

Deliver a reallocation table with these columns: campaign, platform, current daily budget, proposed daily budget, reason. Do not change anything yet. I will mark which lines I accept. Then show me a preview of those exact changes and wait for my confirmation before you apply them.

Strategy

Month-end pacing check across Google Ads and Meta

A projection of total month-end spend against your monthly budget, plus a previewed rebalance that lands total spend within 5% of target.

I have a fixed monthly ad budget of [monthly budget] split across Google Ads and Meta. Check whether I will land on it this month.

1. Run budget pacing for Google Ads. Note which campaigns will overspend and which will underspend by month-end.
2. Pull Meta spend month-to-date for each campaign and ad set, along with their current daily budgets. Project month-end spend at the current daily rate.
3. Show one table with these columns: campaign/ad set, platform, spend so far, projected month-end spend, current daily budget, cost per conversion this month. Add the projected combined total versus my budget.

If the projected total is off by more than 5%, propose new daily budgets for the remaining days that bring it back within 5%. Follow these rules when you rebalance:
- Take money first from campaigns whose cost per conversion is above [target CPA].
- Add money first to campaigns with the lowest cost per conversion that are not already spending their full budget.
- Do not cut any campaign by more than 30%.

List each proposed change with the old value, the new value and the reason. Preview all changes and wait for my approval before applying anything.

Strategy

Check that conversions are comparable before moving budget between channels

A trust table that compares platform-reported conversions to GA4 and shows each channel's comparable cost per conversion.

Before I move any money between Google Ads and Meta, I need to know whether their conversion numbers can be compared at all. Use the last 30 days.

1. Google Ads: audit conversion tracking health. List the primary conversion actions with their counting type (every or one), click-through and view-through lookback windows, and attribution model.
2. Meta: list the pixels with their last fired time, the custom conversions, and the conversion events my campaigns optimize for. Note that Meta's default attribution includes view-through conversions.
3. GA4: pull key events by session source/medium for the same 30 days. Confirm which key events correspond to the platform conversions.
4. Build a table per channel with these columns: platform-reported conversions, GA4 key events from that channel, ratio, spend, platform CPA, GA4-based CPA.

Flag every issue that inflates or hides results:
- Counting set to "every" on lead actions.
- Duplicate or secondary actions being counted.
- A pixel that has not fired in 24 hours.
- A platform-to-GA4 ratio above 1.5.

Finish with a clear answer: which CPA should I use to compare channels, and what must be fixed first. Do not change any setting. If a fix is needed, describe it and I will ask for a preview separately.

Strategy

New client onboarding: one-page competitive landscape brief

A one-page brief mapping who advertises in the client's market, where, with what message, and where the client stands today.

We are onboarding [client name] (domain [client domain], Google Ads account [account]). Build a competitive landscape brief before our kickoff call.

1. Get an account snapshot so we know current spend, conversions and active campaign types.
2. Run a live SERP analysis for the client's 5 most important commercial queries: [5 core queries]. Record which domains show in paid positions and which show in the top 5 organic results.
3. Take the 4 domains that appear most often in paid positions. For each one, pull its paid keyword footprint (keyword count and estimated paid traffic) and its live ads from public ad libraries.
4. Run a keyword gap between the client domain and those 4 competitors.

Deliverable, one page:
- competitor table with these columns: domain, paid presence (queries appeared on out of 5), organic presence, estimated paid keyword count, channels with live ads, main message
- where the client is stronger, where it is weaker and where it is absent
- 5 prioritized questions to ask the client at kickoff, such as margins, which competitor they lose deals to, and which offers they can legally match

Read-only. No changes to the account.

Strategy

Rewrite weak RSA assets to stand out from competitor ads

Compares your responsive search ad copy with competitors' live ads, flags generic or low-CTR assets, and prepares in-place rewrites built on verified facts. Nothing changes until you approve the preview.

In Google Ads account [account], campaign [campaign name], I want our responsive search ads to stop sounding like the competition.

Facts you may use about [client name] (only these, plus anything you can read on our live landing page): [paste real numbers, guarantees, process details, certifications, review counts].

1. Find the campaign and its ad groups and ads with their IDs.
2. Pull the live Google ads of [competitor domains, up to 4] from the Ads Transparency Center, region [country]. List the claims and phrases that recur across them, such as "fast delivery", "best price" or "trusted by X". If a competitor returns no ads, say so and continue with the others.
3. Read our RSAs in this campaign: every headline and description, final URL, Ad Strength and Google's action items. Then pull asset-level performance for the last 60 days, with a 1,000-impression minimum.
4. Read the final URL of each ad as a live page and note any concrete facts it states (numbers, guarantees, delivery times, proof points). Use these together with the facts I pasted above.
5. Flag our headlines and descriptions that either (a) repeat a recurring competitor claim almost word for word, or (b) are in the bottom third by CTR among assets with 1,000+ impressions. If fewer than 6 assets clear 1,000 impressions, use lifetime figures instead and mark the result as low confidence.

For each flagged asset, write one replacement built on a specific fact from my list or the landing page: a number, guarantee, process detail or proof point that competitors do not mention. Never invent a claim. If no fact fits, write "needs a fact from you" instead of a replacement. Keep headlines to 30 characters and descriptions to 90. Keep at least 3 headlines that contain the main keyword.

Pins: if the ad details show which assets are pinned, leave pinned assets untouched. If pin status is not visible in the data, list the assets you plan to change and ask me which of them are pinned before preparing edits.

Show a before/after table per ad with the reason for each swap and the source of the fact used. Then prepare the edits with the partial in-place RSA update (remove and add only the flagged assets) so performance history is kept. Run it as a preview with confirm=false and wait for my approval before applying anything.

Strategy

Meta ad set funnel breakdown: creative, page, offer or checkout?

Each Meta ad set gets a diagnosis of the funnel step where it loses buyers, based on clear rate thresholds.

Strategy

Which offer angle sells? RSA headline offer analysis and rewrite

Tags each RSA headline and description by offer angle, compares conversion economics per angle, and prepares in-place rewrites for the weakest ads using only offers you actually run.

In Google Ads account [account], campaign [campaign name], analyse RSA headlines and descriptions from the last 60 days. Include only assets with at least 1,000 impressions.

If fewer than 10 assets clear 1,000 impressions, drop the floor to 300 and say so. If that is still fewer than 10, use lifetime figures and mark every conclusion as low confidence.

1. Find the campaign and its ads with their IDs, then pull asset-level performance.
2. Tag each asset by offer angle: percentage discount, free shipping or shipping threshold, bundle or gift, guarantee or returns, urgency or stock, installments, or no offer. Show your tagging so I can correct it.
3. Aggregate the results per angle: impressions, CTR, conversion rate, cost per conversion and conversion value per cost. Show how many assets sit behind each angle.
4. Name the winning angle and the losing angle. If an angle has fewer than 3 assets or under 20 conversions, say the comparison is not reliable for it.

Then read the ads in this campaign with their Ad Strength and pick the two ads with the weakest Ad Strength or the lowest value per cost.
- Propose replacing their two worst-performing assets with new variants of the winning angle.
- Keep headlines to 30 characters and descriptions to 90.
- Pins: if the ad details show pinned assets, leave them as they are. If pin status is not visible, list the assets you plan to replace and ask me whether any are pinned before preparing edits.
- Use only offers we actually run, which are: [current offers].

Show the exact before and after copy, then prepare each ad's change with the partial in-place RSA edit (remove the two assets, add the two new ones) so performance history is kept. Run it as a preview with confirm=false and do not apply anything until I approve each ad.

Strategy

Product-level ROAS triage for Shopping and Performance Max

Queries product-level performance across Shopping and Performance Max, sorts every product into heroes, drains, zombies and middle, and turns the split into a campaign and asset group structure plan with differentiated ROAS targets. Read-only.

For Google Ads account [account ID], my target ROAS is [target ROAS, e.g. 4 = 400%].

Step 1. Run account_snapshot (platform=google_ads, recent_days=30) so you know which Shopping and Performance Max campaigns exist, whether they are serving, their budgets and their current bidding.

Step 2. Run google_ads_query with this GAQL:
SELECT segments.product_item_id, segments.product_title, segments.product_brand, campaign.name, campaign.advertising_channel_type, metrics.impressions, metrics.clicks, metrics.cost_micros, metrics.conversions, metrics.conversions_value FROM shopping_performance_view WHERE segments.date DURING LAST_30_DAYS ORDER BY metrics.cost_micros DESC
Convert cost_micros to account currency. If the result looks capped by the row limit, raise limit and tell me how many products you analyzed, so I know whether the long tail is included. If the same item appears in more than one campaign, sum it per product and also note which campaigns it spends in.

Classify each product:
- Heroes: ROAS at or above [target ROAS] with at least 3 conversions.
- Drains: cost above 2x the average cost per product in this dataset and ROAS below 50% of target.
- Zombies: under 100 impressions in 30 days, meaning the product is effectively not served. Note that products with zero impressions do not appear in this view at all; if I paste my full feed item ID list here [optional: paste item IDs], list the missing ones as zombies too.
- Middle: everything else.

