---
title: "Google and Meta Waste: One AI Report, Without the Double Count"
description: "A $500 sale can appear as $1,000 across Google and Meta. Build one waste report through groas, with clear thresholds and changes you approve."
url: "https://groas.com/post/wasted-spend-across-google-and-meta-in-one-report-from-your"
image: "https://pub-87da24ecbbfc4c3bad6875f3aa013712.r2.dev/generated-images/673b3ed4-7225-4b58-aa0b-5efac1129d86.png"
published: "2026-11-19T16:03:24.379Z"
modified: "2026-11-19T16:03:24.436Z"
---

November 19, 2026 · 10 min read

# Google and Meta Waste: One AI Report, Without the Double Count

[](https://groas.com/author/alexander-perelman)

[Alexander Perelman](https://groas.com/author/alexander-perelman)Personally Tested

Head Of Product @ groas

[LinkedIn](https://www.linkedin.com/in/alexander-433793253/)

![Cartoon of two anglers in separate boats proudly posing with the same single fish hooked on both their lines, each claiming the one catch as their own.](https://pub-87da24ecbbfc4c3bad6875f3aa013712.r2.dev/generated-images/673b3ed4-7225-4b58-aa0b-5efac1129d86.png)

In this article

1. [Define waste before asking AI to find it](#define-waste-before-asking-ai-to-find-it)
2. [Pull comparable inputs, not matching-looking columns](#pull-comparable-inputs-not-matching-looking-columns)
3. [The prompt: make the assistant show its rules](#the-prompt-make-the-assistant-show-its-rules)
4. [Worked example: the table supports less than the sales pitch](#worked-example-the-table-supports-less-than-the-sales-pitch)
5. [Find the gap between creating demand and claiming it](#find-the-gap-between-creating-demand-and-claiming-it)
6. [Turn findings into approved changes](#turn-findings-into-approved-changes)
7. [Keep the reason beside the change](#keep-the-reason-beside-the-change)
8. [Know where the evidence stops—and make the smaller decision](#know-where-the-evidence-stopsand-make-the-smaller-decision)

**One $500 order can look like $1,000 in platform revenue.** That is the gap I want a combined Google and Meta waste report to expose before anyone moves another dollar.

Consider a shopper who sees a Meta ad, searches the brand, and buys. If the purchase falls within both platforms’ attribution windows, Meta can claim a view-through conversion and Google can claim the search click. The order is real. Adding the claims together does not create another order.

[Summed platform revenue can run 20–30% above Shopify or CRM actuals](https://www.revenuedrivenmarketing.com/blog/why-meta-and-google-ads-numbers-dont-match/). That reported range is context, not a correction factor I would apply to every account. It also does not mean the same percentage of spend is wasted.

The useful question is what happens between the dashboards: which campaigns create demand, which close it, and which keep spending without a defensible return. I use one assistant reading both accounts through groas to build one waste table, with thresholds I can check. Here is the method, the prompt, and the decision the numbers support.

## Define waste before asking AI to find it

**Waste is spend past an agreed threshold without the required return, after allowing for conversion lag.** It is not simply a row I dislike.

I used to tell clients to pull the last month and kill anything with zero conversions. I was wrong. That rule punishes longer sales cycles and rewards branded search for closing demand another campaign created.

The platforms also use different attribution rules. [Meta’s click and view windows differ from Google’s attribution settings](https://diwizi.com/blog/cross-channel-attribution-google-meta-organic.html), so placing their conversion columns beside each other does not make them comparable. I first state the purchase event, target CPA, reporting windows, and revenue source.

Then I apply platform-specific screening rules:

| Screen                     | Rule                                                                      | What it means                             |
| -------------------------- | ------------------------------------------------------------------------- | ----------------------------------------- |
| Google search-term waste   | Cost exceeds target CPA, with zero conversions after the lag exclusion    | Candidate for a negative keyword          |
| Google watch list          | Cost approaches target CPA, with zero conversions and repeated clicks     | Monitor rather than automatically exclude |
| Meta zero-conversion waste | Spend exceeds twice the ad set’s CPA target, with zero conversions        | Candidate for a pause                     |
| Meta high-CPA test         | At least three conversions, with CPA above three times the ad set average | Review as a potential loser               |

The [Google search-term method uses a 90-day window ending 7–14 days ago](https://ppcleaner.com/blog/google-terms-report-filters-that-surface-waste). That exclusion matters because conversions can arrive after the click and be credited back to its date. Recent spend can look worse before its conversions arrive.

