Switch Language
Toggle Theme

AdSense Data Analysis in Action: Complete Guide from Understanding Reports to Doubling Revenue

Last year, a friend who runs a tech blog complained to me: “I open my AdSense dashboard every day, glance at the revenue number, and close it. I don’t understand those CTR, RPM, CPC metrics. As long as money comes in, that’s all that matters.”

I didn’t say much at the time. Six months later, he came back, this time with obvious anxiety: “Something’s wrong. My traffic went from 2,000 to 5,000 visits per day. Logically, revenue should have doubled, right? But not only did it barely increase, sometimes it even dropped a bit. Is AdSense broken?”

I asked him to send me screenshots of his dashboard. After looking at the data for 5 minutes, the problems were right there:

  • The new traffic mainly came from Facebook shares (this batch of traffic was low quality)
  • Mobile traffic went from 50% to 80%, but mobile RPM was only $1.8 (desktop $4.2)
  • Three basic tutorial pages accounted for 50% of total traffic, with RPM all below $1

Honestly, AdSense had been telling him all this data for a while. He just never looked at it.

This reminds me of when I first started with AdSense. Revenue went from $10 to $15 per day, and I was happy for a week. Then suddenly one day it dropped to $8, and my heart sank—was my site penalized? Were my ads banned? All kinds of worries.

Later I discovered it was just a mobile ad unit that I had accidentally deleted. After restoring it, revenue immediately came back.

You see, the data was always there. It’s like a car’s dashboard—speed, fuel, temperature, every metric is telling you something. If you don’t look at it, it doesn’t mean problems don’t exist.

Essential AdSense Core Metrics

Revenue Formula and Four Core Metrics

First, let’s talk about the most basic thing—how is AdSense revenue actually calculated?

There are actually just two formulas:

Revenue = Impressions × RPM ÷ 1000

Or:

Revenue = Clicks × CPC

Understand these two formulas, and you’ll grasp the underlying logic of revenue.

Now let me explain these four core metrics one by one.

1. Impressions

This is the easiest to understand. When a user opens your page and the ad displays, that counts as one impression.

Many people think the more impressions the better. Actually, it’s not. I’ve seen sites with 100,000 daily impressions but only $50 monthly revenue (all junk traffic), and I’ve also seen premium sites with 5,000 daily impressions making $800 monthly.

Impressions are foundational, but quality is far more important than quantity.

2. CTR (Click-Through Rate)

Calculation formula: Clicks ÷ Impressions × 100%

This metric tells you: out of 100 people who see an ad, how many will click.

Industry benchmark data (compiled from my own observations, you can compare):

  • 0.5-1%: Low, possibly poor ad placement or colors that are too jarring
  • 1-3%: Normal level, most sites fall in this range
  • 3-5%: Pretty good
  • Above 5%: Either your ad placement is excellent, or there might be accidental clicks (like mobile buttons too close to ads)

My tech blog CTR consistently stays between 2.2-2.8%, which is above average.

3. CPC (Cost Per Click)

This is how much money you earn per click.

What determines CPC? Simply put, it’s how much advertisers are willing to pay.

Advertisers in finance, insurance, and SaaS have money and are willing to pay high prices. Entertainment, gaming, and gossip have less competition and lower profits, so CPC is naturally lower.

Here’s reference data:

  • $0.05-0.15: Low-value content (entertainment, gaming, gossip)
  • $0.15-0.50: Average level (lifestyle, food, travel)
  • $0.50-2.00: High-value content (programming, design, marketing)
  • $2.00+: Top tier (finance, investment, enterprise software)

My friend who runs a food blog has a CPC consistently around $0.18. I create tech content, and my CPC reaches $0.45-$0.60. That’s where the gap comes from.

4. RPM (Revenue Per Mille)

This is what I consider the most important metric.

Calculation formula: (Revenue ÷ Impressions) × 1000

RPM essentially quantifies the combined effect of CTR and CPC. It directly tells you: how much money can you earn from 1000 impressions.

$1-3
General Content
News, entertainment, gossip
$3-8
Medium Level
Tech, lifestyle, tutorials
$8-15
Excellent Performance
Niche domains, high-value content
$15+
Top Tier
Fintech, enterprise services

My own data: tech tutorial articles have RPM between $6-12, basic explainer content only $2-4.