Deliver:
1. Totals per group: number of products, share of spend, share of conversion value and ROAS.
2. The top 15 drains with cost, clicks, conversions, average order value (conversion value / conversions), ROAS and a likely offer diagnosis. Base the diagnosis on what the data shows: an average order value too low to cover the cost per conversion, high clicks with very low conversion rate, or spend concentrated in one campaign. Price above market, bundle candidate and entry-item-only are hypotheses, because feed prices, competitor prices and margins are not in this data. Label them as hypotheses unless I paste prices or margins here: [optional: item ID, price, margin %].
3. A structure plan. Explain how to separate heroes, drains and zombies into their own campaigns or asset groups with different ROAS targets, which existing campaigns from Step 1 each group would come from, and the expected impact on spend and conversion value, with the assumptions stated.

This is a plan only. Do not change campaigns, budgets, bidding or listing groups. If I later approve any part of it, show a confirm=false preview of each change first and create any new campaign paused.

Strategy

Pre-launch go/no-go check for lead and call tracking

Audits Google Ads conversions, call reporting, GA4 key events and links, WordPress tags and the contact form before launch, then gives a single go or no-go verdict.

We launch ads for [business website] next week in Google Ads account [account name]. Before we spend anything, confirm that every lead and every call will be counted once and fed to bidding.

Check the following:
1. Google Ads conversion actions and their settings:
   - Lead actions should count "one", not "every".
   - Primary vs secondary should be correct. Use the "included in conversions" flag and the conversion goals (which categories are biddable), not the legacy primary_for_goal flag.
   - Click-through lookback windows should be set to a sensible value.
   - Nothing should be paused or removed. For zero recorded data, keep in mind that a new account with no traffic yet will show zero; flag it as "not yet testable" rather than broken.
   - Note each action's conversion ID and label, so I can match them to the site tags.
2. Call tracking: read the account's call reporting state with a GAQL query on the customer resource (customer.call_reporting_setting.call_reporting_enabled, call_conversion_reporting_enabled, call_conversion_action). Then check whether a "calls from ads" conversion action exists and is primary.
3. GA4: find the property for this site, list its key events, and check whether it is linked to this Google Ads account.
4. WordPress: list every place the site fires GTM or GA4, and flag any duplicate containers or tags.
5. Contact form [form name]: show its required fields and the email address notifications go to. Then prepare ONE test entry with every text field clearly marked "TEST - ignore". This sends a real email to that recipient, so show me the preview first and send it only after I approve. Report the result: mail_sent, mail_failed, validation_failed or spam. mail_sent only proves WordPress handed the mail off, not that it reached the inbox, so ask me to confirm it arrived.

Deliver one table with these columns: item, status (OK / risk / broken / not yet testable), evidence, fix. End with a single verdict: go or no-go.

For any fix that changes settings (call reporting, primary or secondary status, counting type, lookback window), show me the preview of the exact change from the tool and wait for my approval. Do not change anything silently.

Strategy

Competitor ad and SERP landscape before entering a new city

Maps local competitors' ads, paid keywords and the live results page in a new city, then turns the gaps into positioning angles, RSA headlines and auctions to avoid at launch.

We are entering [city] for [service]. The main local competitors are [2-3 competitor domains]. Our 3 core queries are [3 core queries]. Map the landscape before we write a single ad.

First check seo_history and reuse anything recent for these domains and queries, so we do not spend SEO add-on credits twice. Tell me which steps will cost credits before running them.

1. Fetch each competitor's live ads from the Google Ads Transparency Center for [country]. Small local advertisers often have thin or no coverage there; if a result is empty, say so and rely on steps 2 and 3 for their copy.
2. Pull the keywords each competitor bids on in paid search, with search volume, CPC, estimated monthly spend, the ad headlines and descriptions, and the landing pages. An empty result means no ads in this scan, not that they never advertise.
3. Run a live SERP analysis for each core query with location [city]. Note how many ads appear and who runs them, whether the results show a local pack (if the returned data includes it), and what takes space above the organic results.
4. In Google Ads account [account name], find the geo target id for [city] and read the historical metrics for the competitors' main keywords and our core queries in [language]: monthly searches, competition and the top-of-page bid range. This is our auction cost estimate, since competitor-level auction insights are not available through the tools.

Deliver a table with one row per competitor and these columns:
- offer and price claims
- urgency claims (24/7, same day)
- trust signals (reviews, guarantees, licences)
- call to action
- top 10 paid keywords
- what is missing from their ads
Mark "no ad data" where a competitor showed no ads in any source.

Then give me:
- 5 positioning angles nobody in this market is using, each backed by a gap from the table
- 10 RSA headline ideas of 30 characters or less, built on those angles, with the character count next to each
- keywords where 3 or more competitors overlap, with the top-of-page bid range, flagged as expensive auctions we may want to avoid at launch, plus cheaper long-tail alternatives from Keyword Planner

Analysis only. No account changes.

Strategy

Weekly red/amber/green triage of all Google Ads accounts under the MCC

Scores every connected Google Ads account against CPA and ROAS targets with week-over-week change, diagnoses red accounts and writes a management-ready summary. Read-only.

List every Google Ads account under our manager account, including client accounts, and tell me which ones the portfolio report covers. Then run the portfolio report for the last 7 days. For the week-over-week comparison, run a client report per account for 7 days, which gives the % change versus the previous 7 days. For each account, show cost, conversions, CPA, conversion value, ROAS and the % change versus the prior week.

Our targets: [target CPA and target ROAS, one line per account if they differ, or one default for all]. Use these status rules:
- Red: CPA is more than 20% above target, or ROAS is more than 20% below target.
- Amber: the gap is between 10% and 20%.
- Green: everything else.

Judge lead generation accounts with no conversion value on CPA only, and say so in the table. Separately, flag any account whose spend moved more than 30% week over week, whatever its colour.

For each red account, pull an account snapshot with the last 7 days as the recent window. If conversions dropped while clicks held steady, also run a conversion tracking health check before blaming performance. Give the single most likely cause in one line, for example: budget limited, conversion drop, cost spike in one campaign, or possible tracking issue. Name the campaign behind it.

Deliverable:
1. One table sorted from red to green.
2. A 5-bullet summary written for non-specialists that I can paste into my weekly update to management.

This is read-only. Do not change anything in any account.

Strategy

Month-end budget pacing across all accounts with a rebalancing proposal

Projects month-end spend per campaign against each account's monthly cap, then proposes daily budget changes that cut weak-CPA campaigns and add only where budget limits impression share, applied line by line after approval.

List the connected Google Ads accounts. Our monthly spend caps are:
[account name or ID: monthly cap, one per line]

For each account, run budget pacing alerts for the current month with a 10% threshold, and pull a campaign-level performance report for this month for spend to date, conversions and CPA.

Build one table with these columns: account, campaign, daily budget, spend to date, projected month-end spend, account cap and over/under status. Add an account total row and the days left in the month. Flag every campaign projected to finish more than 10% over or under its pro-rata share.

Then propose daily budget changes that bring each account within ±5% of its cap by month-end, calculated as remaining cap ÷ days left. Use these rules:
- When cutting, start with the campaigns that have the weakest CPA this month. Do not judge a campaign with fewer than 3 conversions on CPA alone; say so.
- When adding budget, give it only to Search campaigns where impression share analysis shows the loss is due to budget, not rank. Adding money where the loss comes from rank or quality will not buy more traffic.
- Do not move budget between accounts.
- Respect the safety limits: no single increase above +50%, no more than +100% within 24 hours, no cut deeper than -80%. If a campaign needs more than that, say so and propose pausing instead of a deeper cut.

Show the proposal as a before/after table per campaign, with the expected change in monthly spend. Do not apply anything yet. Wait for my explicit approval. Then apply only the lines I confirm, running each budget change as a preview first and confirming it only after I approve that preview.

Strategy

Conversion tracking standards check across every client account

Audits every Google Ads account under your MCC against one conversion standard (counting, attribution, lookback, primary actions, GA4 link). Returns a deviation matrix and a prioritized fix list, and applies approved fixes only after a preview.

I need to confirm that every Google Ads account under our MCC measures conversions the same way. Without that, CPA comparisons in my management reports are meaningless.

Our standard is:
- Lead actions count "One" (ONE_PER_CLICK).
- Purchase actions count "Every" (MANY_PER_CLICK).
- Data-driven attribution.
- 30-day click-through lookback window.
- Only these actions are primary (included in the Conversions column): [primary conversion actions].

Start by listing every client account under the MCC, including all accessible accounts. Then run these checks for each account, one account at a time:
1. Run the conversion tracking health check for the last 30 days. List any action that is paused or removed, and any primary action that recorded zero conversions. A secondary action with zero data is expected, so only note it.
2. Read the conversion actions and their settings: category, counting type, attribution model, click-through lookback window and whether each one is included in the Conversions column. Also read the conversion goals, since in goal-based accounts the goal category decides what is biddable.
3. List our GA4 properties and the Google Ads links on each one. Map each linked Ads customer ID back to the accounts in the matrix, so every account shows which GA4 property it is linked to, or "not linked".