The [Meta method supplies the twice-target zero-conversion threshold and the three-conversion, three-times-average test](https://www.gomarble.ai/ai-tools/how-to/identify-wasted-meta-ads-spend/). These are operating rules, not proof that every flagged row is useless. I still check relevance, tracking, and campaign context before acting.

The threshold earns a row a review. It does not earn the assistant permission to press every red button.

## Pull comparable inputs, not matching-looking columns

Through groas, I have one assistant read Google search terms and keyword costs alongside Meta spend by ad set and ad, placement costs, and frequency against ROAS. I connect the accounts once; [connecting Meta to ChatGPT without the CSV shuffle](https://groas.com/post/how-to-connect-facebook-ads-manager-to-chatgpt-without) covers that side of the workflow.

Then I stop being the join key. What the deck calls “cross-channel synergy,” I call not exporting two CSVs and VLOOKUPing at midnight.

**One assistant can assemble the evidence; it cannot make incompatible attribution rules disappear.** The report must state its date ranges, exclusions, and conversion definitions before showing a total. I use the Google lag exclusion above and a Meta window that leaves out the most recent estimated days, as specified in the prompt below. Comparable does not mean pretending the windows are identical.

![The same shopper counted once by Meta and once by Google](https://pub-87da24ecbbfc4c3bad6875f3aa013712.r2.dev/generated-images/d4ecb87e-3c59-4fc5-a71c-77efc62a968d.png)

For Meta, I inspect the report in this order:

1. **Budget concentration:** spend share versus ROAS share by ad set.
2. **Purchase-cost outliers:** unusually expensive results relative to the account.
3. **Placements:** spend and cost per result, with zero-result slices shown separately.
4. **Fatigue:** rising frequency alongside falling ROAS, or falling link CTR alongside rising CPM.

That sequence follows the [Meta audit checklist’s budget, cost, placement, and fatigue checks](https://www.fixads.it/blog/meta-ads-audit-checklist). Audience Network rewarded video belongs in the placement review, not on an automatic blacklist.

A default ROAS sort will not answer all those questions. Ask for the checks explicitly.

## The prompt: make the assistant show its rules

I want the assistant to state its math before recommending an action. **A report without visible thresholds is an opinion wearing a spreadsheet.**

The figures below are illustrative inputs, not universal settings. The Google window follows the [search-term filtering method](https://ppcleaner.com/blog/google-terms-report-filters-that-surface-waste); the Meta thresholds follow the [zero-conversion and high-CPA method](https://www.gomarble.ai/ai-tools/how-to/identify-wasted-meta-ads-spend/). I replace the targets with the account’s actual economics before running it.

> Build one waste-review table across Google Ads and Meta through groas.
>
> Use the last 90 days ending 14 days ago for Google search terms. For Meta, use a stated window of 30–90 days, excluding the last 3 days. Explain the chosen windows and flag any longer conversion lag that makes them unsuitable.
>
> For this example, use a $60 Google target CPA. Flag search terms above $60 with zero conversions; watch-list terms spending $30–$60 with zero conversions and at least 3 clicks. Keep search-term detail beneath campaign and ad-group totals.
>
> Use each Meta ad set’s stated CPA target. Flag zero-conversion ad sets only after spend exceeds twice that target. Keep high-CPA tests separate from zero-conversion waste.
>
> Show platform, campaign, ad group or ad set, spend, conversions, target, threshold calculation, verdict, and proposed action. Separate confirmed threshold breaches, watch-list rows, and insufficient data. Do not double-count overlapping rows in the total.
>
> Compare platform-reported revenue with CRM or backend revenue separately. Identify possible attribution overlap without presenting it as proven wasted spend.
>
> Propose changes for approval. Do not execute them.

The [hands-on Google version of this workflow](https://groas.com/post/find-wasted-google-ads-spend-in-one-prompt-a-search-terms) covers the search-term half. This version adds Meta and the cross-platform check without quietly adding their revenue columns together.