Relationships Between Metrics

These metrics aren’t isolated; they’re interconnected.

Look at this relationship chain:

Impressions × CTR = Clicks
Clicks × CPC = Revenue
Revenue ÷ Impressions × 1000 = RPM

To increase RPM (which means increasing revenue), you have three paths:

  1. Increase CTR: Optimize ad placement, adjust ad colors to make them less jarring
  2. Increase CPC: Write high-value content, block low-price ad categories
  3. Improve traffic quality: SEO traffic RPM is typically 2-3 times that of social traffic

Let me show you a real comparison. These are two websites I’ve observed (both friends’):

Website TypeDaily ImpressionsCTRCPCRPMMonthly Revenue
A (Entertainment)100,0000.8%$0.12$0.96$960
B (Tech Blog)30,0002.5%$0.45$11.25$3,375

See that? Site B has only 30% of Site A’s traffic, but 3.5 times the revenue!

This is why I always say: traffic quality > traffic quantity.

Important Change in 2024: Pay Per Impression

If you’re an old webmaster, you might remember that AdSense used to be mainly pay-per-click (CPC model).

But starting in 2024, Google changed the main model to pay per impression (CPM model).

What does this mean?

Before: You only earned money when users clicked on ads. CTR was key.

Now: You earn revenue when users see ads (impressions). Even if CTR is low, as long as impressions are sufficient and traffic quality is high, you can still make money.

How this affects us:

  1. Impressions become more important—before you might pursue precise small traffic, now pageviews (PV) are also crucial
  2. RPM becomes the core metric—it comprehensively reflects impression quality and click performance
  3. Mobile traffic value increases—before, mobile had low CTR and low revenue; now with pay per impression, mobile can monetize well too

My own data comparison:

  • 2023 (old model): Mobile RPM $2.1, Desktop RPM $5.8
  • 2025 (new model): Mobile RPM $3.8, Desktop RPM $6.2

Mobile improved significantly.

Honestly, this change is beneficial for most webmasters. No need to obsess over “how to improve CTR” anymore, just focus on creating good content and improving traffic quality.

How to Discover Problems Through Data

Building a Data Monitoring Habit

Many people don’t not know they should look at data; they don’t know what to look at or how to look at it.

I spend 30 minutes weekly doing data checks, looking at 5 dimensions. Let me share with you.

Dimension 1: Time Dimension

Open AdSense dashboard → Reports → View by date

I look at:

  • Revenue trends for the last 7 days (any unusual fluctuations)
  • Compare to the same period last week (week-over-week)
  • Compare to the same period last month (month-over-month)

What counts as unusual?

  • Single-day revenue suddenly drops more than 20%
  • Revenue declines for 3 consecutive days
  • RPM suddenly drops significantly

Last November I encountered this. One day revenue suddenly dropped from $25 to $15. My first reaction was: Oh no, was my site penalized?

After calming down and looking at the data, impressions hadn’t changed, still around 5,000. But CTR dropped from 2.2% to 0.9%.

I kept investigating. Found that mobile CTR dropped from 2.1% to 0.8%.

Looked closer. Turns out a mobile in-feed ad had been accidentally deleted. After restoring it, revenue immediately returned to $25.

Dimension 2: Device Dimension

Reports → Platforms → Mobile devices vs Desktop

Key things to watch:

  • How big is the RPM gap between mobile and desktop
  • Is mobile CTR too low

Normal situation: Mobile RPM should be 50-70% of desktop RPM.

If your mobile RPM is below 50% of desktop, you need to optimize.

Previously my mobile RPM was only $2.3, desktop $8.5. Gap was too big. After optimizing ad placement and types, mobile RPM rose to $4.8. I’ll explain this case in detail later.

Dimension 3: Traffic Source Dimension

This needs to be viewed together with Google Analytics 4 (GA4).

AdSense only tells you “revenue data,” GA4 tells you “traffic sources.” Combining both helps you find root causes.

My experience:

  • SEO traffic: RPM usually highest, $4-10
  • Direct visits: RPM medium, $3-6 (returning users)
  • Social media: RPM lowest, $0.8-2 (short user dwell time)

Once my friend’s blog traffic went from 2,000 to 5,000. He was happy. I looked at the data: all new traffic came from Facebook, RPM only $0.8 (his original SEO traffic RPM was $3.8).