Deliverable: a matrix with accounts as rows and these columns: lead counting type, purchase counting type, attribution model, lookback window, primary actions, zero-data primary actions and GA4 link. Mark every deviation from our standard. Below the matrix, list the fixes in priority order. Put a primary action recording zero data first, since it misleads Smart Bidding.

For each fix, say how it will be applied:
- Counting type or lookback window: set_conversion_settings.
- Primary or secondary role: update_conversion_goal in goal-based accounts, update_conversion_action otherwise.
- Attribution model, pausing or removing an action: these cannot be changed through the API, so list them as manual steps in the Google Ads interface.

Do not change anything yet. Once I approve specific fixes, show a preview of each call and apply it only after I confirm.

Strategy

Build one master negative keyword list from waste shared across accounts

A vetted list of non-converting search terms wasting money in two or more accounts, applied as a shared negative list only after approval.

Our accounts keep paying for the same junk queries, such as jobs, free, DIY and competitor support terms. I want one standard shared negative keyword list across all Google Ads accounts under the MCC.

Steps:
1. In each account, list the existing shared negative keyword lists and read their contents.
2. Run smart negative keyword suggestions for the last 90 days in every account.
3. Keep only the terms that meet all of these conditions:
   - They spent money with zero conversions in 2 or more accounts.
   - They have not converted in any account over the last 90 days. Confirm this with the search terms report.
   - They do not contain our brand or product names: [brand and product terms].
   - They are not already covered by an existing negative.

Deliverable: a table with these columns: term, suggested match type (exact, phrase or broad), number of accounts affected, total 90-day spend and an example search term. Prefer phrase match unless a broader match would clearly block converting traffic.

Do not create or edit anything yet. After I approve the final list, show a preview of the manage_shared_negative_list call for each account, including the list name "[master list name]" and the campaigns it will attach to. Apply only after I confirm.

Strategy

Forecast the daily budget you need for your peak month

Compares last year's peak weeks, impression share lost to budget and rank, and keyword volume seasonality with today. Builds three budget scenarios with stated assumptions, and previews the recommended budget change before applying it.

Help me set the daily budget for campaign [campaign name] for my peak month, [month].

1. Find the campaign and pull last year's weekly performance for that month and the 4 weeks before it: cost, clicks, conversions, conversion rate and cost per conversion, segmented by week. Compare this with the last 30 days. If the campaign did not run last year, say so and use the closest comparable campaign I name: [fallback campaign or "none"].
2. Pull search impression share, and the share lost to budget versus rank, for last year's peak month and for the last 30 days.
3. Take my top 10 keywords by cost. Use their Keyword Planner historical metrics for my market to estimate how much search volume rises in that month compared with the current month.

Then build three scenarios: conservative, base and aggressive. For each scenario, give:
- daily budget
- expected daily clicks
- expected daily conversions
- expected cost per conversion

State your assumptions in one short table before the scenarios:
- CPC: my current CPC rising by 10-20% in peak. Say which figure each scenario uses.
- Conversion rate: last 30 days for conservative, last year's peak-month rate for aggressive, and the midpoint for base, unless the data argues otherwise.
- Clicks: capped by the extra impression share actually available, since budget can only buy what is lost to budget, not what is lost to rank.

Reject any scenario where the projected cost per conversion goes above [target CPA]. These are estimates built from the data, not a platform forecast, so label them that way.

Also show this month's pacing, so I know where I stand today.

Show the recommended budget change as a preview and wait for my confirmation before applying it. A single increase above +50% is blocked, so if the target is higher, propose a staged schedule of increases with dates.

Strategy

Client portfolio triage: scale, hold or fix each Google Ads account

Compares the last 30 days with the previous 30 across all connected Google Ads client accounts, checks tracking health and impression share lost to budget, and labels each account Scale, Hold, Fix or Data first. Read-only.

Run a monthly triage across all my connected Google Ads client accounts. Compare the last 30 days with the previous 30 days. Targets: [target CPA] and [target ROAS]. If some accounts have their own targets, use these instead: [account: target list].

1. List the connected accounts. Run the portfolio report for 30 days and for 60 days. Get the previous 30 days for each account by subtracting the 30-day figures from the 60-day figures, then calculate CPA, ROAS and % change yourself. Name any account the report skipped as inaccessible.
2. Run conversion tracking health (30 days) for every account, because the Data first label depends on it.
3. Run campaign-level impression share for every account that passes the result test for Scale or Hold. Report IS lost to budget as a spend-weighted average across its Search campaigns. Mark accounts with no Search campaigns "n/a".

Classify every account with this rule:
- Data first: any primary (included in Conversions) action that is paused, removed or has recorded zero data. This label applies regardless of results.
- Scale: CPA at or below target, or ROAS at or above target. Conversions must be stable or rising, tracking healthy, and impression share lost to budget above 15%.
- Hold: on target but IS lost to budget at or below 15%, or fewer than 30 conversions in the period (too little data to push).
- Fix: CPA more than 25% above target, ROAS more than 25% below target, or conversions down more than 20%.
If an account records no conversion value, judge it on CPA only and say so.

Deliver one table sorted by spend with these columns: account, spend, conversions, CPA, ROAS, % change, IS lost to budget, label, one-line reason. Below the table, name the three accounts where extra budget is most likely to pay back, and say why.

This is read-only. Do not change anything.

Strategy

Where should the next Google Search budget go? Budget-limited vs rank-limited

A budget allocation plan that gives extra spend only to Search campaigns that are on target and truly limited by budget, previewed before applying.

For [client account], I have about [extra daily budget] per day to add. Decide which Search campaigns should get it.

Use the last 30 days. For each enabled Search campaign, pull:
- spend, conversions and CPA
- impression share, IS lost to budget and IS lost to rank
- this month's pacing projection

Decision rule:
- A campaign is eligible for more budget only if all three are true: CPA is within 10% of [target CPA], it has at least 20 conversions, and IS lost to budget is above 10%.
- If IS lost to rank is higher than IS lost to budget, do not add budget. Flag the campaign as a bid or Quality Score problem instead.
- If pacing already projects an overspend, exclude the campaign.

Split the extra budget across eligible campaigns in proportion to their IS lost to budget. Cap each increase at 30% per step so Smart Bidding does not swing.

Show a table with these columns: campaign, current daily budget, proposed budget, IS lost to budget, CPA, reason. List excluded campaigns separately with the reason for each.

Prepare the budget changes as a preview and wait for my approval before applying anything.

Strategy

Meta ad set rules: scale, keep, refresh or kill

Applies a clear scale, keep, refresh or kill rule to every active Meta ad set using 7-day, 28-day, week-over-week and lifetime data, then previews budget and status changes for your approval.

Apply a scale, keep, refresh or kill rule to every active ad set in [Meta ad account]. Use the last 7 days for the decision and the last 28 days as context. Benchmark: [target CPA].

1. List campaigns and ad sets. Note each ad set's optimization goal, start date and budget, and whether its campaign carries the budget (campaign budget optimization). Budgets in these lists are in cents, so convert them before proposing amounts.
2. Pull ad set level insights for the last 7 days and for a 28-day date range: spend, results, cost per result, CTR, frequency and CPM. Count "results" as the action that matches each ad set's optimization goal and say which action you used.
3. For the week-over-week trend, pull the last 14 days in 7-day increments at ad set level and compare cost per result and CTR between the two weeks.
4. Pull lifetime results per ad set (maximum date range) for the learning check.
5. For ad sets that might need a refresh, pull ad level CTR and frequency for the last 7 days, and the ad list with ids and statuses.

Rules, checked in this order:
1. Kill: spent more than 2x the target with zero results, or cost per result more than 50% above target in both the 7-day and the 28-day window. Propose pausing it. If the ad set started less than 7 days ago, flag it instead of pausing.
2. Refresh, not kill: cost per result rose versus the previous week, frequency is above 3.5 and CTR fell week over week. Name the specific ads to replace. Propose pausing a tired ad only once its replacement is live, and tell me what new creative to brief.
3. Scale: 7-day cost per result at least 20% below target, at least 15 results and frequency under 3. Propose a budget increase of no more than 20%.
4. Keep: within 20% of target, or fewer than 50 results since launch (likely still in learning).

Ad sets in campaigns with a campaign-level budget cannot be scaled at ad set level. Flag those and propose no ad set budget change.

Output a table sorted by spend with these columns: ad set, spend, results, cost per result, frequency, CTR trend, label, exact change (with the new budget in currency units).

Show all budget and status changes as a preview, and apply only the ones I approve.

Strategy

ChatGPT Ads test verdict: go, extend or stop

A one-page go, extend or stop verdict for a ChatGPT Ads test, benchmarked against the client's Google Search CPA, with tracking checked first.