## Worked example: the table supports less than the sales pitch

Here is an illustrative report, not a client result. The Google example uses the target in the prompt; the Meta examples use the [published $900-target scenarios](https://www.gomarble.ai/ai-tools/how-to/identify-wasted-meta-ads-spend/). The placement row is a review example, not an additional saving.

| Platform / slice                     | Spend in window              | Conversions   | Threshold calculation                     | Verdict                                           |
| ------------------------------------ | ---------------------------- | ------------- | ----------------------------------------- | ------------------------------------------------- |
| Google search terms                  | $480 across 6 terms          | 0             | Each term exceeds the $60 target          | Review relevance; add confirmed junk as negatives |
| Google watch-list terms              | $30–$60 each, with 3+ clicks | 0             | Not above the $60 target                  | Watch                                             |
| Meta retargeting                     | $4,800                       | 0             | Above twice the $900 target               | Pause candidate after checks                      |
| Meta broad prospecting               | $2,100                       | 2; CPA $1,050 | Too few conversions for the high-CPA test | Hold; insufficient evidence                       |
| Meta Audience Network rewarded video | $640                         | 0             | No finite cost per conversion to compare  | Investigate; do not automatically exclude         |

**The defensible action set is the $480 in confirmed junk terms and the $4,800 zero-conversion ad set.** The prospecting row does not become waste because its CPA looks uncomfortable. The placement row needs investigation, and it may overlap an ad set already shown. Adding it again would inflate the total.

Those two action rows contain $5,280 of spend in their respective reporting windows. That is not automatically $5,280 saved next month: the windows differ, and historical spend is not a forecast. I ask for an expected monthly saving separately, with the calculation shown.

If the row cannot name its threshold, it does not get to touch budget.

![A miniature control room with Google and Meta levers feeding one cash shredder](https://pub-87da24ecbbfc4c3bad6875f3aa013712.r2.dev/generated-images/2d460afd-c9c1-4cd6-9313-6392c5189d26.png)

## Find the gap between creating demand and claiming it

The combined table matters because a campaign can look efficient while depending on spending elsewhere. I look for these patterns:

- **Overfunded closers.** Branded search and retargeting can show attractive CPAs while harvesting demand created elsewhere. Cutting prospecting to feed those closers can undermine the system that makes them look good.
- **Prospecting feeding retargeting.** A weak-looking prospecting row may refill the audience that retargeting converts. The [cross-channel attribution discussion describes retargeting CPA rising after prospecting cuts](https://www.revenuedrivenmarketing.com/blog/why-meta-and-google-ads-numbers-dont-match/). I treat that dependency as something to examine, not a guaranteed outcome.
- **Overlapping credit.** A Meta view and a Google brand click can both receive credit for the same purchase. That proves overlapping claims, not that either touch was unnecessary.

**Double-counted revenue is not, by itself, evidence of double-paid waste.** This distinction keeps an attribution cleanup from turning into an indiscriminate budget cut.

I use [consistent UTMs, a neutral reporting layer such as GA4, and CRM or backend revenue as the sales ground truth](https://diwizi.com/blog/cross-channel-attribution-google-meta-organic.html). Platform revenue stays visible, but I do not sum it and call the result business revenue.

Branded search taking a large share of Google spend while Meta prospecting frequency rises and ROAS falls deserves investigation. It does not establish which campaign caused the problem. Where appropriate, [feeding CRM outcomes back through Google enhanced conversions and Meta Conversions API](https://www.revenuedrivenmarketing.com/blog/why-meta-and-google-ads-numbers-dont-match/) helps both platforms optimize against those outcomes. It does not make their attribution claims additive.

I judge closers on their incremental contribution and creators on new-customer cost. The combined report points to the question; it does not manufacture the answer.

## Turn findings into approved changes

A waste table that never produces an action is a second dashboard. **Every proposed change needs a reason, a scope, and approval.**

I turn the reviewed rows into these moves:

1. **Exclude confirmed irrelevant Google queries.** Propose exact-match negatives at the appropriate scope. A term that fails in one campaign does not automatically belong on an account-wide exclusion list.
2. **Pause qualifying Meta ad sets.** Check the conversion event, lag, and threshold before pausing a zero-conversion row.
3. **Propose budget moves separately.** Favor prospecting with the lowest supported new-customer cost, not retargeting simply because its reported ROAS is prettier.

For each move, I ask the assistant for the exact proposed command, the expected monthly saving with its calculation, and a review date. Nothing runs on autopilot for this first pass.

The approval line matters more than the automation.

![Paper-cut control panel with pause, negative keyword, and budget-shift actions awaiting approval](https://pub-87da24ecbbfc4c3bad6875f3aa013712.r2.dev/generated-images/7097bc28-db96-464e-9c09-2aaba38e4ea6.png)

## Keep the reason beside the change

Every proposed fix records the number observed, the threshold applied, and who approved it. I also require `reason`, `approved_by`, and `review_on` in the action log.