Traffic +150% Revenue Only +22%
Real case of low-quality traffic

Dimension 4: Geographic Dimension

Reports → Countries

CPC and RPM vary significantly between countries.

Benchmark data:

  • US, Canada: RPM $5-20
  • UK, Germany, France: RPM $3-12
  • Japan, Korea, Singapore: RPM $2-8
  • Southeast Asia, South Asia, Africa: RPM $0.5-3

My website gets 60% traffic from US (RPM $8.5), 30% from Southeast Asia (RPM $1.8), 10% from China (RPM $0.6).

If your traffic suddenly increases but revenue doesn’t, it might be increased traffic from low-value regions.

Dimension 5: Page Dimension

Reports → Content → URLs

This is particularly important. Find out which pages are dragging you down.

I will:

  1. List top 20 pages by traffic
  2. Check RPM for each page
  3. Mark pages with RPM below $2 (or below 50% of average)

Once I found 5 Python basic tutorials accounting for 45% of total traffic but RPM only $1.2-$1.8 (other tech articles RPM $6-12).

The problem: Basic tutorial readers are mostly students, advertisers aren’t willing to pay high prices (CPC only $0.08-$0.15).

Optimization solution: Add advanced technical content (cloud computing, DevOps). After 3 months, advanced content traffic went from 30% to 55%, overall RPM increased from $3.2 to $6.8.

Three Common Problems’ Data Characteristics

Problem 1: Revenue Suddenly Drops

Data characteristics:

  • Total revenue drops 20-50%
  • Could be impressions, CTR, or CPC having issues

Diagnosis steps (check in this order):

Step 1: Check impressions
Did traffic decrease?
→ Yes: Go to GA4 to check traffic sources, see which channel dropped
→ No: Proceed to Step 2

Step 2: Check CTR
Did click rate decrease?
→ Yes: Check if ad placement was adjusted or ads accidentally deleted
→ No: Proceed to Step 3

Step 3: Check CPC
Did price per click decrease?
→ Yes: Might be seasonal factors (Q4 advertisers have more budget, higher CPC; Q1 less budget, lower CPC)
→ No: Proceed to Step 4

Step 4: Check traffic quality
Compare RPM from different sources, see if junk traffic surged

The most typical example I encountered was what I mentioned earlier—accidentally deleting a mobile ad unit, CTR immediately halved.

Problem 2: Traffic Increases But Revenue Doesn’t

Data characteristics:

  • Impressions increase 30-50%
  • Revenue increases less than 10% (or doesn’t change)
  • RPM significantly decreases

Diagnosis steps:

Step 1: Compare RPM of new vs old traffic
Create segments in GA4 (new users vs returning users)
Compare with AdSense data

Step 2: Check traffic sources
Where is new traffic mainly coming from?
Is social media traffic surging? (Usually low RPM)

Step 3: Check content types
Are new pages low-value content (entertainment, gossip)?

Real case (still my friend):

His blog went from 2,000 to 5,000 daily visits:

  • Impressions: 6,000 → 15,000 (+150%)
  • Revenue: $18 → $22 (+22%)
  • RPM: $3 → $1.47 (-51%)

Problem: All new traffic came from Facebook. Users had short dwell time, high bounce rate, RPM only $0.8.

Optimization: Reduced social media promotion, focused on SEO. After 3 months, daily visits returned to 3,000, but revenue rose to $32/day (RPM $3.5).

You see, sometimes traffic decreases but revenue actually increases.

Problem 3: Some Pages Monetize Poorly

Data characteristics:

  • Some high-traffic pages have RPM below $2
  • These pages account for 30-50% of total traffic
  • Dragging down overall RPM

Diagnosis steps:

Step 1: Check "by URL" report in AdSense dashboard
Step 2: Find top 20 pages by traffic
Step 3: Mark pages with RPM <$2
Step 4: Analyze characteristics of these pages
  - What's the content theme? (Entertainment? News?)
  - Are ads placed? Is placement appropriate?
  - What's user behavior like? (Dwell time, bounce rate)

My tech blog encountered this problem. 5 Python basic tutorials accounted for 45% of traffic, RPM only $1.2-$1.8.