We ran a ChatGPT Ads test for [client], and I need a go, extend or stop decision.

1. Pull daily delivery for the test campaigns since launch: impressions, clicks, CTR, spend, conversions and cost per conversion. Break it down by country and device.
2. Check conversion tracking: are the pixel and conversion events attached to these campaigns, and are they receiving events?
3. If impressions are low, run the delivery diagnosis.
4. Pull the client's Google Ads Search CPA for the same dates as the benchmark.

Decision rule:
- Fix first: tracking is not receiving events or delivery blockers are present. Give no verdict until these are fixed, and list the fixes.
- Go: cost per conversion within 30% of the Google Search CPA, with at least 15 conversions. Propose a 25% budget step-up.
- Extend: fewer than 15 conversions, but CTR and cost per click are in line. Propose two more weeks at the current budget.
- Stop: spend above [test budget] with cost per conversion more than 2x the Google Search CPA.

Deliver a one-page summary with the numbers, the verdict and the next step.

Show any budget or status change as a preview, and apply it only after my approval.

Strategy

Move budget from weak campaigns to budget-limited ROAS winners

A ranked list of campaigns that lose sales to budget caps, the donors that fund them, and previewed budget changes on Google and Meta.

Strategy

Free up brand search budget where organic already wins

Match brand queries across Search Console and Google Ads, check whether competitors are bidding on your brand, and preview a 14-day brand budget cut that shifts money to high-ROAS non-brand campaigns.

Check whether our brand search campaigns are paying for clicks we would get organically anyway. Our brand terms contain [brand name]. Use the last 60 days.

1. Run gsc_list_sites and pick the property for [our domain]. With gsc_performance (dimension query, query filter [brand name], days 60), list brand queries with clicks, impressions, CTR and average position.
2. With search_terms_report (days 60, sort by cost), list the search terms that contain the brand term, with cost, clicks, conversions and conversion value. Use list_entities and get_report at campaign level to identify which campaigns are brand campaigns, with their daily budgets, cost and ROAS.
3. Competitor check on our brand. Run seo_history first so we do not pay twice for data we already have. Then:
   - Run seo_paid_keywords for [main competitor domain] and tell me whether any of the keywords it bids on contain our brand term.
   - Run seo_serp_analysis for our top 3 brand queries by paid cost and note any paid results from other advertisers.
   - Run competitor_ads for [main competitor domain] (platform google) to see what their live ads say.
   - Run impression_share_analysis at campaign level for the brand campaigns. High top-of-page and absolute-top share with little rank loss is a sign we are not being challenged.
   All of these are snapshots of a single moment, not bidding history. Treat "no competitor found" as "none seen in this scan" and say so in the table.

Deliverable: a table matched by query with these columns: organic position, organic CTR, organic clicks, paid clicks, paid cost, paid ROAS, competitor seen (yes / no / not checked), and a verdict.
- Reduce: organic position is 1.5 or better, organic CTR is above 30%, and no competitor was seen on the term in this scan.
- Keep: a competitor was seen, organic position is worse than 3, or the term was not covered by the competitor checks.

Then estimate how much monthly budget the Reduce terms free up, based on their share of brand campaign cost. Name the non-brand campaigns with ROAS above [target ROAS] that should receive it. A single budget increase above +50% is blocked, so split larger moves across days if needed.

Propose a 14-day test with brand campaign budgets cut by 50%. Record today's baseline now: average daily brand clicks from Search Console plus brand paid clicks from the search terms report. Success means total brand clicks (paid plus organic) drop by less than 10% versus that baseline. I will ask you to re-run steps 1 and 2 on day 14 and compare. Show every budget change with set_campaign_budget as a preview (confirm=false) and wait for my approval before applying anything.

Strategy

Carve out a capped ChatGPT Ads test budget with kill criteria

Check ChatGPT Ads purchase tracking, fund a capped 21-day test from your weakest Google Ads and Meta campaigns, and get clear day-7 and day-21 kill and scale checks, all previewed before anything changes.

I want to test ChatGPT Ads as a new channel without hurting overall ROAS. Plan it step by step.

1. Readiness check. Run manage_chatgpt_ads_accounts (action list) to confirm the account is connected, then chatgpt_ads_conversion_setup with check_recent_events=true. Report:
   - whether a pixel exists
   - whether an order_created conversion event setting exists, and which campaigns it is attached to (none is fine if we have no campaigns yet, it gets attached when the campaign is built)
   - whether any events arrived in the recent test window, and whether they carry a value, if the event data shows it
   Decision rule: if there is no pixel or no order_created event setting, stop and list exactly what is missing. If both exist but no recent events showed up, do not treat that as broken (the window is only about 15 minutes). Tell me to place a test order and re-run the check, and continue planning steps 2 to 5 in the meantime, clearly marked "pending tracking confirmation".
2. Funding. Pull the last 30 days at campaign level from Google Ads (get_report, platform google_ads) and Meta (meta_campaign_insights, date_preset last_30d, level campaign). Rank campaigns by ROAS and propose taking [test budget] from the lowest ones. For Meta, check with list_meta_campaigns whether the budget sits on the campaign; if it does not, use list_meta_adsets and cut at ad set level instead. Show a table: source campaign, platform, current daily budget, proposed daily budget, ROAS. Note that a single cut deeper than 80% is blocked on Google Ads, so pause instead of cutting that far.
3. Test design. Run for 21 days. Show the current limits with chatgpt_ads_spend_limits first. Then propose an account spending limit window equal to [test budget] for the 21 days (end_date is exclusive). Spending limits need an admin key with billing permission and postpaid billing; if the tool says they are not available, propose capping the test campaign itself with a lifetime budget of [test budget] instead, created paused.
4. Kill and scale rules. Nothing runs these automatically, so give me two short check-ins I will ask you to run on day 7 and day 21:
   - Day 7: chatgpt_ads_insights for the test campaign. If CTR is under 0.5% or there are zero purchases (conversions=true), propose a pause with chatgpt_ads_status as a preview. If impressions are near zero, run diagnose_chatgpt_ads_delivery before judging.
   - Day 21: if ROAS is at or above 80% of [target ROAS], propose scaling. If it is below 50%, propose stopping. In between, recommend another 7 days with no budget change.
5. Daily read: list the metrics I should check in chatgpt_ads_insights (spend, impressions, clicks, CTR, purchases, purchase value) and which ones need conversions=true.

Preview the spending limit and every budget cut (confirm=false). Wait for my approval before creating or changing anything. Anything new on ChatGPT Ads is created paused.

Strategy

Build a week-by-week peak season budget calendar per channel

Combine last year's Google Ads, Meta and TikTok results with search demand seasonality to build a weekly peak season budget calendar, a 10% reserve and check-in triggers you run during the season.

Help me plan budget for our peak season, [peak period dates], with a total budget of [total peak budget].

1. Find the accounts with list_accounts, list_meta_ad_accounts and list_tiktok_ad_accounts.
2. Pull last year's daily spend, conversion value and ROAS for the same period plus the 4 weeks before it, on Google Ads, Meta and TikTok. Use get_report with start_date and end_date; check describe for each platform to see if a daily breakdown is available. If it is not, use google_ads_query (segments.date), meta_insights_query (time_increment 1) and tiktok_report_query (stat_time_day). Roll the days up into weeks yourself.
   If a channel was not running last year or returns no data for those dates, say so plainly. For that channel, use its last 8 weeks as a baseline and mark its numbers as "no seasonal history".
3. Search demand seasonality for [main product keywords] (up to 5). Run seo_history first to reuse any stored result, then seo_keyword_trends (needs SEO add-on credits). If the add-on is not available, use keyword_historical_metrics instead with language_id [language id, e.g. 1037 Turkish] and the market's geo id; it is monthly rather than weekly, so say which resolution you used. I want to see when demand starts rising and when it peaks.

Deliverable:
- A weekly table for each channel with last year's spend share, ROAS, and the week where ROAS peaked.
- A proposed calendar with columns for week, channel, budget and share of total. Follow these rules:
  - Ramp prospecting 2 to 3 weeks before demand rises, so learning finishes before the peak.
  - Weight each channel's peak-week budget by last year's ROAS during that week. For channels with no seasonal history, weight by their recent ROAS and cap them at their recent spend level plus 50%.
  - Hold back 10% as a reserve to deploy mid-period.
- In-flight triggers as a checklist. Nothing runs automatically, so tell me to come back every 3 days during the peak and ask you to pull the last 3 days per channel with get_report:
  - if a channel's 3-day ROAS is more than 20% above [target ROAS], propose releasing part of the reserve to it
  - if it is more than 30% below target for 3 days, propose moving its budget to the strongest channel
  Note that budget increases above +50% in a single step are blocked on Google Ads, Meta and TikTok, so plan reserve releases in steps.

This is planning only. Make no changes now.

Strategy

Find high-intent local keywords competitors bid on and you don't

A ranked shortlist of competitor paid keywords with local volume and suggested match types, added only after your approval.

