Thanks for your patience.

1. Forum Generator 2.0
2. WP Guardian 1.4
3. Referer Guardian 1.3

Forum Generator 2.0

Over the last 24 hours, the basic framework that will become Forum Generator 2.0 has been written. We’re going to continue with the same UI stylistically, but a lot of changes are happening under the hood. A massive overhaul of the gathering system is my biggest priority at the moment, along with adding support for various forum platforms other than phpBB3. While phpBB has a lot to offer, it is limited by a lack of basic plugin support. After spending some time working with the bbPress and vBulletin forum platforms more closely, this limitation is blatantly obvious. As development continues, my plan is to continue updating the blog with what I’m finding so that others who are developing plugins for forum platforms can gain from what will most likely be a major learning experience for me.

WP Guardian 1.4

I’m going to begin integrating some of the features requested in this post, as well as changes and bug-fixes we’ve discovered since the last release. One thing that I plan on working on is the UI. Specifically, I’d like to integrate some of the design features used in Referer Guardian to make it easier for users to assign links to advertisements and advertisements to posts, pages, and categories. Once the feature set of 1.4 becomes set in stone, I’ll update the blog.

Referer Guardian 1.1

The Referer Guardian launch was a success. We’ve received a handful of feature requests for the next version from customers. One big new feature will be the inclusion of basic statistics into the Referer Guardian control panel. This tracking isn’t intended to compete with any of the various analytics packages available, but will allow Referer Guardian users the ability to see how their campaigns are performing in real-time.

Other News

During non-development hours in the next week or so, I’m going to put together a thorough tutorial (or potentially series of tutorials) on utilizing WP Guardian and Referer Guardian together. This probably should have been done before launch, but the main focus at that point in time was testing. We are also requesting that customers of any of our products provide us feedback ASAP for any new features they’d like to see in the next releases of all three of our current products.

That is all for now. I hope you all are having a good December and looking forward to a great start to the new year.

• More adaptive to different permalink structures.
• You can now redirect through pages AND posts

Customers, please let me know if you do not receive an update email.

About half of the announcement list emails weren’t ever sent. Everyone expecting them should be receiving them within the next 10-15 minutes. Sorry for the inconvenience!

The long and tall of it is that we’re going to launch Referer Guardian at 2PM EST today. Fifteen hours later than we anticipated, but we need to get some sleep and no one likes reading email early in the morning anyway.

Thanks for understanding,
bhz

Just a quick update for everyone waiting for Referer Guardian to launch.

We made a great discovery yesterday during the final round of testing. We’ve tested and scrutinized the human generated clicks themselves extensively without ever having a leak, but had never seriously abused a Referer Guardian guarded iframe. So, a good friend was kind enough to lend us a 1×1 iframe on a site that gets hundreds of thousands of hits a day. We routed this iframe through Referer Guardian and to a tracking page where we recorded the referer as well as the header information (which carries browser information, among other things.) What better way to do a final test than to put it through some insane abuse?

The results were very, very good and we learned something that, as far as we know, no one else has ever discovered. After well over 100,000 iframe loads, we had a leak rate of ~0.03%. That means that for every 100,000 visitors that hit the iframe, we recorded 30 leaks. Now, leaks are a scary thing when you’re developing (or using) something to hide your referer, but this isn’t a surprise. Iframes are notoriously hard to deal with because each browser handles them a little bit differently.

When we began to analyze the logs, we saw a very clear pattern for the tiny amount of referers that did leak through; they were all (as in 100%) based on the Apple WebKit browser framework, which is used by both Safari and Chrome. What was even more strange was the fact that there were literally thousands of other users, using the same browsers and operating systems, whose referers were cloaked perfectly. Our first thought was security software, but the fact that a majority of these leaks were from iPhones and Mac users indicates otherwise.

We also noticed that almost all of these leaks happened in pairs in very close succession to one another. After putting some serious thought into the issue, what we’re almost positive is occurring is the result of pre-fetching and caching. Under certain, rare circumstances (such as an extremely fast refresh or back button hit) some browsers will load the end result of a series of redirections rather than follow the redirection path itself. We have been unable to recreate this event using the same browsers (including the iPhone) after many, many, many tries, but we’re almost certain this is the case.

Since we are unable to control browsers’ behavior at the system-level, we’ve added the option to block all WebKit-based browsers. The referer is checked all along the cloak path, so it isn’t possible for a browser that either sends no referer or an unexpected referer to actually pass through Referer Guardian directly. We added the option specifically for people who choose to iframe.

For the sake of comparison, we also tested the currently most popular referer cloaking plugin, CPA-Redirector, and found that a little more than 5% of referers passed through during the same iframe test with very little pattern to the browsers or operating systems of those leaks. This means that under a strenuous iframe test, CPA-Redirector was 167 times more likely to leak the referer than Referer Guardian.