Why so low? Many student readers, advertisers unwilling to pay high prices.

Two optimization directions:

  1. Delete or rewrite this low-value content (aggressive)
  2. Add high-value content to dilute traffic share of low RPM pages (moderate)

I chose option 2. Added cloud computing and DevOps advanced content, overall RPM increased from $3.2 to $6.8 (+112%).

GA4 + AdSense Joint Analysis

Looking at AdSense data alone, you only know “revenue dropped.” But you don’t know why.

Adding GA4, you can see the complete picture.

Key metric comparison:

ProblemAdSense MetricGA4 Metric
Poor traffic qualityRPM decreasesBounce rate >70%, dwell time <30s
Irrelevant contentLow CTRHigh exit rate
Poor device experienceLow mobile RPMHigh mobile bounce rate
Poor traffic sourceLow social traffic RPMLow social traffic conversion

How to link AdSense and GA4?

Step 1: GA4 → Admin → AdSense linking
Step 2: Select your AdSense account → Save
Step 3: View AdSense data in GA4
   Reports → Monetization → AdSense revenue
   Can view RPM by page, source, device

After linking, you can see RPM for each traffic source and each page in GA4.

Comparing data, problems become obvious.

Real Cases - Data-Driven Optimization Decisions

I’ve talked about theory and methods. Now let me share 3 real cases, each with complete data, diagnosis process, and optimization results.

Case 1: Mobile RPM from $2.3 to $4.8 (+108%)

Background

This is my own tech tutorial blog. Daily visits around 3,000, mobile traffic accounts for 75%.

Problem discovered: Mobile revenue far lower than desktop.

Data Snapshot

Mobile (before optimization):
- Impressions: 6,000/day
- CTR: 0.9%
- CPC: $0.35
- RPM: $2.3

Desktop:
- Impressions: 2,000/day
- CTR: 2.8%
- CPC: $0.40
- RPM: $8.5

Mobile accounted for 75% of traffic but RPM was only 27% of desktop. This wasn’t normal.

Problem Diagnosis

I investigated following this logic:

Step 1: CTR gap too large (0.9% vs 2.8%). Problem is with ad display.

Step 2: Check mobile ad placement. Found all ads were below the fold (users had to scroll down to see them).

Step 3: Check ad types. All fixed-size ads (300x250), poor display on small screens.

Found it. Mobile users couldn’t see ads, or even when they saw them didn’t want to click (poor display).

Optimization Solution

I made 3 adjustments:

Adjustment 1: Reorganize ad placement

  • Add banner ad below header (above the fold)
  • Insert in-feed ad at 30% of article (appears naturally during reading)
  • Remove those unseen ads below the fold

Adjustment 2: Change all ads to responsive

  • Auto-adapt to mobile screens
  • More natural display

Adjustment 3: Enable anchor ads

  • Fixed display at bottom of screen
  • Doesn’t affect reading experience

Optimization Results (After 4 weeks)

+108%
Mobile RPM Increase
Mobile (after optimization):
- Impressions: 6,000/day (unchanged)
- CTR: 2.1% (up 133% from 0.9%)
- CPC: $0.38 (up 9% from $0.35)
- RPM: $4.8 (up 108% from $2.3)

Monthly revenue change:
- Before: Mobile $414 + Desktop $170 = $584/month
- After: Mobile $864 + Desktop $170 = $1,034/month
- Growth: +$450/month (+77%)

Honestly, when I saw the data I was pleasantly surprised. Same traffic, just adjusting ad placement and types, revenue increased 77%.

Key Insights

  • Mobile traffic accounting for 70-80% is normal, must optimize
  • Ad placement more important than quantity (removing low-performing positions actually increased revenue)
  • Responsive ads are essential for mobile

Case 2: Increasing CPC from $0.15 to $0.38 (+153%) Through Content Optimization

Background

This is a friend’s lifestyle blog. 5,000 daily visits, not small traffic, but revenue never took off.