Strategy

Replace generic RSA headlines that sound like every competitor

Underperforming headlines that duplicate competitor copy, with differentiated replacements shown as a preview before any edit.

Pull the RSA asset performance for campaign [campaign name] over the last 60 days. Only consider headlines with at least 500 impressions.

Flag a headline as weak if it meets either condition:
- its CTR is more than 20% below the ad's average
- it has clicks but its conversion rate is below half of the ad's average

Next, pull the live search ads of [competitor domains] from the ad libraries. Mark every one of my headlines that says essentially the same thing as a competitor headline, for example "24/7 service", "affordable prices" or "expert team" with no proof attached.

For each headline that is weak, generic, or both, propose a replacement that:
- keeps the same theme
- adds something specific and checkable: a response time, a guarantee term, a review count, years in business, or a district name
- fits within 30 characters

Do not touch headlines that perform well, even if they are generic. Keep pinned positions as they are.

Show a before/after preview for each ad with its current Ad Strength. Edit the existing ads in place, so they keep their history, and only after I approve each change.

Strategy

Decide what to do with competitor brand searches in your account

Spend and conversions from competitor brand queries, a check on who bids on your brand, and a keep-or-exclude decision per term.

Pull my search terms for the last 90 days and find every term that contains one of these competitor brand names: [competitor brand names]. Include misspellings.

For each term, show: cost, clicks, conversions, cost per conversion and CTR. Compare each with my account average cost per conversion.

Then check the paid keywords of [competitor domains] and tell me whether any of them bid on my brand name, [my brand name].

Apply this decision rule to every competitor brand term:
- Cost per conversion up to 1.5x my average, with at least 2 conversions: keep it, and suggest a dedicated ad group so I can control bids and copy.
- Spend above 2x my average cost per conversion with zero conversions: add it as a phrase match negative keyword at campaign level.
- Anything in between: keep watching, no action yet.

If competitors bid on my brand, tell me whether my brand campaign's impression share looks affected.

Show the proposed negative keywords as a preview with their estimated monthly savings. Do not add anything until I approve.

Strategy

Find paid clicks I already win organically and free up that budget

Matches strong organic queries from Search Console with paid search terms and keywords. Checks the live Google results page for competitor ads on each query, then previews bid cuts and pause tests that free budget without losing clicks.

I want to stop paying for clicks I would get organically anyway, and move that money to channels that bring new customers. Use the last 90 days. My Search Console property is [site], and my Google Ads account is [account].

1. From Search Console, pull queries with their clicks, impressions, CTR and average position. Keep the queries where my average position is 3 or better and organic CTR is above 15%.
2. From Google Ads, pull the search terms report and the keyword report for the same 90 days. Match them to those queries by exact text or a close variant. For each match, include clicks, cost and conversions, and note whether it is a keyword in my account or only a search term triggered by a broader keyword.
3. Read each matching campaign's bid strategy from the account snapshot. Keyword bid changes only take effect on manual CPC or Enhanced CPC. On Smart Bidding campaigns, use pause or negative instead of a bid cut.
4. Competitor presence: Google does not expose auction insights per query here, so check the live results page instead. First check SEO history so I do not pay twice for a query that was already scanned. Then run a SERP analysis for the 20 overlapping queries with the highest paid cost, plus [my brand name], and record whether competitor ads appear and which domains run them. Treat this as a snapshot of today's results page, not a 90-day average, and mark queries outside the top 20 as "not checked".

Deliver a table with these columns: query, organic position, organic CTR, organic clicks, paid clicks, paid cost, paid conversions, keyword or search term, bid strategy, competitor ads on SERP (yes/no/not checked), recommendation. Use these recommendations:
- Keep: competitor ads appear on the SERP, the query was not checked, or paid cost per conversion is below [target CPA].
- Reduce bid by 30%: no competitor ads, organic position 1-2, and the keyword is on manual CPC or Enhanced CPC.
- Pause as a 4-week test: no competitor ads, organic position 1, and paid cost above [minimum spend] in the period. If the query is only a search term with no keyword of its own, propose it as an exact match negative on that campaign instead.

Total the monthly spend these actions would free, using the 90-day cost divided by 3. Then show previews of the keyword bid changes, the keyword pauses and any negatives, and wait for my approval before applying anything.

After 4 weeks I will come back. Then pull the same queries from Search Console and Google Ads for the 4 weeks before and the 4 weeks of the test, and compare total clicks (paid plus organic) and conversions so I can decide whether to keep the changes.

Strategy

Is my Google Ads spend adding leads beyond what Google Maps brings?

Lines up 6 months of Business Profile actions and search terms against Google Ads spend, conversions and search terms. Estimates how much paid spend looks incremental and which paid terms cover searches your free listing misses. Read-only.

I run a local business at [business location]. I want to know whether my Google Ads budget brings customers I would not get from my free Business Profile listing anyway. Use the last 6 months.

1. Business Profile actions by month. Find my location with gbp_list_locations. gbp_insights only returns a total for the last N days, so call it six times with days=30, 60, 90, 120, 150 and 180. Subtract each total from the next one to get calls, direction requests and website clicks for each 30-day block. Label the blocks by date range, not calendar month.
2. Business Profile search terms. Pull the terms people used to find the listing with gbp_search_keywords, covering the last 6 months. This data lags about 2 months, so state which months it actually covers.
3. Google Ads by the same blocks. Pull spend, clicks, conversions and cost per conversion for the last 180 days. Use performance_report at campaign level with a monthly segment if it supports one. Otherwise use google_ads_query with date ranges that match the six 30-day blocks from step 1. Pull search terms for the same 180 days with search_terms_report, and read campaign targeting (locations and radius) with get_campaign_targeting.
4. Put both sources in one table with one row per 30-day block: Profile calls, directions, website clicks, Ads spend, Ads clicks, Ads conversions. Note whether blocks with higher ad spend show higher total actions (Profile plus Ads), or whether Profile actions drop when Ads rise, which would suggest ads are taking credit for customers who would have found me anyway.
5. Compare the Profile search terms with the paid search terms, using only the months both sources cover. List:
   - paid terms that already bring strong Profile impressions (likely overlap)
   - paid terms where the Profile does not appear at all (likely incremental)
   - paid terms that convert but fall outside my targeted locations or radius, if any.

Deliver a short verdict:
- An estimated range for the share of ad spend that looks incremental. Say plainly that this comes from correlation over 6 blocks, not proof.
- Which paid terms to keep funding: the ones the Profile does not cover.
- Whether part of the budget, up to [max amount to shift], would be better spent elsewhere.
- A simple test I could run later to measure true incrementality, for example pausing the overlapping terms for a few weeks and watching Profile actions. Describe it only, do not set it up.

This is analysis only. Do not change any campaign or the Business Profile.

Strategy

Build a 12-month budget split by season and channel

Turns 12 months of search demand, GA4 channel data and last year's Google Ads and Meta cost per conversion into a month-by-month budget split, with clear rules applied in a fixed order. Planning only.

Help me plan next year's ad budget of [annual budget] month by month, split between Google Ads and Meta. My business sells [main product or service] in [target country or city]. My main conversion is [main conversion, e.g. lead form or purchase].

1. Search demand. Find the geo id for my market with geo_target_search. Then get month-by-month search volume for my 5 main keywords, [main keywords], with keyword_historical_metrics for the last 12 months, using that geo id and the right language_id (1037 for Turkish). Sum the 5 keywords per month and turn the result into a demand index where the average month equals 100. If any keyword has no volume data, cross-check its seasonality with seo_keyword_trends (it takes up to 5 keywords and costs SEO add-on credits, so tell me before calling it).
2. GA4. Pull sessions and key events for the last 365 days with ga4_report, dimension=date and dimension2=channel. Group the days into months and keep paid search, paid social, organic search and direct.
3. Ad platforms. For the same 12 months:
   - Google Ads: monthly spend, conversions and cost per conversion. Use performance_report with a monthly segment if it supports one, otherwise google_ads_query with segments.month.
   - Meta: use meta_insights_query with fields spend, actions and cost_per_action_type, time_increment=monthly, at account level. Use the action type that matches my main conversion.

Build the plan with these rules, applied in this order:
1. Base split: each month's budget = annual budget × (that month's demand index ÷ sum of all indexes).
2. Cost check: in months where last year's blended cost per conversion (Google Ads plus Meta) was more than 40% above the yearly average, cut that month by 15%. Redistribute the freed budget to the other months in proportion to their demand index.
3. Cap: no month may exceed 2x the lowest month. If a month breaks the cap, trim it to the cap and spread the excess across the months below the cap. Repeat until it holds. The 12 months must still add up exactly to [annual budget].
4. Channel split: Google Ads gets more weight in high-demand months because it captures existing searches. As a default, scale Google Ads' share linearly from 50% in the lowest-index month to 80% in the highest-index month. Meta never gets less than 20% of any month's budget.
If two rules conflict, say which one won and why in the reason column.