How is that for extensive research?

<?php 
	if(isset($_GET['kw'])){
		$keyword = urlencode($_GET['kw']);
		echo 'Keyword Suggestions for: <strong>'.$keyword.'</strong>';
	} else {
		echo 'No Keyword Selected!';
		exit();
	}
	$url = 'http://google.com/complete/search?output=toolbar&q='.$keyword;
	$xml = simplexml_load_file($url);
	$data = $xml->xpath("/toplevel/CompleteSuggestion/suggestion");
	$queries = $xml->xpath("/toplevel/CompleteSuggestion/num_queries");
	$i = 0;
	$count = count($data)-1;
	while($i <= $count){
		echo $data[$i]['data']. ' - '. $queries[$i]['int'];
		echo "<br />";
		$i++;
	}
?>

Just save this to a .php file, upload, and visit http://path.to/the/file?kw={keyword} where {keyword} is the keyword or phrase you’re looking to get results for. With a little bit of PHP know-how, it should be very easy to extend this code to meet your needs.

Hope it helps!

As we mentioned earlier, WP Guardian 1.3 includes the following features and improvements:

• Category-based blocking
• Page-based Blocking
• Optional Close button
• Increase Range of Time Before Ad Pops

We’ve also included an experimental method to automatically add the wmode="transparent" parameter to all embedded videos. You’ll find instructions on how to unlock that feature in the updated instruction manual.

We’ve implemented the following improvements and features:

• Category-based blocking
• Page-based Blocking
• Optional Close button
• Increase Range of Time Before Ad Pops

If you have any questions about the 1.3 release, please comment here. We’ll be sending out an email blast and the updates as soon as we’re done finishing up documentation and the final round of testing.

I hope all of our US customers had a great Thanksgiving and that everyone is getting pumped up for the holiday season sales rush. From where we’re sitting, this season is looking to be a very profitable one.

A little less than a week ago, the popular dating site PlentyOfFish launched a beta self-serve advertising network. We decided to give the network a quick test to see how well their traffic performed in comparison to other social networks. While the result was disappointing, it is important to keep in mind that the network is still in beta, so they know there is room for improvement. Hopefully this post is seen as (mostly) constructive criticism and not an outright indictment.

Before I begin, I should also point out that, despite our namesake and all associated negative connotations, all ads were compliant with affiliate network standards and serious effort was put forth to comply with PoF’s ad network’s somewhat vague standards.

The Setup

Budget: ~$500.00
Tracking: Prosper202 (Congrats, Wes and Steven!)
Offers: Dating, self-improvement (teeth whitening and weight loss), grants/loans.
Campaigns: 10, 3-5 ads per campaign. All with a history of converting well on other ad networks in similar segments.

The Network

The sign-up process is simple. My account was approved and funded within 15 minutes. To avoid confusion, it is a total pain in the ass to actually find the network from the main PlentyOfFish domain. The address of the ad network is http://ads.pof.com/. The minimum network deposit is $25.00, which is entirely reasonable.

Once the account was setup and funded, I created my first campaign, promoting a dating network. PoF’s advertising model is bid-based CPM (cost-per-mille/1,000 impressions.) The minimum bid is $0.15 per thousand impressions. They don’t provide much information with respect to competition within your bid range or how much others bidding on similar segments are bidding. I’ll explain why this sort of blind bidding process can be a big pain in the ass later in the post.

A quick look at their advertising guidelines reads like a list of ever profitable campaign I’ve ever run. Their restrictions get particularly vague with regards to the promotion of other dating sites. No free dating sites are allowed. No dating sites that are considered “risque” are allowed. To me, passing some sort of high-handed moral judgment on how “adult” a lander for a site that is 18+ seems silly, but I understand wanting to control your network.

The segmentation, which is what originally attracted us to the network, has some serious potential. You can select the user segment that you’d like to advertise to based on the following criteria:

• Zip/Postal Code
• State/Province
• Country
• Age
• Gender
• Education Level
• Profession
• Has children
• Car
• Body Type
• Drinking Habits
• Ethnicity
• Height
• Income
• Login Count
• Marital Status
• Religion
• Search Type/Looking For
• Smoking Habits
• Session Depth

I’ve emphasized two separate segmentation options here; Login Count and Session Depth. In an ideal situation, these two segments could be used to target “engaged” users. Engagement is a sort of unquantifiable metric that helps advertisers gain insight on how likely a person is to click on an ad. The level of engagement is absolutely key in CPM advertising.

The Login Count segment is easy to understand. It allows you to choose to target users who have logged in at least a certain number of times. This can help eliminate users who may only be logging in for the first time and are more interested in checking out the dating and social networking aspects of the site, ignoring the ads entirely.

