The core motivation for the experiment was to test the prevalent assumption that people coming from search engines are the ones most likely to click on ads. To this end, I used a little piece of JS that would present different ads to people based on how they arrived on this site – either directly, from a search engine, from another kind of external site, or from a different page on the same site. All the ads were visually identical but were tracked separately.

(See the original announcement and the first summary for more details.)

The visitor segmentation script has been running uninterrupted for more than five months, allowing me to collect CTR and CPM data on more than 10 000 AdSense clicks. Overall, the trends look stable enough that running it for a while longer probably wouldn’t change the results. So now is a good time to post one final summary and declare the experiment concluded.

Internal Traffic Wins

At least on paper, it does. Out of all possible traffic sources, internal traffic – i.e. people who browse more than one page into the site – has the highest click-through rate and the highest eCPM. However, it also accounts for the least number of ad impressions. In a nutshell, internal traffic is valuable but rare.

Conversely, search engine traffic comes third in terms of CTR and eCPM, but brings the most money due to a high number of impressions.

The full results are below. To comply with AdSense ToS which prevent me from displaying the actual numbers, the results have been normalized to display the relative CTR and eCPM of various traffic sources.

Source Data

The above results are based on:

• > 18 000 clicks.
• > 400 000 impressions (73% search engine traffic).

All data was collected during an uninterrupted 168 day period (2010.03.09 – 2010.08.24) on this very same site. Results may be different for other sites.

Does this unconventional conclusion remain true  in the long-term? Today I’m going to post a follow-up study that discusses just that. The long-term results were… different.

Newly Charted

As of today, the visitor segmentation script has collected 52 days’ worth of data. This is five times the duration of the previous experiment (which lasted just over a week), so the results presented below are presumably much more accurate that those I posted last time.

Without further ado, here’s the updated CTR chart :

Units hidden to comply with AdSense ToS.

The updated study shows that visitors arriving from external sites (except search engines) are those who yield highest click-through rate. In the previous analysis, this group occupied the second-to-last spot. Internal visitors, who were the previous “leaders” – if one may use that word to describe people very likely to click on Internet ads – now fall into the second place. Search engine visitors come third. Finally, direct visitors – i.e. people who arrive via browser bookmarks or by typing in the URL – are still the least likely to create advertising income.

Here’s a chart comparing the old and new results :

Darker shades represent the previous results, lighter ones – updated research.

Of course, we also need to consider eCPM. eCPM represents the average earnings per 1000 impressions. Again, I can’t show you the actual numbers due to AdSense rules, but this chart should give you a general idea of the relative financial value of various traffic sources :

Interestingly, this is pretty close to what we saw in the previous study – internal visitors generate the most valuable clicks. External and search visitors share the second spot, and direct traffic is the least valuable advertising-wise.

Here’s a comparison with the previous results :

The past is dark, the present light. Future not shown.

Conclusions

Despite the updated rankings, the overall conclusion is the same one I arrived to last time – on this site, the CTR of search, external and internal visitors is very close. Showing ads only to one kind of visitors wouldn’t make much sense.

Likewise, the main take-away for other AdSense users is still this : don’t rely blindly on stuff you read on a blog somewhere; test any suggested optimization techniques yourself.

I think that the fact that I got slightly different results this time only reinforces the validity of that advice

The core motivation for the experiment was to test the prevalent assumption that people coming from search engines are the ones most likely to click on ads. To this end, I used a little piece of JS that would present different ads to people based on how they arrived on this site – either directly, from a search engine, from another kind of external site, or from a different page on the same site. All the ads were visually identical but were tracked separately.

The Numbers Are In

As it turns out, search engine visitors are not the ones most likely to click on ads – when different types of traffic are ranked by the click-through ratio (CTR), search visitors actually fall into the second place. The most eager ad-clickers are the people entering the page via an internal link. Here’s a comparison chart :

My apologies for hiding the units. AdSense prohibits one from publicly disclosing their CTR and related figures.