Deliver a 12-row table with these columns: month, demand index, total budget, Google Ads, Meta, reason. Below the table, list the assumptions I should revisit each quarter, including any months where data was missing or thin. Planning only, no changes to any account.

Strategy

Competitor offer and messaging matrix from live ad libraries

Pulls competitors' live ads from the Google Ads Transparency Center and the Meta Ad Library, then builds a side-by-side matrix of promises, offers, CTAs and proof, plus table-stakes messages and open angles for your client.

For my client [client name], pull the live ads of these competitors from the public ad libraries. For each competitor I give both the website domain and the brand or Facebook Page name, because Google's library searches best by domain and Meta's by brand name:
[competitor 1: domain / brand name]
[competitor 2: domain / brand name]
[up to 5 competitors]
Market: [country code, e.g. TR]

For each competitor:
- Run competitor_ads with platform=google, query = the domain, region = the market above.
- Run competitor_ads with platform=meta, query = the brand name, region = the market above. If Meta returns nothing or returns ads from a different advertiser with a similar name, try the domain or an alternative spelling once, and tell me which query matched. If the Meta library is unavailable because no Meta account is connected, say so and continue with Google only.
- Keep only ads that clearly belong to that competitor. If you are unsure an advertiser is the right one, list it separately instead of mixing it in.

Build one table with one row per competitor and these columns:
- number of live ads found, split by library (Google / Meta)
- formats (text, image, video, carousel), only if the library returns format information; otherwise write "not returned"
- main promise or value proposition, quoted from the ad copy
- offer type (discount %, free trial, free shipping, guarantee, none)
- price or price anchor if shown in the copy
- call to action wording
- proof elements used (reviews, numbers, awards, client logos)
- the oldest ad still running, with its start date only if the library returns one

Under the table:
1. List the 3 messages that appear in at least half of the competitors. These are table stakes. Our ads must match them or explain why we don't.
2. List up to 5 angles that no competitor or only one competitor uses. Rate each angle for fit with [client name]'s positioning ([one-line positioning statement]): high, medium or low.
3. If start dates are returned, flag every competitor ad running for 60+ days. Long-running ads are likely profitable, so describe exactly what they do: hook, offer, format and CTA. If no start dates come back for a competitor, say that ad age could not be measured for them instead of guessing.

Use only what the ad libraries return. Note that these libraries show live creatives, not spend, targeting or results. Mark anything you infer as an inference. This is research only; do not create or change anything in my accounts.

Strategy

Competitor paid keyword gap, turned into an add list for approval

A ranked list of keywords competitors pay for that the client does not cover, checked against search volume, with a keyword addition preview.

Run a paid keyword gap for Google Ads account [account] against these competitor domains: [competitor domains, up to 3].

Steps:
1. Pull the paid keywords each competitor bids on.
2. Pull our keywords and our search terms from the last 90 days.
3. Keep only the competitor keywords that are not already a keyword in our account and have not appeared as a search term with 20 or more impressions. Those are real gaps.
4. Remove brand terms of any competitor and anything outside [client's product or service scope].
5. For the remaining keywords, get historical metrics: average monthly searches, competition and top-of-page bid range.

Deliverable: a ranked table of up to 30 keywords with these columns: keyword, which competitors bid on it, monthly searches, top-of-page bid range, suggested match type (exact or phrase) and the existing ad group it fits best. Rank by searches multiplied by relevance. Exclude anything whose high top-of-page bid is above [max acceptable CPC].

Then prepare a preview of adding the top 15 to the matching ad groups. Do not add anything until I approve the exact list.

Strategy

Who is bidding on our brand? Brand conquesting check and defense plan

A finding of which competitors appear on the client's brand searches, what their ads say, and whether the brand campaign is losing share.

Check whether competitors are conquesting the brand of [client name] in Google Ads account [account]. Our brand campaign is [brand campaign name].

1. Run a live SERP analysis for the brand name and its 3 most common variants, for example brand + reviews, brand + price and brand + login. List every paid advertiser that is not us.
2. For each of those advertisers, pull their live ads from public ad libraries. Note whether they use our brand name in the copy and which comparison or switching angle they push.
3. For the brand campaign over the last 30 days, compared with the previous 30, report: search impression share, absolute top impression share, share lost to rank, CPC and conversions.
4. From the search terms report, flag brand searches that combine our brand with a competitor name or with "alternative".

Decision rule: if absolute top impression share is below 90% or CPC rose more than 20% period over period, call it active conquesting. Recommend specific defensive steps, such as bid strategy target, copy that answers the competitor's angle, or sitelinks to comparison pages. Do not change anything yet. List the recommended changes so I can approve them one by one.

Strategy

Cross-channel funnel leak finder: where buyers drop off

Checks that purchase tracking works, then builds a step-by-step e-commerce funnel per channel from GA4 events, compares it with the previous 30 days and pinpoints the biggest leak. Analysis only.

Map my e-commerce funnel for the last 30 days and compare it with the previous 30 days. GA4 property: [property]. Do not touch any settings. This is analysis only.

1. Check tracking first. Use Google Ads conversion tracking health and the GA4 key event list to confirm that purchase is recording. If purchase shows zero or broken data, stop and tell me before analysing anything.
2. Pull the unified overview for Google Ads, Search Console and GA4 for the last 30 days, plus the GA4 overview KPIs.
3. From GA4, pull event counts split by event name and session source/medium (dimension event, second dimension source_medium). Do this once for the last 30 days and once for the last 60 days, and derive the previous 30 days as the 60-day figure minus the 30-day figure. The funnel steps are view_item, add_to_cart, begin_checkout and purchase. These do not need to be key events; total event counts are enough.
4. If a step does not appear at all, list it as missing and build the funnel from the steps that exist. If only purchase exists, fall back to sessions → purchase by source/medium and landing page, and tell me which ecommerce events I should add.
5. Build one table per channel (group source/medium into paid search, organic search, paid social, direct, referral, email, other). Show the count at each step and the step-to-step rate, for both periods.
6. Flag any step whose rate fell more than 20% versus the previous period. Also flag any channel whose rate at a step is more than 30% below the site average.

Finish with:
- The single biggest leak, estimated in lost purchases: for the flagged step, (site-average rate minus channel rate) × the channel's volume entering that step, carried through to purchase at the channel's own downstream rates. Show the calculation and call it an estimate.
- The most likely cause, chosen from traffic quality, product page, checkout or tracking, with the evidence from the data that points to it and what the data cannot prove.
- Three concrete checks to confirm the cause before we add budget.

Strategy

TikTok hook, offer or landing? Per-ad funnel diagnosis

Each TikTok ad is classified by where it fails, from thumb-stop to purchase, with previewed pause proposals for clear losers.

Strategy

Keyword demand and starting budget for a new local Search account

A keyword table grouped into ad groups with volumes, bid ranges, match types, starter negatives and a realistic daily budget.

I am planning a new Google Ads Search account for [service, e.g. emergency plumbing] in [city]. Nothing gets built yet. This is demand research only.

1. Find the location IDs for the city and its main districts.
2. Generate keyword ideas from my seed terms: [3-5 seed keywords]. Include city and district modifiers plus "near me" variants.
3. Pull historical metrics for the shortlist.

Deliver one table with these columns: keyword, avg monthly searches, competition, top-of-page bid range (low and high), intent (urgent job / price research / DIY / job seeker), recommended match type (exact or phrase only), and proposed ad group.

Rules:
- Drop keywords under 20 monthly searches unless they are clearly urgent and high intent.
- Put DIY, free, salary, course and job queries in a separate starter negative keyword list.
- Keep the plan to 3-6 ad groups, each built around one service.

Then estimate a starting daily budget that buys at least 10 clicks a day across the core ad groups at the high end of the top-of-page bid range. Tell me how many ad groups that budget can actually support without starving any of them.

Strategy

Rank service and area combinations to launch first from local data

Combines Business Profile search terms, Search Console queries and Keyword Planner volumes to rank the top 10 service + area combinations and the first move for each.

We want to expand from our current area into [new service area]. Before choosing campaigns, see where demand already exists, starting with our own data and filling the gaps with Keyword Planner.

1. For Business Profile location [location name], read:
   - The services and categories listed.
   - The search terms people used to find us over the last 6 months, including the low-volume ones. Note that this data lags about 2 months and low-volume terms come back as "fewer than N".
   - Calls, direction requests and website clicks. The tool returns totals for a window, so read the last 30, 60, 90 and 180 days and derive the approximate month-by-month trend from the differences.
2. From Search Console property [site], pull the last 180 days by page and query together. Keep queries that mention a district, neighbourhood or city name, or a specific service. Position only exists for queries that already earn impressions; mark everything else "no organic data".
3. For combinations that involve [new service area] or show little or no impressions in our own data, get demand from Google Ads account [account name]:
   - Find the geo target ids for [new service area] and our current area.
   - Run Keyword Planner with service + area seeds in [language] for those geo ids to get average monthly searches and competition.
   - For the shortlisted combinations, read the monthly search history to separate seasonal from steady demand.

