How to Use Claude for Google Ads Analysis

Analyze this campaign data and identify all campaigns with ROAS below 2
Sort by spend from highest to lowest
For each campaign include: campaign name, total spend, conversions, revenue, ROAS, and conversion rate
At the end, calculate how much budget is being wasted on underperforming campaigns
Recommend which ones to pause immediately vs which ones to optimize

Compare performance this week vs last week for all campaigns
Create a table showing: campaign name, spend change %, conversion change %, CPA change %, ROAS change %
Flag any campaign where CPA increased more than 20% or ROAS dropped more than 15%
For flagged campaigns, suggest possible causes based on the data patterns

Which campaigns should I pause based on this data
Only include campaigns with at least $500 spend in the date range
Criteria for pausing: ROAS below 1, or CPA more than 2x account average, or zero conversions
For borderline cases, explain why you would or wouldn't pause
Calculate total budget I would save by pausing recommended campaigns

Find all keywords with CTR below 2% and at least 1000 impressions
For each keyword include: keyword, match type, impressions, clicks, CTR, conversions, cost
Group them by ad group so I can see patterns
Identify which keywords are dragging down ad group quality scores
Recommend action for each: pause, change match type, or improve ad relevance

Analyze the search terms report and find converting search terms that aren't exact match keywords yet
Sort by conversions, then by conversion value
For each search term show: the search term, matched keyword, match type, conversions, CPA, ROAS
Output a list formatted for Google Ads Editor bulk upload to add as exact match keywords
Include recommended starting bids based on current performance

Identify all wasted spend on search terms
Criteria: cost greater than $50 and zero conversions
Group these search terms by theme so I can create organized negative keyword lists
For each theme, write the negative keywords in phrase match and exact match format
Calculate total wasted spend and project monthly savings if I add these negatives

Compare headline performance across all responsive search ads
Analyze which headlines have the highest CTR and which have the highest conversion rate
Note: high CTR doesn't always mean high conversions
Create a ranking of top 10 best performing headlines and bottom 10 worst
Identify patterns in what makes headlines work: length, CTA type, numbers, questions, etc
Suggest 5 new headline variations based on the winning patterns

Find all ad groups where we only have 1-2 active ads running
List them with: ad group name, campaign name, number of active ads, ad group spend, ad group conversions
For each ad group, look at the existing ad copy and suggest 2-3 new ad variations to test
Make the new variations meaningfully different, not just word swaps
Prioritize ad groups by spend so I fix the highest impact ones first

Identify all campaigns that are limited by budget
Show: campaign name, daily budget, average daily spend, impression share, impression share lost to budget, estimated clicks lost per day
Calculate: if I increased budget to eliminate lost impression share, how many additional clicks and conversions could I expect
Factor in current conversion rate to project incremental conversions
Rank campaigns by opportunity size

Calculate the incremental CPA if I increase budget by 20% on the top 5 performing campaigns
Use the last 30 days conversion rate and factor in diminishing returns
Assume a 10% efficiency loss for each 20% budget increase as we reach less qualified traffic
Show the math: current spend, current conversions, current CPA, projected new spend, projected new conversions, projected new CPA
Only recommend the increase if projected CPA stays within 15% of current CPA

Break down performance by device type: desktop, mobile, tablet
For each device show: impressions, clicks, CTR, conversions, conversion rate, CPA, ROAS, spend share
Flag any device where CPA is 50% higher than account average or ROAS is 50% lower
Calculate how much I'm overspending on underperforming devices
Recommend bid adjustments as percentages for each device
If mobile is underperforming, note whether landing page speed could be a factor

Analyze performance by hour of day and day of week
Create a heatmap-style breakdown showing conversion rate and CPA for each hour
Identify the top 5 best performing hours and the 5 worst performing hours
For worst performing hours, calculate cost during those hours vs conversions
Recommend specific ad schedule bid adjustments or hour exclusions
Factor in that some hours may have low volume, so flag any recommendation based on less than 100 clicks