While this might not be obvious from the chart, the actual CTR disparity between internal, external and search traffic is very low (less than one percent). However, direct visitors – e.g. people accessing the site via a bookmark or clicking a link in an email – are significantly less likely to click on ads.

CTR alone doesn’t give us the full picture – it tells us how many clicks we can expect to get if we display our ads to a certain type of visitor, but not what those clicks are worth. To judge the financial value of a visitor we need to consider CPM :

Again, the specific numbers have been hidden to conform with the AdSense ToS.

The same trend is even more pronounced here – people clicking on internal links also deliver the most valuable AdSense clicks, people arriving from search engines come second, and external & direct visitors are, on average, the least likely to bring in oodles of cash.

Conclusions

Over the course of the experiment, the pages that had the ad-varying script installed received about 23 000 impressions and over 1500 AdSense clicks. I feel that those numbers are big enough to call the results statistically significant.

Given that, and how close the CTR values of search, external and internal visitors are for this site, it wouldn’t really make sense to show ads only to one of those groups. However, I can safely stop showing AdSense to people who visit the site via bookmarks and not lose anything much, income-wise.

Now the question that’s probably on your mind is : “How does this translate to my site?” Should you show ads to everyone, or just a certain kind of visitor? Now that I’ve demonstrated that the folk wisdom of only showing ads to search visitors is not necessarily valid, how are we to maximize our financial fortunes?

It depends

I’m not going to substitute one piece of dubiously specific advice for another. Instead, the take-away of this exercise is :

Don’t take AdSense micro-optimization advice on trust. Test it, and use what works for you!

Note : The keyword generator will only work in modern browsers, e.g. Firefox, Opera and Google Chrome. Internet Explorer is not supported.

Here’s a screenshot :

Using the keyword generator is pretty straightforward – just enter a seed keyword in the text box and click “Generate Suggestions”. This will load up to 10 similar keywords from Google Suggest and add them all to the results table. You can repeat the process with a different keyword to get even more suggestions, or click one of the “Dig” links to generate related keywords.

You can specify multiple seed keywords by separating them with a comma. Alternatively, you can also click “Expand textbox” to transform the single-line input box into a multi-line text area, then enter one seed keyword per line.

Clicking on the “Copy Keywords” link will pop up a new text box containing the entire list of suggested keywords in an easily “copy & paste”-able form. Finally, clicking “Clear” will empty the results table so that you can start afresh.

With all this commotion and ad-blocker evangelizing going on, one has to wonder – is AdBlock on the way to becoming the norm, and thus completely devaluing online advertising?

No. Not even close.

I’ve run the numbers. If current Internet usage trends persist, people who use AdBlock Plus (and other ad-blocking plugins) will always be a tiny minority. Sure, it might have some impact on the ad revenue of websites that target the tech-savvy crowd. But advertisers in general will probably always see AdBlock as just a minor annoyance – if that.

See the stats below.

AdBlock Plus Popularity

By last count, there were ~1,734 million Internet users in the world. AdBlock Plus, the most popular ad-blocking plugin, has ~10 million active users. Thus, only about 0.6% of all Internet users have AdBlock installed.

Growth Rates

In the last two years, the number of Internet users has been increasing by about 19.7 million each month, and this growth is accelerating. I wasn’t able to find complete historical data for AdBlock Plus, but judging by these stats (see the monthly view) the number of AdBlock users is currently increasing by about 315 thousand per month. And it seems to be slowing down lately.

With a growth rate disparity this huge, AdBlock will never catch up.

Sources : InternetWorldStats.com, AdBlock Plus usage statistics

Viralogy Script

About two weeks ago, I received an offer to bundle a social media tracking script from Viralogy with my Broken Link Checker plugin. This script, which is properly called the “Dynamic Insights Tracker”, tracks the flow of visitors and analyses their preferences and social media activity. If installed on your blog, it would record various data about your visitors and report it back to Viralogy. In turn, this information could then be used to present the tracked visitors with a personalized experience on other sites.