Data Snapshot

Overall data (before optimization):
- Impressions: 15,000/day
- CTR: 1.8% (normal level)
- CPC: $0.15 (low)
- RPM: $2.7

Analysis by content type:
Content TypeTraffic ShareCPCRPM
Entertainment gossip50%$0.08$1.4
Food recommendations30%$0.18$3.2
Health wellness20%$0.42$7.5

The problem was obvious: 50% of traffic on entertainment gossip content, CPC only $0.08.

Problem Diagnosis

Entertainment gossip content:

  • Large traffic (easy to write, easy to spread)
  • But extremely low CPC (advertisers unwilling to pay high prices)
  • Dragging down overall revenue

Health wellness content:

  • Small traffic (requires expertise)
  • But high CPC ($0.42)
  • Unfortunately only 20% share

Optimization Solution

Adjust content strategy (a long-term process):

Step 1: Reduce low-value content updates

  • Entertainment gossip from 5 posts per week down to 1

Step 2: Increase high-value content

  • Health wellness from 2 posts per week to 4
  • Add new finance and education themes

Step 3: SEO optimize high-value content

  • Focus on optimizing keywords for health and finance articles
  • Internal linking to guide traffic to high RPM pages

Step 4: Wait for traffic structure adjustment

  • This process takes 3-6 months (SEO takes time)

Optimization Results (After 6 months)

+153%
CPC Increase
$0.15 → $0.38
+195%
RPM Increase
$2.7 → $7.98
+136%
Monthly Revenue Growth
$1,215 → $2,870
-20%
Traffic Change
Quality up, quantity down

Analysis by content type:

Content TypeTraffic ShareCPCRPM
Entertainment gossip25% (↓50%)$0.08$1.4
Food recommendations30% (unchanged)$0.18$3.2
Health wellness35% (↑75%)$0.42$7.5
Finance education10% (new)$0.65$11.7

See, traffic dropped 20% but revenue increased 136%.

This case made me deeply understand one thing: traffic quality is far more important than traffic quantity.

Key Insights

  • CPC is determined by content theme (not something you can arbitrarily control)
  • Rather than writing 100 low-value content pieces, better to write 10 high-value ones
  • Long-term strategy: Focus on high-value areas you’re good at

Case 3: Discovered and Removed Low-Performing Ad Units, Overall Revenue Increased 73%

Background

This is another friend’s programming tutorial site. Problem: Many ad units placed, but RPM never high.

Data Snapshot

View by ad unit (many people don’t know about this feature):

Ad UnitImpressionsCTRRPM
Below header5,0002.8%$9.8
Middle of article5,0003.2%$11.2
Sidebar top5,0001.5%$5.2
Sidebar middle5,0000.6%$2.1
Sidebar bottom5,0000.3%$1.0
Footer5,0000.2%$0.7

Problem Diagnosis

Looking at this data, the problem is obvious:

  • Below header, middle of article: High CTR, high RPM (good positions)
  • Sidebar top: Medium CTR and RPM (acceptable)
  • Sidebar middle, bottom, footer: Extremely low CTR, extremely low RPM (dragging down performance)

These 3 positions (sidebar middle, bottom, footer) have impressions but almost no one sees or clicks them. They’re not only not contributing revenue, they’re actually lowering overall RPM.

Optimization Solution

My recommendation to him was simple: Remove low-performing ad units.

Remove: Sidebar middle, bottom, footer
Keep: Below header, middle of article, sidebar top

From 6 ad units down to 3.

He was worried at the time: “Half the ads are gone, won’t revenue drop?”

I said: “Look at the data. Those 3 positions have CTR only 0.2-0.6%, RPM only $0.7-$2.1. Removing them won’t significantly impact revenue, and user experience will actually improve a lot.”

Optimization Results (After 2 weeks)

+73%
Revenue Growth After 3 Months
Ad UnitImpressionsCTRRPM
Below header5,0003.1%$10.8
Middle of article5,0003.5%$12.2
Sidebar top5,0001.8%$6.3

Impressions halved, revenue almost unchanged.

More importantly: User experience significantly improved (50% fewer ads), bounce rate decreased, dwell time increased, SEO ranking actually improved.

After 3 months, traffic naturally grew 15%. Combined with higher RPM, total revenue increased 73%.