**A reversible change still needs an explanation.** Native platform histories and undo options have different limits; [this change-log overview describes the differences between Google and Meta histories and reversal support](https://marketsplash.com/ads-agent-change-log/). I do not want the rationale scattered across those interfaces.

If a paused ad set turns out to support demand elsewhere, I want to find the decision and reverse it without a forensic dig through two platforms. Logging the action is useful. Logging why it happened is what makes the next decision better.

## Know where the evidence stops—and make the smaller decision

**One table is not complete visibility.** Google hides some low-volume search terms for privacy, so visible query spend will not necessarily reconcile to campaign spend. That is a reporting limit, not permission to label the difference waste.

The [search-term reporting limitations also affect campaign coverage](https://ppcleaner.com/blog/google-ads-search-terms-report-filters-that-surface-waste): grouped insights are not interchangeable with raw queries, and Display and Video do not supply the same search-term rows. Meta audience overlap is not exposed through the API in this workflow. Similar targeting can suggest overlap; it cannot prove the same buyer was unnecessarily reached twice.

Small samples are the other trap. A couple of conversions or an ad set still in learning is not enough to justify a confident loser label. I mark insufficient evidence, consider consolidation where appropriate, and avoid turning uncertainty into an automatic pause.

For low-spend accounts, this may be more reporting machinery than the decision needs. If tracking is broken, I fix that first. No prompt makes missing purchases reappear.

The worked example supports a narrow decision: exclude the confirmed irrelevant Google terms, pause the qualifying zero-conversion Meta ad set after checks, and hold the prospecting row with insufficient evidence. Any budget transfer needs its own case. None of those decisions proves that every remaining campaign contributes to the funnel.

I run the combined report monthly through groas, with explicit thresholds and CRM revenue kept separate from platform claims. The point is not a bigger waste number. It is a smaller set of changes I can defend.

Separate reports leave the gap between platforms unexamined. One report makes it reviewable. **Before buying more ads on the strength of summed Google and Meta revenue, reconcile the sales—and act only on the waste the evidence supports.**

## Frequently Asked Questions

### Why does my Google and Meta revenue total more than my actual Shopify sales?

The same purchase can be claimed by both platforms. A shopper who sees a Meta ad, searches the brand, and buys may fall inside both attribution windows, so Meta claims a view-through conversion and Google claims the search click. Adding their claims together double-counts real orders, which is why summed platform revenue can run 20-30% above Shopify or CRM actuals.

### What counts as wasted ad spend?

Waste is spend past an agreed threshold without the required return, after allowing for conversion lag. It is not simply any campaign row you dislike. Killing anything with zero conversions in the last month is too crude a rule, because it punishes longer sales cycles and rewards branded search that closes demand other campaigns created.

### Can I compare Meta and Google conversion columns side by side?

Not directly, because the platforms use different attribution rules. Meta's click and view windows differ from Google's attribution settings, so placing their conversion columns beside each other does not make them comparable. A combined report should state its date ranges, exclusions, and conversion definitions before showing any total.

### What thresholds should I use to flag wasted spend in Google and Meta?

For Google search terms, flag anything costing more than your target CPA with zero conversions, and watch-list terms approaching the target with repeated clicks. For Meta, flag zero-conversion ad sets only after spend exceeds twice the ad set's CPA target, and review tests with at least three conversions but a CPA above three times the ad set average. The threshold earns a row a review, not an automatic action.

### Should an AI assistant automatically pause my ad campaigns?

No. The assistant should state its math, including thresholds and calculations, before recommending an action, and propose changes for approval rather than executing them. For each move, ask for the exact proposed command, the expected monthly saving with its calculation shown, and a review date. The approval line matters more than the automation.

### Can cutting prospecting to fund branded search hurt my results?

Yes, it can. Branded search and retargeting can show attractive CPAs while harvesting demand that prospecting created, so cutting prospecting to feed those closers can undermine the system that makes them look good. Retargeting CPA has been observed to rise after prospecting cuts. Judge closers on their incremental contribution and creators on new-customer cost.

### Why should I log the reason behind every ad account change?

Logging the action is useful, but logging why it happened is what makes the next decision better. Native platform histories and undo options have different limits between Google and Meta, so the rationale should not be scattered across those interfaces. Each entry should record the number observed, the threshold applied, and who approved it, along with reason, approved_by, and review_on fields.

### Does the search terms report show every query people typed?

No. Google hides some low-volume search terms for privacy, so visible query spend will not necessarily reconcile to campaign spend, and grouped insights are not interchangeable with raw queries. Display and Video also do not supply the same search-term rows. That is a reporting limit, not permission to label the difference waste.