Here’s a simplified example :  say you have a  blog about the Kindle and other e-ink devices, and you’ve got the tracker script installed. When someone reads an e-book reader review on your blog, the tracker script will take note of that. If the same person later goes on to visit an e-commerce site that deals in a variety of gadgets, the Viralogy API will enable that site to find out that the visitor might be interested in e-book readers, and present them with a dynamically optimized store page that puts the Kindle front and centre.

Naturally, I have been assured that privacy matters a lot to the people at Viralogy Inc, and that nothing untoward would happen to anyone’s personal information. This seems about as believable as the Google Analytics privacy policy.

What’s In It For Me?

Money. Well, duh.

According to Viralogy, I would receive a small bi-annual payment per each blogger who installs my plugin(s) and enables the script. Given that I still don’t have any reliable statistics on how many people actually use my plugins, it’s not really possible to tell how much cash that would amount to. By a very rough estimate, it could be anywhere from $50 to $5000. This uncertainty vexed me.

What’s In It For You?

Not much.

I’m sure I could make up some plausible-sounding bulshytt about how installing the script demonstrates your gratitude towards the plugin developer, and how having the script installed allows you to express that gratitude without actually giving the dev. any money yourself. And it does sound plausible, now that I think of it. But in the end, the tracker script wouldn’t give you any direct benefits.

The survey results also show a general lack of enthusiasm for third-party scripts :

Q : I have received an offer to bundle a social media analysis script from Viralogy.com with Broken Link Checker. Your thoughts?

Include the script - 16%, Don't include the script - 39%

People who chose “Other” mostly asked for more information about the script, or said they would be okay with the script being included if actually enabling it was optional. Of course, it would definitely be optional – that’s the only non-evil way to include the script.

Q : If the aforementioned script *was* included, would allow the plugin to install/enable it?

Yes - 39%, No - 61%

Conclusions

Tricky tricky. My gut says “maybe”, so let’s do it the hard way and evaluate the pros & cons of this offer.

Pros :

• Me : A vaguely defined amount of cash.
• Users : None.
• Everyone else : The beginnings of personalized web-browsing. Yes, this is a good thing. It’s also pretty much inevitable in the long-term – if not by Viralogy, then certainly by the future efforts of Google/Amazon/Microsoft.

Cons :

• Me : Potential for negative publicity. That could even be useful, but meh.
• Users : None that I can see. The script would be optional anyway, so those who don’ t like it could simply not enable it. Perhaps a minority would have some ideological objections? Either way, the user response was generally unenthusiastic.
• Everyone else : Another tracking script for paranoid Internet users to worry about. Compare with Google Analytics & co.

The costs are relatively small, and the benefits are likewise not very impressive. Conclusion : Not worth it.

Perhaps some other time.

Feature Highlights

• Automatically cloak links in any part of your site. This means you can also cloak links in the sidebar, footer, etc.
• Comes with multiple cloaking techniques, like the ability to display the cloaked link in a frame or hide the HTTP referer info.
• Improved click statistics + optional Google Analytics integration.
• Gives you more control over which links get cloaked – e.g. you can cloak a bunch of links based on their domain name.
• Backwards compatible with the free version – the settings will be imported automatically and any links you’ve cloaked using the <!–cloak–> tag or the “Cloak All” mode will continue to be cloaked with no additional effort on your part. The old “static” cloaked links can be imported with a few clicks.
• A new cool name – Eclipse Link Cloaker

You can find a more complete list of features (along with screenshots and tutorials) on the plugins website.

Wait, “Premium”?

Yes, that curious adjective means the plugin also comes with a price-tag. Not to worry though – during the initial release period you can grab the new plugin for a meagre sum of $25 and apply it’s awesome cloaking power to any number of sites. Sorry folks, the introductory offer is over and the price has gone up to $57! But there’s still a 100% money-back guarantee, so you wouldn’t really be risking anything.