Key Insights

  • Less is more: Removing low-performing ad units is a win-win for user experience and revenue
  • Focus on individual ad unit CTR and RPM, not total impressions
  • Ad units with CTR below 0.5%, boldly remove them

Actionable Optimization Decision Process

Weekly Data Analysis Checklist (5-Step Method)

Step 1: Check Overall Trends (5 minutes)

Action steps:
1. Open AdSense dashboard
2. Reports → Select "Last 7 days"
3. Look at total revenue, RPM, CTR, impressions

Checklist:
□ Is total revenue normal? (Compare to last week)
□ Did RPM drop &gt;10%?
□ Is CTR abnormal? (&gt;5% might be accidental clicks, &lt;0.5% needs optimization)
□ Did impressions fluctuate significantly? (&gt;20% needs attention)

If all 4 items are normal, you can relax. If there’s an anomaly, continue investigating.

Step 2: Device Dimension Analysis (5 minutes)

Action steps:
Reports → Platforms → Compare mobile and desktop

Checklist:
□ Is mobile RPM &lt;50% of desktop?
□ Is mobile CTR &lt;1%?
□ Does mobile traffic account for &gt;70%?

If mobile RPM is too low and traffic share is high, this is your biggest optimization opportunity.

Refer to Case 1’s optimization methods (adjust ad placement, use responsive ads).

Step 3: Page Dimension Analysis (10 minutes)

Action steps:
Reports → Content → URLs

Checklist:
□ List top 20 pages by traffic
□ Mark pages with RPM &lt;$2 (or below 50% of average)
□ What percentage of total traffic do these pages represent?

If low RPM pages account for more than 30%, consider:

  • Add high-value content to dilute low RPM page traffic
  • Or rewrite/optimize these pages

Step 4: Ad Unit Analysis (10 minutes)

Action steps:
Reports → Ad units

Checklist:
□ List CTR and RPM for all ad units
□ Mark ad units with CTR &lt;0.5%
□ Mark ad units with RPM below average

Ad units with CTR below 0.5%, boldly remove them. Refer to Case 3.

Step 5: Create Optimization Plan (10 minutes)

Based on above data, choose 1-2 optimization points:

Priority ranking:
1. Fix obvious problems (like extremely low mobile RPM)
2. Remove low-performing ad units (CTR &lt;0.5%)
3. Adjust content strategy (increase high CPC themes)
4. Test new ad placements

Only change one thing at a time. After changes, wait 2-4 weeks, look at data, then decide next step.

Create Optimization Plans Based on Abnormal Metrics

Anomaly 1: RPM Suddenly Drops >20%

Optimization steps:
1. Check which dropped: impressions, CTR, or CPC
2. Compare device, region, page dimensions
3. Find specific cause (ad unit changes, traffic source changes, content theme changes)
4. Restore or adjust configuration

I encountered this last November. Finally found I had accidentally deleted an ad unit. After restoring, it immediately came back.

Anomaly 2: CTR Suddenly Drops

Optimization steps:
1. Check if ad placement was accidentally deleted or adjusted
2. Check which dropped: mobile or desktop
3. Restore high CTR ad units
4. Test new ad colors and styles

Anomaly 3: Traffic Increases But Revenue Doesn’t

Optimization steps:
1. Compare RPM of new vs old traffic
2. Check traffic sources (GA4)
3. If low-quality traffic, consider:
   - Reduce promotion on that channel
   - Or: Optimize user experience and ad layout for that traffic

Refer to Case 2. My friend encountered this exact problem.

Avoid 4 Common Mistakes

Mistake 1: Only Look at Total Revenue, Not Metric Details

Wrong approach: Only look at “how much did I make today” each day

Correct approach: Regularly check CTR, RPM, CPC trends

Consequence: When problems arise, you don’t know the cause and can’t optimize

Mistake 2: Frequently Adjust Ad Placement

Wrong approach: Adjust daily without waiting for data to stabilize

Correct approach: After each adjustment, observe for at least 2-4 weeks

Consequence: Can’t determine which adjustment was effective (too many variables)

I’ve seen people change ad placement 3 times in a week. Data got completely messy, couldn’t tell which version was better.

Mistake 3: Pursue High CTR at the Expense of User Experience

Wrong approach: The more ads the better, ads everywhere

Correct approach: Balance user experience and revenue

Consequence: High bounce rate, low return rate, long-term revenue decline

Some websites get 8-10% CTR, but it’s all accidental clicks. User experience is terrible, traffic eventually declines.