## Run Your Marketing On groas

Plug ChatGPT, Claude or any AI agent into groas and let it do the work of a whole marketing team: $19 a month flat with every feature included, and autopilot whenever you’re ready.

[Start For $19](https://groas.com/pricing)

[Or have groas run it for you](https://groas.com/done-for-you)

## Structured data

```json
{"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://graph.groas.com/entity/groas","name":"groas","url":"https://groas.com/","description":"groas is the marketing infrastructure for your own AI: it connects your AI to all your marketing and gives it proprietary marketing data collected 24/7. Plug Claude, ChatGPT or any AI agent into groas and run all your marketing from it, for a tiny fraction of the usual cost, with the groas app there whenever you want to see more. It is the layer ChatGPT, Claude and any AI assistant run a business’s online marketing on. Through groas an assistant can do anything you could do natively in Google Ads, Bing Ads, Meta Ads and ChatGPT ads, on your website and in AI search, and, where the owner switches it on, can run connected campaigns by itself, so AI agents do the work a whole marketing team used to. groas also gives those agents large-scale proprietary marketing data it collects around the clock, on any schedule the owner chooses: every ad competitors run on Google and Meta, their landing pages and full funnels with the offers and prices in them, Reddit customer research, AI search answers from ChatGPT, Google AI Overviews, Gemini, Claude and Perplexity, and SEO data from Semrush and Ahrefs, backlinks included ($19 a month, flat, with everything included, for unlimited teammates on one shared workspace and unlimited ad accounts and websites, delivered as an MCP server). Separately, its done-for-you service has senior people and proprietary quantitative systems run paid ads, SEO and AI search for businesses, and white label for agencies, from $5,000 a month.","logo":{"@type":"ImageObject","url":"https://groas.com/icon-512.png","width":512,"height":512},"sameAs":["https://www.linkedin.com/company/groas/"]},{"@type":"WebSite","@id":"https://groas.com/#website","name":"groas","url":"https://groas.com/","inLanguage":"en-US","publisher":{"@id":"https://graph.groas.com/entity/groas"}},{"@type":"WebPage","@id":"https://groas.com/post/wasted-spend-across-google-and-meta-in-one-report-from-your#webpage","url":"https://groas.com/post/wasted-spend-across-google-and-meta-in-one-report-from-your","name":"Google and Meta Waste: One AI Report, Without the Double Count","description":"A $500 sale can appear as $1,000 across Google and Meta. Build one waste report through groas, with clear thresholds and changes you approve.","inLanguage":"en-US","isPartOf":{"@id":"https://groas.com/#website"},"about":{"@id":"https://graph.groas.com/entity/groas"},"primaryImageOfPage":{"@type":"ImageObject","url":"https://pub-87da24ecbbfc4c3bad6875f3aa013712.r2.dev/generated-images/673b3ed4-7225-4b58-aa0b-5efac1129d86.png"},"breadcrumb":{"@id":"https://groas.com/post/wasted-spend-across-google-and-meta-in-one-report-from-your#breadcrumb"},"dateModified":"2026-11-19T16:03:24.436Z","mainEntity":{"@id":"https://groas.com/post/wasted-spend-across-google-and-meta-in-one-report-from-your#article"}},{"@type":"BreadcrumbList","@id":"https://groas.com/post/wasted-spend-across-google-and-meta-in-one-report-from-your#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://groas.com/"},{"@type":"ListItem","position":2,"name":"Blog","item":"https://groas.com/blog"},{"@type":"ListItem","position":3,"name":"Google and Meta Waste: One AI Report, Without the Double Count","item":"https://groas.com/post/wasted-spend-across-google-and-meta-in-one-report-from-your"}]},{"@type":"BlogPosting","@id":"https://groas.com/post/wasted-spend-across-google-and-meta-in-one-report-from-your#article","url":"https://groas.com/post/wasted-spend-across-google-and-meta-in-one-report-from-your","mainEntityOfPage":{"@id":"https://groas.com/post/wasted-spend-across-google-and-meta-in-one-report-from-your#webpage"},"isPartOf":{"@id":"https://groas.com/blog#blog"},"headline":"Google and Meta Waste: One AI Report, Without the Double Count","description":"A $500 sale can appear as $1,000 across Google and Meta. Build one waste report through groas, with clear thresholds and changes you approve.","datePublished":"2026-11-19T16:03:24.379Z","dateModified":"2026-11-19T16:03:24.436Z","image":{"@type":"ImageObject","url":"https://pub-87da24ecbbfc4c3bad6875f3aa013712.r2.dev/generated-images/673b3ed4-7225-4b58-aa0b-5efac1129d86.png"},"author":{"@id":"https://groas.com/author/alexander-perelman#person"},"publisher":{"@id":"https://graph.groas.com/entity/groas"},"keywords":"meta, google, waste, report, spend, conversions, without, target","wordCount":2039,"inLanguage":"en-US"},{"@type":"Person","@id":"https://groas.com/author/alexander-perelman#person","name":"Alexander