Miscellaneous Notes

Phew. I hope that didn’t sound too much like sales talk. Anyway, there are a few more minor points I’d like to mention :

• There will be an affiliate program with 50%-70% commissions. It’s already functional now (so if you have any special ideas, let me know), I just haven’t set up the public-facing affiliate how-to and such.
• The plugin will have free, automatic updates and can be updated using the same interface that is used to update plugins hosted on wordpress.org.
• Obviously, there’s a little bit of DRM involved.
• mod_rewrite/pretty permalinks are no longer required. The plugin automatically detects what kind permalink structure the user has and alters it’s cloaked URLs appropriately.

If you have any questions or doubts, feel free to leave a comment.

Well, this post is actually my attempt to answer a burning question : “How can we download stuff for free and ensure that musicians, actors, programmers and other people who create stuff still get paid?”. It’s a longish, wistful rant on intellectual property and conflicts between scarcity-based economy and post-scarcity society (The Web), with a few constructive suggestions tacked on in the end. Comments are welcome.

The Problem

What do the following examples have in common :

• Lawsuits against filesharers and sites that enable filesharing. Case in point : The Pirate Bay trial.
• Fledging attempts to make the Internet fully monitored, supposedly to stop copyright infringement.
• People who write great OSS/freeware tools still need to have a “day job” to pay the bills.
• Ridiculously cumbersome DRM.
• Buying something through an online content delivery service can cost more than buying the same thing in a physical store. Case in point : Steam.
• The fact that the owner of 4chan is in debt.

Any ideas? Well, here’s mine : all of these problems are caused by a conflict between scarcity and post-scarcity.

We’re currently at an unique point in time, a time when there still exists a scarcity of many resources but a few resources are becoming effectively free (disk space, computing power, bandwidth… well, almost). This leads to a conflict of intuitions. On the one hand, people realize that copying a file or loading a website costs practically nothing (uses non-scarce resources), so they don’t want to pay for downloaded movies/music/software or pay for using a site like 4chan or Wikipedia.

On the other hand, the producers of value – actors, musicians, programmers, webmasters – have to expend scarce resources to create stuff, even if the end-result is stored in a non-scarce medium. People still need equipment like computers, music instruments, brushes, filming sets, video cameras and other tools of their craft. They also need food, shelter, clothes, transportation, time and perhaps web hosting. You can download [The.Dark.Knight[2008]DvDrip-aXXo.avi] for free, but it cost millions to produce.

In other words, creators need a way to obtain scarce resources in exchange for non-scarce resources. They’ve spent time and money to make something people like and they feel they should get something tangible in return (also, they want resources to make more stuff in the future). Of course, this is intuitively unacceptable to consumers who wonder “Why should I pay for stuff I can get for free?”.

And so it goes.

Bad Solutions

So far, all the attempts to deal with this conflict have only lead to partial solutions at best and crazy abolish-human-rights projects at worst.

• Artificial scarcity. Yeah, I’m looking at you, DRM. Trying to turn a non-scarce resource into a scarce one seems downright criminal when you consider that this thing we call “progress” normally leads to less scarcity, not more. Also, I’m sure you’re aware of the fact that DRM has so far proven ineffective at deterring piracy and might even push more people towards using torrents as it’s more convenient than dealing with idiotic file protection schemes.
• Selling merchandise and tickets for live performances. This is a fine practice in it self, but it only works for certain niches and levels of popularity. A band like Sonata Arctica can probably make a decent amount of cash on t-shirts and tickets, but I doubt anyone would pay to see the author of the WordPress plugin WP Super Cache programming “live” (no offense Donncha ).
• Get a job. Of course, the programmers, writers, musicians and everyone else could just get a “normal” job and use that income to support their art, right? That works in theory, but this solution creates more problems than it solves. First, you have to find a job that doesn’t suck, which is no easy feat. Second, you’d have less time to spend on the thing you actually want to do. And finally, there’s one drawback that most people fail to consider – if your day job is in the same field as your “art” (e.g. a programmer working on commercial software to support his open-source projects), chances are it will be subject to the same scarcity vs. non-scarcity problem. Getting a job only shifts the problem to the employer, not eliminate it.
• Advertising. This is another partial solution that only works for certain niches. How would product placement work in music, and would you really want it to? The truth is people hate ads and can go to great lengths to get rid of them. A website with AdSense on it might get away with it for a while (though a lot of geeks seem to think in an ideal world everybody would be using Adblock), but adware is usually mercilessly eradicated and commercial breaks in streamed videos are seen as a deadly sin. In summary : advertising works, but it’s not the best solution.