Mistake 4: Ignore Mobile Optimization

Wrong approach: Only optimize desktop (because you use a computer)

Correct approach: Prioritize mobile optimization (accounts for 70-80% of traffic)

Consequence: Miss the biggest revenue improvement opportunity

Look at Case 1. After mobile optimization, my revenue increased 77%.

The Right Way to Do A/B Testing

If you want to optimize scientifically, you can do A/B testing.

What to Test?

  • Ad placement (Header vs middle of article)
  • Number of ads (2 vs 3 vs 4)
  • Ad colors (default vs custom blended)
  • Ad types (display ads vs in-content ads)

How to Test?

Step 1: Determine test goal (increase RPM)
Step 2: Choose one variable (ad placement)
Step 3: Create version A and version B
Step 4: Run for at least 2-4 weeks (accumulate enough data)
Step 5: Compare RPM and CTR
Step 6: Keep better performing version
Step 7: Continue testing next variable

Precautions

  • Test only one variable at a time (don’t change placement and color simultaneously)
  • Run long enough (at least 2 weeks, preferably 4 weeks)
  • Too little traffic (daily PV <500) not suitable for A/B testing
  • Record results of each test (create an Excel sheet)

I have an Excel sheet recording all tests from the past year:

DateTest ContentRPM ChangeConclusion
2024-03Header ad placement+18%Adopted
2024-05Mid-article ad+25%Adopted
2024-07Sidebar ad color-5%Abandoned
2024-09Mobile anchor ad+35%Adopted

This way you know what works and what doesn’t.

Conclusion

By now, you should understand: AdSense data analysis isn’t some advanced technique, it’s just regularly checking dashboard data, finding problems, making adjustments.

Let me recap the core points:

Four Core Metrics: Impressions, CTR, CPC, RPM. Among them, RPM is most important—it comprehensively reflects your monetization capability.

Five Monitoring Dimensions: Time, device, traffic source, geography, page. Spend 30 minutes weekly looking at these 5 dimensions’ data, problems basically can’t escape.

Three Real Cases:

  • Mobile optimization: RPM from $2.3 to $4.8 (+108%)
  • Content optimization: CPC from $0.15 to $0.38 (+153%)
  • Removing low-performing ad units: Overall revenue increased 73%

These all actually happened, data’s right there.

Honestly, for AdSense optimization, data analysis is the most fundamental and most important step. Without data support, you’re just adjusting blindly. With data, you know where the problem is and which direction to optimize.

3 Things You Can Start Today:

  1. Open AdSense dashboard, record current data

    • Write down your CTR, RPM, CPC
    • Compare with benchmark data in this article
  2. Check mobile vs desktop RPM difference

    • If mobile RPM is below 50% of desktop, this is your biggest optimization opportunity
  3. Find top 10 pages by traffic, mark those with RPM below average

    • See what percentage of total traffic these pages represent
    • Consider adding high-value content to dilute low RPM page traffic

Following the methods in this article, over the next 4 weeks, your revenue should be able to increase 20-50%. No exaggeration, I’ve seen too many cases.

Finally, let me ask you a question: What’s your current RPM? Have you discovered any problems through data analysis?

Feel free to share your data and experiences in the comments. Let’s communicate and improve together.

FAQ

What's the most important metric in AdSense dashboard?
RPM (Revenue Per Mille) is the most important comprehensive metric. It directly tells you how much you can earn from 1000 impressions, comprehensively reflecting the effect of CTR and CPC.

Benchmark data reference:
• General content (news, entertainment): $1-3
• Medium level (tech, lifestyle): $3-8
• Excellent performance (niche domains): $8-15
• Top tier (fintech): $15+

If your RPM is below $3, you need to optimize from three aspects: traffic quality, content theme, and ad placement.
Is it normal for mobile RPM to be much lower than desktop?
Normally, mobile RPM should be 50-70% of desktop RPM. If it's below 50%, optimization is needed.