Perelman","jobTitle":"Head Of Product @ groas","description":"My background includes Goldman Sachs and Stanford Computer Science.","image":"https://groas.com/media/blog/4f17cac3a81acc224cc1b2cfab8f94bbc295a3089d13e8421125771ef8d4064f.jpg","worksFor":{"@id":"https://graph.groas.com/entity/groas"},"sameAs":["https://www.linkedin.com/in/alexander-433793253/"]},{"@type":"FAQPage","@id":"https://groas.com/post/wasted-spend-across-google-and-meta-in-one-report-from-your#faq","mainEntity":[{"@type":"Question","name":"Why does my Google and Meta revenue total more than my actual Shopify sales?","acceptedAnswer":{"@type":"Answer","text":"The same purchase can be claimed by both platforms. A shopper who sees a Meta ad, searches the brand, and buys may fall inside both attribution windows, so Meta claims a view-through conversion and Google claims the search click. Adding their claims together double-counts real orders, which is why summed platform revenue can run 20-30% above Shopify or CRM actuals."}},{"@type":"Question","name":"What counts as wasted ad spend?","acceptedAnswer":{"@type":"Answer","text":"Waste is spend past an agreed threshold without the required return, after allowing for conversion lag. It is not simply any campaign row you dislike. Killing anything with zero conversions in the last month is too crude a rule, because it punishes longer sales cycles and rewards branded search that closes demand other campaigns created."}},{"@type":"Question","name":"Can I compare Meta and Google conversion columns side by side?","acceptedAnswer":{"@type":"Answer","text":"Not directly, because the platforms use different attribution rules. Meta's click and view windows differ from Google's attribution settings, so placing their conversion columns beside each other does not make them comparable. A combined report should state its date ranges, exclusions, and conversion definitions before showing any total."}},{"@type":"Question","name":"What thresholds should I use to flag wasted spend in Google and Meta?","acceptedAnswer":{"@type":"Answer","text":"For Google search terms, flag anything costing more than your target CPA with zero conversions, and watch-list terms approaching the target with repeated clicks. For Meta, flag zero-conversion ad sets only after spend exceeds twice the ad set's CPA target, and review tests with at least three conversions but a CPA above three times the ad set average. The threshold earns a row a review, not an automatic action."}},{"@type":"Question","name":"Should an AI assistant automatically pause my ad campaigns?","acceptedAnswer":{"@type":"Answer","text":"No. The assistant should state its math, including thresholds and calculations, before recommending an action, and propose changes for approval rather than executing them. For each move, ask for the exact proposed command, the expected monthly saving with its calculation shown, and a review date. The approval line matters more than the automation."}},{"@type":"Question","name":"Can cutting prospecting to fund branded search hurt my results?","acceptedAnswer":{"@type":"Answer","text":"Yes, it can. Branded search and retargeting can show attractive CPAs while harvesting demand that prospecting created, so cutting prospecting to feed those closers can undermine the system that makes them look good. Retargeting CPA has been observed to rise after prospecting cuts. Judge closers on their incremental contribution and creators on new-customer cost."}},{"@type":"Question","name":"Why should I log the reason behind every ad account change?","acceptedAnswer":{"@type":"Answer","text":"Logging the action is useful, but logging why it happened is what makes the next decision better. Native platform histories and undo options have different limits between Google and Meta, so the rationale should not be scattered across those interfaces. Each entry should record the number observed, the threshold applied, and who approved it, along with reason, approved_by, and review_on fields."}},{"@type":"Question","name":"Does the search terms report show every query people typed?","acceptedAnswer":{"@type":"Answer","text":"No. Google hides some low-volume search terms for privacy, so visible query spend will not necessarily reconcile to campaign spend, and grouped insights are not interchangeable with raw queries. Display and Video also do not supply the same search-term rows. That is a reporting limit, not permission to label the difference waste."}}]}]}
```