A Better Solution?

Any proposed solution needs to answer two basic questions :

• Where does the money come from? As much as we’d like that, we can’t really get everything for free. So even if we invent a perfect system for dealing with postscarcity in a world that still depends largely on a scarcity-based economy, somebody will have to pay = provide the required scarce resources to the people producing non-scarce stuff. This could be either the consumers, or advertising income.
• Who gets the money, and how much? This is a problem of resource allocation. You’ve probably heard various rumors that record companies bag most of the income and hand only pennies to the artists. Another example would be mediocre videogames that cost as much as good ones, and the fact that you have to pay up front for movie tickets. Ideally, what you pay should be commensurate with how valuable the product is to you.

My proposal : Whuffie + real money

Imagine a membership service called “PostRep” (from post-scarcity and reputation). This service takes a small subscription fee, say, $20-$30/month. This is where the money comes from (advertising might also be used – more on that later). In return, every member of the system gets the ability to hand out and receive reputation points. You could also call it Whuffie, karma, upvotes, tokens or whatever else you think is cool and marketable. Team accounts can also be created. Each team has to decide on their own how to redistribute points received by a team account as the system doesn’t impose any particular algorithm there.

You can hand out as many points as you like – there is no limit – and you can do it manually and/or automatically. For example, you could give 100 points to the programmer who wrote an open-source program you use, and your music player could automatically give 1 point to the author(s) of each track you listen to. Downloading stuff via the PostRep system would also automatically assign a (configurable) amount of reputation to the creator of the file. This should be fairly easy to integrate with existing rating systems, so that when you rate a movie on IMDB, you also automatically send a proportional amount of points to the team who made it.

What (else) is in it for me?

As a member of this imaginary system, you get free and easy access to the non-scarce values created by any member of the PostRep online system. Music, movies, software, books, whatever else that can be expressed and transmitted digitally. No DRM and other bullshit. Creating a system that actually works would be a huge engineering challenge, but it’s definitely feasible (see also : Steam and similar services).

As a side-effect of the reputation system you can also get a very flexible rating scheme for digital content and even other people. Hell, you could even implement the “right-handed vs. left-handed Whuffie” idea and see how much reputation points where assigned by people you like vs. the people you dislike. Of course, the amounts would need to be normalized based on the total number of points handed out by each individual, but that would be pretty easy from the engineering perspective.

Where’s the money?

So what happens to your subscription fee? Simple – each month, PostRep redistributes the funds to other members proportionally to the amount of rep. points/Whuffie/whatever you gave them. This also means that you get a smaller or larger sum as well, depending on how much rep you got for your work. Popular artists, OSS programmers and the owners of popular websites could even make a decent income.

Advertising

A more controversial approach would be to make the system partially or wholly ad-supported. Marketers would jump on the possibility to access data on per-user ratings and point transactions, as it would give them the ability to create finely targeted advertising. However, as I mentioned before, advertising isn’t generally well-regarded by users. Making it opt-in might help.

And where’s the catch?…

Of course, there are several disadvantages to this idea, like political feasibility and how the PostRep earnings of popular artists would compare to their current income. I’ve also been (intentionally) vague on how exactly the reputation points would be calculated and normalized. However, I won’t go into details on that today – this post is already long enough as it is Feel free to post any criticisms and suggestions in the comment section.