Combine all sources into a ranked list of the top 10 service + area combinations. For each one show:
- monthly searches (Keyword Planner) and impressions found in our own data, with the source labelled
- current organic position and ranking page, if any
- whether the service is already listed on the profile
- recommended first move: paid search ad group, local landing page, or Business Profile post or service entry

Rank by demand first. Break ties in favour of combinations where we have no organic page in the top 10.

Separately, list any service people search for that is missing from our profile's service list. This is analysis only. Do not edit the profile or the Google Ads account.

Strategy

90-day budget, lead forecast and decision rules for a new city

Benchmarks an existing Google Ads campaign, compares keyword volume between cities and builds a three-scenario 90-day budget and lead model with checkpoint rules. Read-only.

We are opening [new city]. Use our existing market as the benchmark: campaign [existing campaign] in [existing city]. Our goal is [monthly lead goal] leads a month. Language of our ads: [language].

1. Pull the last 6 months of the existing campaign:
   - CPC, CTR and conversion rate.
   - Cost per conversion, split by calls and form leads. First list the conversion actions and their categories. If calls and forms are separate actions, run a GAQL query on the campaign segmented by conversion action name. If they share one action, say so and give one blended CPA.
   - Weekday performance, with a GAQL query segmented by day of week.
   - Search impression share, and whether lost share came from budget or rank.
2. Take our top 15-20 keywords by conversions from the keyword report. Look up the geo IDs for both cities and pull monthly search volumes for those keywords in each city, using the right language ID. Show the new city as a % of the existing one, and flag any seasonality. If Keyword Planner returns no data for this account, say so and ask me for a market-size ratio (population or households) to scale with instead.

Build a 90-day model for the new city in three scenarios: conservative, base and optimistic. For each month show budget, expected clicks, leads and cost per lead. In the first 30 days, assume CPC is 20% higher and conversion rate 20% lower than the benchmark. Cap clicks at what the new city's search volume can realistically supply. Say clearly if the lead goal is unreachable at any realistic budget.

Then write decision rules:
- Day 14: check search terms, negatives and Quality Score.
- Day 30: if cost per lead is above 1.5x the benchmark with 15+ conversions, fix the landing page and ads before adding budget.
- Day 60: move to a Target CPA bid strategy if we have 30+ conversions.
- Day 90: scale, hold or stop, with the numbers that decide each.

Planning only. No changes.

Strategy

Cross-channel spend vs conversion share by business unit

A per-business-unit view of Google, Meta, LinkedIn and Microsoft showing where spend share outruns conversion share, with reallocation candidates.

Management wants to know whether our channel mix makes sense for each business unit. Use the last 30 days.

1. List every connected account on Google Ads, Meta, LinkedIn and Microsoft Advertising. Map each account to a business unit using this list: [account to business unit mapping]. List any account you cannot map instead of guessing.
2. Pull spend, conversions and CPA per account, using the platform-reported conversions.
3. For each business unit, build a table with one row per channel. Show spend, conversions, CPA, share of the unit's spend and share of the unit's conversions.

Decision rules:
- If a channel's share of spend exceeds its share of conversions by more than 10 points, mark it as a candidate to cut.
- If the reverse is true by more than 10 points, mark it as a candidate to scale.
- If a channel has fewer than 30 conversions in the period, mark it "insufficient data" instead.

Add a short caveat that each platform counts conversions with its own attribution, so the shares are directional. Also flag any channel whose conversions look implausible, for example zero conversions alongside meaningful spend.

Deliverable: the tables plus 3 to 5 reallocation recommendations with suggested amounts. Recommendations only. Do not change any budget.

Strategy

Estimate what one quarter of LinkedIn ads will cost in buying season

Audience size, suggested bids, minimum budgets and a 90-day cost estimate for three LinkedIn targeting versions.

Strategy

Wind down seasonal keywords on Google and Bing as demand drops

Uses Keyword Planner history to find keywords whose demand usually falls sharply next month, matches them to Google Ads and Microsoft Advertising performance, and previews pauses, bid cuts and budget reductions for your approval.

My season for [product/service] ends in [month]. My market is [country], language [language]. Plan the wind-down on Google Ads and Microsoft Advertising, so I stop paying peak prices into falling demand.

1. Confirm the accounts with list_accounts and list_microsoft_accounts. Take a snapshot of both to get current campaign budgets and bid strategies.
2. Pull the last 30 days of keyword performance:
   - Google: the keyword report
   - Bing: keyword-level entity performance, plus the campaign, ad group and keyword lists so you have the ids needed for changes
3. Find the geo id for [country] and run Keyword Planner historical metrics on the keywords from both platforms, covering at least the last 24 months. Using the same two months in previous years, flag keywords whose monthly searches usually drop more than 40% from this month to next month.
   Keyword Planner has no Bing data, so use the Google volume as a proxy for Bing keywords. Match on the keyword text after lowercasing and trimming match-type symbols. Mark every Bing row as "Google volume proxy" and list Bing keywords with no matching text separately.
4. For each seasonal keyword, give: platform, volume this month versus next (historical average), cost, conversions and cost per conversion.
5. Assign one action to each keyword:
   - Pause: zero conversions and a volume drop of more than 40%
   - Lower bid by 20-30%: converting, but demand is falling
   - Keep: evergreen demand
   Keyword bids only apply under manual or enhanced CPC. If a campaign uses an automated strategy, do not propose keyword bid cuts there. Fold that demand drop into the budget recommendation instead.
6. List the campaign budgets that should come down, and by how much, based on the expected volume drop of the keywords that carry their spend. Keep each cut within the budget guardrails, and suggest pausing instead of any cut deeper than 80%.

Show every change as a preview, grouped by platform: keyword pauses and bid changes first, then budgets. Do not apply anything until I confirm, and apply only what I approve.

Strategy

TikTok creative fatigue check: pause, refresh or iterate

A per-ad fatigue verdict based on weekly CTR and CPA decay, plus a brief for the next video concepts, with pauses held for approval.

Check creative fatigue in [TikTok ad account] for the last 21 days, broken down by ad and by week.

Include every active ad with at least [minimum spend] in spend over the period. For each ad, show weekly spend, impressions, CTR, conversion rate, CPA, and the change from week 1 to week 3.

Classify each ad:
- Fatigued: CTR is down more than 30% from its best week and CPA is up more than 25%. Recommend pausing it.
- Declining: only one of those two thresholds is hit. Keep it running and recommend testing a new hook in the first 2 seconds.
- Healthy: neither threshold is hit. If its CPA is at least 20% below the ad group average, flag it as a candidate for new variations.

Then count, per ad group, how many healthy ads would remain after the proposed pauses. If fewer than two would remain, say the ad group needs new creative before anything is paused there.

Deliver:
1. The table.
2. A short creative brief for the next three video concepts, based on what the healthy ads have in common (hook, format, length, offer).

Prepare the pause actions as a preview and wait for my approval.

Strategy

LinkedIn lead quality problem: pivot targeting, adjust bid or pause

A data-backed decision on a weak LinkedIn campaign, comparing lead fit, audience size and bid range before proposing any change.

[LinkedIn campaign] has been running for at least 30 days, and the client says the leads are poor. Help me decide whether to pivot the targeting, change the bid, or pause.

Steps:
1. Pull the last 30 days: spend, impressions, CTR, leads and cost per lead. Add a breakdown by job function and seniority if it is available.
2. Download the leads and assess fit: what share matches [ideal job titles] and the target company size. Then calculate cost per qualified lead.
3. Run an audience forecast for the current targeting, and get the suggested bid range for that audience.

Decision rule:
- Pivot targeting if fewer than 40% of leads fit the ICP. Propose a narrower audience, forecast its size (it must stay above 50,000 members) and show its bid recommendation.
- Adjust the bid if leads fit but delivery is limited and our bid sits below the suggested range.
- Pause if fit is under 40% and either the narrower audience would drop below 20,000 members or cost per qualified lead exceeds [max cost per qualified lead].

Give me the recommendation with the supporting numbers. Show any targeting, bid or status change as a preview and wait for my approval.

Strategy

Find each channel's diminishing-returns point before moving budget

A per-channel table of weekly spend vs ROAS and marginal ROAS, with a scale / hold / cut verdict and a proposed budget split.

Strategy

Month-end pacing check across all paid channels

A projection of month-end spend per channel against the total cap, plus previewed daily budget changes that land on target while protecting ROAS.

Our total paid media cap this month is [monthly budget]. Check where we will land.

1. Get month-to-date spend and the projected month-end spend for each channel.
   - Google: use budget_pacing_alerts.
   - Meta, TikTok and Microsoft Ads: use their reports with the current daily budgets.
   - ChatGPT Ads: use the account spending limits.
2. Build one table with these columns: channel, MTD spend, MTD ROAS, projected month-end spend, share of the total cap, and over/under in currency.
3. If the projected total lands outside plus or minus 3% of the cap, propose daily budget changes for the remaining days that bring it back within range. Follow these rules:
   - Cut first from campaigns with MTD ROAS below [target ROAS].
   - Add first to campaigns above target that are spending their full budget.
   - Keep each daily budget change at 25% or less.

