Performance reviews for Google Ads
Give your AI tool the reports you review each week. It says when a drop began and which campaign carries it, then judges keywords, Quality Score, ads, devices, locations, schedule, landing pages, audiences, placements and video against the account’s own numbers.
npx skills add parallel-ai-inc/google-ads-skills -gInstalls all 13 skills in Claude Code, Codex, Cursor and other tools. Steps for each tool
Google Ads performance review
The answer
Conversions fell 24% last week because Plumbing | Emergency’s conversion rate dropped from 2026‑11‑06 while its cost held, so check that campaign’s page and call tracking first. Across the month the clearest waste is the job seeker keyword “plumbing jobs” (640.00), Gary, Indiana outside the service area (1,000.00) and the hydro jetting page (520.00), and the water heater page is the weakest landing page.
Weekly performance review
Cost fell 7% but conversions fell 24%, so cost per conversion rose from 70.81 to 86.80. Plumbing | Emergency carries 10.70 of the 12.52 lost conversions: its conversion rate is the main lever, and its daily conversion rate shifted on 2026‑11‑06, from 9.6% to 4.7%. Drain Cleaning | Old is paused, so its drop is not a loss.
- Last 7 days, 2026‑11‑06 to 2026‑11‑12, against the 7 before, 2026‑10‑30 to 2026‑11‑05.
- Account: cost 3,425.03 (was 3,680.96, −7%). Conversions 39.46 (was 51.98, −24%). Cost / conv. 86.80 (was 70.81, +23%). Conv. rate 8.3% (was 10.2%). Clicks 477 (was 512). Cost per conversion against the week before: Worse, conversion rate.
- Without brand: cost 3,365.38 (was 3,618.96, −7%). Conversions 30.05 (was 42.46, −29%). Cost / conv. 111.99 (was 85.23, +31%). Conv. rate 6.9% (was 9.1%). Clicks 435 (was 469). Cost per conversion against the week before: Worse, conversion rate.
- Brand campaigns: cost 59.65 (was 62.00, −4%). Conversions 9.41 (was 9.52, −1%). Cost / conv. 6.34 (was 6.51, −3%). Conv. rate 22.4% (was 22.1%). Clicks 42 (was 43). Cost per conversion against the week before: In line.
What to give it
Any of the reports you review, as CSV. The weekly review needs the campaign report by day for at least 14 days, and 28 for the trend. The monthly reviews take 30 days of the keyword, ads, device, landing page, audience, location, ad schedule, placement and video reports, each from its own place in Google Ads.
The skill also asks for the brand campaigns and brand terms, so cheap brand conversions do not pull every baseline down, and what the business does not serve, such as places, services or hours.
What you get back
performance-review.xlsxThe weekly review and trend, a tab for each segment review with a verdict on every row, the actions, the rules used and the reviews not run.performance-review.mdThe summary. The answer, the weekly review, each segment review with what to do, and what the reports agree on across reviews.
In the chat, your AI tool answers what you asked first, then the drop and where it started, and the clearest waste across the month.
Every segment gets a verdict against the account’s own numbers
Each location, device, hour, page and audience is judged against the account’s own baseline, and only once it has spent enough to judge. Brand is set apart, so its cheap conversions do not flatter the baseline.
| Location | Cost | Conversions | Cost / conv. | Verdict | Wasted |
|---|---|---|---|---|---|
| Chicago, Illinois, United States | 7,000.00 | 74.00 | 94.59 | In line | |
| Evanston, Illinois, United States | 3,300.00 | 42.00 | 78.57 | Better | |
| Skokie, Illinois, United States | 3,300.00 | 34.00 | 97.06 | In line | |
| Gary, Indiana, United States | 1,000.00 | 0.00 | No conversions | 1,000.00 | |
| Wilmette, Illinois, United States | 200.00 | 1.00 | 200.00 | Not enough data to judge |
How it decides
Scripts compute every number, and your AI tool states no figure they did not print. Each rule it applies is labelled Official, from Google’s documentation, or Practitioner rule, from Parallel AI’s operating knowledge.
- A segment is judged only once it has spent enough to judge, and a wide gap on little volume is marked as worth watching, not as a verdict.Practitioner rule
- Money is never added across reviews, because the same click has a device, an hour and a location.Practitioner rule
- Quality Score is a diagnostic reported for Search keywords. It is not an input to the auction or a figure to average.Official
- A location outside the service area is excluded, with the location option set to people in or regularly in the targeted area.Practitioner rule
What it does not cover
- Why a drop happened. The performance diagnosis skill in the pack splits the change by lever and lines up the account changes before it.
- Targets and industry figures. Segments are judged against the account’s own numbers in the same file.
- Ad fatigue over time, the content of a page or a video, and policy detail. Those need the account itself or the asset.
- Conversions still arriving. Recent days can gain conversions reported late.
When you want it on live accounts
The skill reviews one account’s exports at a time. Parallel AI runs the weekly review on your live Google Ads accounts, every client at once, and makes the changes it points to in the account.
The page for this job in Parallel AI is Audit every Google Ads account you run, then fix what matters.
The guide behind this job is Google Ads clicks but no conversions, and what to check first.
Install
In Codex, Cursor, Gemini CLI, GitHub Copilot, Windsurf, OpenCode and other tools that read Agent Skills, the open skills CLI installs every skill in the pack.
npx skills add parallel-ai-inc/google-ads-skills -gIn Claude Code, add the pack as a plugin from a session.
/plugin marketplace add parallel-ai-inc/google-ads-skills/plugin install parallel-ai-for-google-ads@parallel-ai
To install this skill alone, name it.
npx skills add parallel-ai-inc/google-ads-skills --skill performance-reviews-for-google-ads -gAdd -a with a tool’s id, such as codex or cursor, to install for that tool alone. Without -g, the skills install into the current project.
Questions and answers
Does the skill cost anything?
No. The skills are free to use for any Google Ads work, for yourself or for clients. They may not be republished as your own skills. Parallel AI is a separate product, with a 5‑day free trial.
Does it change my Google Ads account?
No. It never connects to an account on anyone’s behalf and never changes one. It reads the reports you give it and writes files. Where a job ends in a change, it writes a Google Ads Editor file, and you check it there before posting.
Where does my data go?
The skill’s scripts make no network calls, and nothing goes to Parallel AI. Your AI tool sees the files you give it, as it would any file, under that tool’s own terms. If the export is a client’s, check your agreement with that client before you give it to an AI tool.
Do I need every report?
No. Each report unlocks its own review, and the skill runs the ones your files allow and lists the rest with the report that would unlock them.
Which AI tools can run it?
Tools that read Agent Skills and can run Python, such as Claude Code, Codex, Cursor, Gemini CLI, GitHub Copilot, Windsurf and OpenCode. The scripts use Python 3’s standard library only.
Is Northside Plumbing a real business?
No. Northside Plumbing is the pack’s made up sample account, with problems planted for each skill to find. Run the skill on the sample files and you get the same figures as this page.
Google Ads is a trademark of Google LLC. Parallel AI is not affiliated with or endorsed by Google.
Run the weekly review on every account
Parallel AI reads your live Google Ads accounts each week, finds where a drop began and makes the changes it points to in the account, across every client at once.
Starter $20, Pro $60 and Max $200 a month
You connect Google Ads first, then add a card to start the trial.