Hmm, I really need to find some illustrative pictures next time.

But enough about that. If we lay aside the morality debate, do you know how big is the real impact of Adblock users on your site’s financial success? Perhaps you’re only “losing” 0.1% of your potential ad impressions. Or it might be over 50%. In any case, it wouldn’t hurt to have some actual numbers, right? I’ll show you how to get them.

Nuts And Bolts

We’ll need two things – a way to detect the Adblock extension, and a way to track the number of visitors that use it. For the first part, there’s a simple JavaScript solution posted in a StackOverflow thread that we can adapt for our needs. With a few minor changes, this script seems to work nicely with the latest version of FF and correctly detects Adblock Plus 1.0.1.

Tracking the Adblock users is a bit trickier. There are two choices here – either write a specialized web stats tool (a marvelous waste of time), or try to integrate the detection script with your existing analytics software (requires grokking the API). I decided to go the second route and use Google Analytics custom segments for this task. If you’re already using GA on your site, creating a custom segment is relatively painless (more on this later) and you get all the advanced reporting with no additional hassle.

Bringing It All Together – The Tracker

Place the below piece of code on every page where you want to track Adblock users. It doesn’t matter where exactly you put it, as long as there’s the GA tracker code on the same page, too.

Sidenote for WordPress users : The easiest way to install this script on your site is to put it in your theme’s footer. Go to Appearance -> Editor, click on Footer and paste the code just before the </body> tag.

<script type="text/javascript">
/* <![CDATA[ */
 
function checkAdblock() {
	if ( detectAdblock() ){
		trackAdblock('Adblock Detected');
	} else {
		trackAdblock('No Adblock');
	}
}
 
//Run checkAdblock when page is loaded
if(document.addEventListener)
	document.addEventListener("DOMContentLoaded", checkAdblock, false);
 
//adapted from http://stackoverflow.com/questions/2472/how-can-i-tell-if-a-web-client-is-blocking-ads
function detectAdblock(){
	//Create an invisible test image 
	var AbpImage = document.createElement("IMG");
	AbpImage.id = 'abp_detector';
	AbpImage.src = '/images/ad/banner.gif';
	AbpImage.style.width = '0px';
	AbpImage.style.height = '0px';
	AbpImage.style.top = '-1000px';
	AbpImage.style.left = '-1000px';
	document.body.appendChild(AbpImage);
	//Check if Adblock has tried to hide the image
	var AbpDetected = (ws_getComputedStyle(document.getElementById('abp_detector'),'display') == 'none');
	//Be nice - remove the image 
	document.body.removeChild(AbpImage);
	return AbpDetected;
}
 
//http://www.robertnyman.com/2006/04/24/get-the-rendered-style-of-an-element/
function ws_getComputedStyle(oElm, strCssRule){
	var strValue = "";
	if(document.defaultView && document.defaultView.getComputedStyle){
		strValue = document.defaultView.getComputedStyle(oElm, "").getPropertyValue(strCssRule);
	}
	else if(oElm.currentStyle){
		strCssRule = strCssRule.replace(/\-(\w)/g, function (strMatch, p1){
			return p1.toUpperCase();
		});
		strValue = oElm.currentStyle[strCssRule];
	}
	return strValue;
}
 
//Record Adblock state  
function trackAdblock(state){
 
	function readCookie(name) {
		var nameEQ = name + "=";
		var ca = document.cookie.split(';');
		for(var i=0;i < ca.length;i++) {
			var c = ca[i];
			while (c.charAt(0)==' ') c = c.substring(1,c.length);
			if (c.indexOf(nameEQ) == 0) return c.substring(nameEQ.length,c.length);
		}
		return null;
	}
 
	var current_cookie = readCookie('__utmv');
	if ( (current_cookie==null) || (current_cookie.indexOf(encodeURIComponent(state)) == -1) ){
		//Set the user-defined segmentation variable with the Google Analytics tracker
		if (typeof pageTracker != 'undefined'){
  			pageTracker._setVar(state);
	  	}
	}
}
 
/* ]]> */
</script>

Sorry for the lack of indentation – it looks like my syntax highlighter plugin has a few glitches.