Main optimization directions:
• Ad placement: Ensure above the fold visibility, insert below header or at 30% of article
• Ad type: Use responsive ads that auto-adapt to mobile screens
• Ad format: Enable anchor ads (fixed at bottom of screen) to increase exposure

Case data: Through these three adjustments, I increased mobile RPM from $2.3 to $4.8 (+108%), monthly revenue grew 77%.
Traffic increased but revenue didn't—what's the reason?
This is usually caused by a surge in low-quality traffic. Need to check two dimensions:

1. Traffic sources (check in GA4):
• SEO traffic: RPM usually $4-10 (highest quality)
• Direct visits: RPM $3-6 (medium)
• Social media: RPM $0.8-2 (lowest)

2. Content types:
• Entertainment gossip: CPC $0.05-0.15
• Tech tutorials: CPC $0.50-2.00
• Finance: CPC $2.00+

Real case: Friend's blog traffic went from 2,000 to 5,000 (+150%), but revenue only from $18 to $22 (+22%), because all new traffic came from Facebook with RPM only $0.8. Later reduced social promotion, focused on SEO, traffic returned to 3,000 but revenue rose to $32/day.
Are more ad units always better?
No, fewer but better ad units work better. The key is individual ad unit CTR and RPM, not total quantity.

Judgment criteria:
• Ad units with CTR &lt;0.5%: Boldly remove
• Ad units with RPM below average: Consider removing or adjusting
• Poor user experience positions (sidebar bottom, footer): Usually extremely low CTR

Real case: Friend's site went from 6 ad units down to 3, removing sidebar middle, bottom, and footer three low-performing positions (CTR only 0.2-0.6%). Result: RPM increased from $5.0 to $9.77 (+95%), 3 months later revenue grew 73%.

Principle: Remove low-performing ad units → User experience improves → Bounce rate decreases → SEO ranking improves → Traffic increases → Revenue increases.
How to quickly find the cause of revenue decline?
Check in this order step by step, 5-10 minutes to pinpoint the problem:

Step 1: Check impressions
• Did traffic decrease? → Go to GA4 to check traffic sources

Step 2: Check CTR
• Did click rate decrease? → Check if ad placement was accidentally deleted or adjusted

Step 3: Check CPC
• Did price per click decrease? → Might be seasonal factors (Q4 more budget, Q1 less budget)

Step 4: Check traffic quality
• Compare RPM from different sources → See if junk traffic surged

My own experience: One day revenue suddenly dropped from $25 to $15. After checking, found CTR dropped from 2.2% to 0.9%. Further investigation revealed mobile in-feed ad was accidentally deleted. After restoring, revenue immediately returned to $25.
How often should I check data?
Recommend spending 30 minutes weekly for systematic checks, specific process:

• Step 1: Check overall trends (5 minutes)
Total revenue, RPM, CTR, impressions—are they normal

• Step 2: Device dimension analysis (5 minutes)
Mobile vs desktop RPM difference

• Step 3: Page dimension analysis (10 minutes)
Find top 20 pages by traffic, mark low RPM pages

• Step 4: Ad unit analysis (10 minutes)
Check CTR and RPM for each ad unit

• Step 5: Create optimization plan (10 minutes)
Choose 1-2 optimization points, only change one thing at a time

Note: After each adjustment, observe for at least 2-4 weeks before making next change, avoid too many variables making it impossible to judge effectiveness.
What types of content have the highest CPC?
CPC is determined by advertiser bids—high-profit industries are willing to pay high prices:

• $2.00+ (top tier): Finance, investment, insurance, enterprise software, SaaS
• $0.50-2.00 (high value): Programming, design, marketing, cloud computing, DevOps
• $0.15-0.50 (average): Lifestyle, food, travel, health, education
• $0.05-0.15 (low value): Entertainment, gaming, gossip, news

Real case: Friend's lifestyle blog had 50% traffic on entertainment gossip (CPC $0.08), overall CPC only $0.15. Through 6 months content adjustment, reduced entertainment content, increased health wellness (CPC $0.42) and finance education (CPC $0.65), overall CPC increased from $0.15 to $0.38 (+153%), monthly revenue grew from $1,215 to $2,870 (+136%).

Recommendation: Focus on high-value areas you're good at. Rather than writing 100 low-value content pieces, better to write 10 high-value ones.

19 min read · Published on: Jan 8, 2026 · Modified on: Jan 22, 2026

Comments

Sign in with GitHub to leave a comment

Related Posts