Show each change as a preview with the current and new daily budget and the projected effect on the total. Do not apply anything until I confirm.

Strategy

Rebalance prospecting vs retargeting spend on Google and Meta

The current prospecting vs retargeting spend split with ROAS and frequency per bucket, plus a recommended target split with previewed changes.

Retargeting ROAS usually looks great because it harvests buyers that prospecting created. Check whether we are overfunding it.

Last 30 days:
1. Meta: classify each ad set as retargeting or prospecting.
   - Retargeting: targets website, engagement or customer list custom audiences.
   - Prospecting: everything else, including lookalikes and broad targeting.
   - Report spend, ROAS, frequency and CPM per bucket.
2. Google Ads: classify campaigns and ad groups the same way, based on attached remarketing or Customer Match lists. Report spend and ROAS per bucket.

Deliverable:
- A table per platform with these columns: bucket, spend share, ROAS, and frequency (Meta only).
- Decision rules:
  - If retargeting takes more than [max retargeting share] of spend, or Meta retargeting frequency is above 4 over 7 days, retargeting is saturated. Shift budget to the prospecting ad sets with the best ROAS.
  - If prospecting ROAS is below 60% of [target ROAS], keep the current split. Flag the creative or audiences as the problem instead of the budget.
- Specific budget moves per ad set or campaign.

Preview every change and wait for my approval.

Strategy

Map competitors' live ad offers and find your positioning gap

Pull local competitors' live Google and Meta ads, map their offers, urgency claims, trust signals and CTAs, compare them with your own RSA copy and find the claims you can own.

Use the public ad libraries to research how my local competitors advertise. Competitors: [3-5 competitor domains or brand names]. Service area: [city or district].

For each competitor, run competitor_ads on platform google and on platform meta (Meta needs my connected Meta account), with region set to my market. Optionally, if I have SEO add-on credits, check seo_history and then run seo_paid_keywords for each competitor domain to get the headlines and descriptions they run on Google Search and the landing pages they send clicks to.

Give me one table with these columns:
- platform
- live ads returned (say if the result hit the limit, so the count is a minimum)
- start date of the oldest active ad, only if the library returns dates; otherwise write "not shown"
- main offer (free quote, discount, fixed price, free inspection)
- urgency claims (24/7, same day, "arrives in X minutes")
- trust signals stated in the ad copy (years in business, licences, review counts, guarantees). Only record what the ads actually say; do not look up or estimate review counts
- call to action (call, form, WhatsApp, booking), from the CTA button or the copy

Leave a cell as "not in library data" rather than guessing.

Then read my active Google Ads responsive search ads: find them with list_entities (entity ads), read the copy with get_ad_details, and use rsa_asset_performance to see which of my headlines and descriptions actually earn clicks and conversions. Compare my claims with theirs. Finish with three short lists:
1. Claims every competitor makes. These are table stakes, so flag any I am missing.
2. Two or three claims nobody makes that I could credibly own, and the proof I would need to back each one up.
3. Competitor offers I should not match, and why (margin, capacity, or risk of attracting bad leads).

This is read-only research. Do not change anything in my account.

Strategy

Build a 12-month demand and competitor activity calendar

Combines 12-month search demand, competitor ad activity and your monthly Google Ads results into one planning calendar. It sets budget weights per month and flags when new copy and offers need to be ready.

Help me plan my budget around seasonality. This is a planning document. Do not change any budgets or campaigns.

0. Check seo_history first. If a trend or paid keyword run for these keywords or competitors can still be reused, reuse it.

1. Demand:
- Get the Google Trends interest for my 5 core service keywords in [country or region]: [core keywords].
- Then look up the geo id for [country or region] and pull Keyword Planner historical metrics for the same keywords, covering the last 12 months with language [language], so I also see real monthly search volumes.
- For each month, build a demand index where my peak month = 100. Base it on the combined monthly volumes, and use Trends as a cross-check. Remember that Trends gives relative interest only.

2. Competitors: pull the live ads of [competitor brand names or domains] from the Google Ads Transparency Center for region [country code].
- If the results include first-shown or start dates, use them to show which months each competitor launches seasonal campaigns or promotions, and what they offer.
- If dates are missing or a competitor has no data, which is common for small advertisers, write "no timing data". Don't invent it.
- Then run competitor paid search for each competitor domain to see their current ad headlines, offers and landing pages. Treat this as a snapshot of today, not a history.

3. My performance: run a GAQL query on my Google Ads account for the last 12 full months, segmented by month (segments.month). Pull cost, conversions and cost per conversion, and calculate my annual average cost per conversion.

Set budget weights with these rules:
- Raise the weight in months where demand is above 80 and my cost per conversion stays within 20% of my annual average.
- Lower it in months where demand is below 50 or my cost per conversion is more than 40% above average.
- Otherwise keep it neutral. Show the weights so they add up to 100% across the year.

Also list:
- The 2-3 months where I should have new ad copy and offers ready, based on the demand peaks and when demand starts climbing.
- How many weeks before each peak competitors usually start. Only give this where start dates were returned, and label it an estimate. Where no dates exist, use the month demand starts rising as the lead-time signal instead.

Do not change any budgets or campaigns.

Strategy

Which offer actually sells on Meta: hook vs offer vs page

Meta ads grouped by offer type with a verdict on whether weak results come from the creative hook, the offer itself or the landing page.

Analyse which offers work in Meta ad account [ad account] over the last 60 days.

1. List all ads that spent at least [minimum spend].
2. Read each creative with get_meta_ad_creative. Tag its offer as one of: percentage discount, free shipping, free trial or demo, bundle or gift, or no explicit offer.
3. Pull these metrics per ad with meta_insights_query: spend, impressions, link CTR, landing page views, results and cost per result.
4. Aggregate the metrics by offer type.

Apply these decision rules per offer type, using [target cost per result] as the target:
- Link CTR below the account median: the hook or creative is the problem.
- Link CTR above the median but cost per result above target: the offer or the landing page is the problem.
- Landing page views below 70% of link clicks: page speed is the problem.

Deliverables:
1. An offer-type table with the metrics.
2. A one-line verdict per offer type.
3. Two next tests, each with a clear hypothesis and success metric.

Do not pause or edit anything. If you recommend pausing ads, list them with their numbers and wait for my approval. Once I approve, show a preview of each change before you apply it.

Strategy

Lead quality audit across Google, Meta and LinkedIn forms

Valid lead rate and cost per valid lead by platform and campaign, with form changes that would raise lead quality.

Management is questioning lead quality. Audit the in-platform lead forms for the last 30 days.

1. Pull the leads from three sources:
   - Google Ads lead forms with list_lead_form_submissions.
   - Meta instant forms with download_meta_leads.
   - LinkedIn Lead Gen Forms with download_linkedin_leads.
2. Mark a lead as invalid if any of these apply:
   - The phone number has the wrong digit count or obvious placeholder digits.
   - The email is fake or a test entry.
   - The name is meaningless.
   - It duplicates another lead, within or across platforms.
   - It gives a disqualifying answer to the qualifying question.
3. Pull spend per campaign from get_report, meta_campaign_insights and linkedin_campaign_analytics.

For each platform and campaign, calculate:
- Total leads
- Valid lead rate
- Cost per lead
- Cost per valid lead

Flag every campaign with a valid lead rate below 60%.

Deliverables:
1. A comparison table ranked by cost per valid lead.
2. The three main junk patterns you found.
3. Suggested form changes, such as a higher-intent qualifying question or a required field. Write these as recommendations only.

Report counts only. Do not show names, phone numbers or emails in the output. Do not change any form or campaign.

Strategy

Verify conversion tracking before presenting funnel numbers

A pass/fail measurement checklist across Google Ads, GA4 and the site tags, showing how errors distort reported CPA.

Before I present funnel results to management, check that the conversions in [Google Ads account] and GA4 can be trusted. Use the last 30 days.

Run these checks:
1. Find primary conversion actions that are micro steps (page views, button clicks, scrolls) and inflate the Conversions column. Use list_conversion_actions and list_conversion_goals.
2. Find lead actions set to count every conversion instead of one, and any unusual lookback windows. Use get_conversion_settings.
3. Find actions that are active but recorded zero conversions in the period. Use conversion_tracking_health.
4. Look for duplicate GA4 or Ads tags firing on [website]. Use wordpress_tracking and list_gtm_tags.
5. Compare Ads conversions with GA4 key events for the same event. Flag any gap above 25%.

Deliverables:
1. A pass/fail checklist with the evidence for each check.
2. An estimate of what CPA would be with only real business conversions counted as primary.
3. A prioritised fix list.

If a fix requires changing a conversion action, for example moving it to secondary, show me a preview of the exact change and wait for my approval before applying it.

These prompts work best when your AI assistant is connected to your accounts with Opus Growth.

Connect your accounts free