Viewing The Results

Now that you’ve installed the tracker, you’ll probably want to verify that it’s working and see the promised statistics. To do this, you need to create a custom segment in your Google Analytics report. Okay, that’s an easy one.

1. Go to “Advanced Segments” and click on “Create a new customised segment”.

2. Expand the “Visitors” list and drag the “User-Defined Value” block to the empty box on the right. Select “Matches exactly” from the Condition dropdown, and enter “Adblock Detected” in the Value field. Finally, enter a suitable name in the Name segment : field and click on Create Segment. You can also click the Test Segment button to check the segment before saving, but this will only work if the tracker has already been running for a while and has recorded some hits.

3. Now, go back to the Dashboard and select the newly created segment from the “Advanced Segments” box.

The result :

Note that you’ll probably need to wait for a few hours – or even a full day – before the new stats show up in Google Analytics.

Final Mini-Rant

On a more personal note, it appears that only 5% of my visitors use Adblock. I believe this hardly merits more than a “meh” reaction

And in case anyone cares about my stance on the ethics of ABP : at this point in time, there is no proper solution. Everybody wants free stuff. Everybody wants money. Consider the digital piracy debate – Adblock vs. Anti-Adblock is a more subtle manifestation of the same issues. And it doesn’t look like there’s a way out, not until we hit some kind of Utopia.

Here are some tips that I figured out from my experimentation.

Use section targeting

Section targeting allows you to emphasize or downplay certain parts of you page for the purpose of AdSense ad selection. This can improve your ad relevance and result in a higher CTR. To emphasize some content wrap it in these tags :

<!-- google_ad_section_start -->
...content here...
<!-- google_ad_section_end -->

To make the AdSense bot ignore some part of your content, use something like this :

<!-- google_ad_section_start(weight=ignore) -->
...ignored content here...
<!-- google_ad_section_end -->

For this blog I’ve edited the theme to add the “ignore” tags around my sidebar, page footer and most of the header. I also added inclusive tags to some individual posts. The Google AdSense Targeting plugin does some of this automatically, but I prefer to do it manually as it gives me greater control. Note that section targeting can take a while to come into effect.

Break up untargeted words

If you see that AdSense is picking up certain words in your posts and assigning them too much importance (like getting blog-related ads when the post is about widgets and only mentions blogging once), there’s another trick to try.

In addition to section targeting, you can also break up the untargeted words on your page by using the <span> tag. Example : “bl<span>og</span>”

The tags have no effect on how the page looks to a human visitor. On the other hand, the AdSense mediapartners bot will think “bl” and “og” are two separate words (so you might get ads about Ogs ).
Note that this might be considered a slightly “grayhat” technique by some.

Get advanced AdSense tracking

The statistics available in the AdSense account are very general and not that useful for advanced optimization. For one, per-URL and per-keyword CTR stats would help a lot in seeing which pages are performing well and which need to be improved.

Currently I’m using the AdSense Click Tracker plugin to track AdSense clicks with Google Analytics. This approach works, but it only catches around 40% of the clicks and I can’t seem to get certain data from GA reports (like the aforementioned per-URL CTR). So I’m still looking for a better tracking script.

Do what professionals do (not what they say)!

There are lots of AdSense e-books and video courses out there (some of them are pretty good), but what I’ve found really useful is examining how successful webmasters optimize their pages for AdSense. You can see real examples of how ads are integrated into the content, analyze AdSense block placement and so on, and, best of all, it’s free

Keep in mind that you need to pick the site to analyze carefully. Some highly successful websites are deliberately under-monetized, at least in terms of contextual ads, as they can afford it easily. I’d also suggest you look for websites with target audience similar to yours for better results.