Session Depth may be a little more confusing to some. As I understand it, session depth reflects the number of pages or clicks a user has made on the site during their current session. Similar to Login count, this can help separate the people who are just logging on to check a message from those who are actively surfing the site.

One missing targeting segment that is available for their premium ad service is targeting based on interest and profile content. I made contact with PoF ad managers a few weeks ago to try and promote a client’s e-commerce business. They mentioned the self-serve option as an alternative. Without interest targeting, it is absolutely not.

I created a few campaigns, targeting to similar demographics that I have on Facebook and other social networks. PoF offers a few more metrics than some networks, so I had high hopes that this would keep my ad spend down. The key to successful CPM campaigns is knowing your target audience.

Because they don’t provide any real information as to the number of users an ad will reach within a segment, nor what a reasonable bid is for a particular segment, I ranged my bids from $0.15 (the minimum) to $0.50 to try to get a decent sample size while doing testing.

The Ad Review Process

It was during the ad review process that the first major shortcomings of the system became apparent. Ad approval times ranged from an hour (for updates to pre-existing campaigns) to five hours (for new campaigns.) They state in their FAQ section that approval can take up to 24 hours, because they need to make sure that your landers are compliant with their rules. IMHO, this delay is unacceptable.

My first few ads were rejected (after a few hours) for vague reasons. According to their FAQ, “If denied, an explanation will be provided to help you better design your ads in the future.” The reality of the situation was, “If denied, an explanation that doesn’t provide any real information will be posted to the History section of your account panel, but you won’t be notified, so keep refreshing, sucka!”

After a few failed attempts, I was able to stumble into a handful of landing page/ad text combos that met network standards. Forget about using even remotely piquant language. Even phrases that have made it passed US TV censors seem to be off-limits. This aspect wouldn’t be as frustrating if the exact reason an ad was denied was provided OR the review time wasn’t as long. Once again, it is important to note that the network is still in beta and these sorts of network management issues shouldn’t be too hard to iron out.

The Results

Average Impressions: 150,000
Average CTR: A gruesome 0.009%. I’d be embarrassed about this if similar ads targeted at similar segments hadn’t performed much, much better on other networks.
Average Conversion Rate: A very telling 0.005%. These same offers are converting at nearly 500x this rate on other ad networks.
Profit: A few hundred dollars in the red. I should note that I’ve had individual campaigns over the course of my career that failed miserably. Everyone does. That said, I’ve never had a string of campaigns fail like this, especially ones that are performing well on other networks.

At first, some of the campaigns seemed to be performing quite well. A few were seeing a reasonable amount of clicks and converting at 3-5% for the first hour or two of the campaign going live. However, almost across the board, CTR and conversion numbers declined rapidly after an ad had been live for 3-4 hours. The segments I was targeting for some campaigns were (purposefully) broad enough that ad overlap shouldn’t have been an issue on a network as large as PoF.

In a few cases, campaigns stopped being displayed altogether with no explanation. Increasing the bids made no difference whatsoever and, as I’ll describe in the next section, without a way to contact an ad manager, I never figured out how to get them to start again. Adding new campaigns and waiting for their (re-)approval seemed the only way to restart certain campaigns. There were no common trends across these “stopped” campaigns that I could see that might provide some sort of explanation.

User engagement was poor across the board. I combined both my own landers and direct linking and found that most users were only sticking around for a couple of seconds, including a handful of repeat visitors.

The Problems

1. Budgets and Account Replenishment On more than one occasion, campaigns went over-budget by 5-10%. Not too bad, but not too good either. My account also needed to be manually refunded after it sat at $0 for hours, even though I had setup automatic replenishment at $5.00.
2. Where are my ads? Some ads simply stopped being displayed. No reason given. According to the management panel, they were approved and active. Increasing bids didn’t seem to make a difference. Why?
3. Bid Surfing There is no way to tell a reasonable bid range. Blind bidding can work, but not when the ad delivery itself is questionable.
4. Help please?! I pride myself on not asking for too much help from ad account managers, so when I need it, I really need it. I couldn’t find any contact information listed on the actual PoF network site. I emailed various people who I’ve had contact with regarding their premium ad network and received absolutely no response. I understand that they are in beta still, but (unless you’re Google) you need to give your advertisers SOME WAY to get their questions answered. A vague FAQ isn’t going to cut it.

The Highlights

1. Promising segment targeting options They’ve made a solid attempt to introduce new/uncommon segments to advertisers which, in the long run, will probably prove powerful.
2. Low competition… for now at least. Many campaigns seemed to get a decent amount of impressions for low bids. Without knowing how accurate the ad delivery system is, it is hard to tell how competitive segments are.
3. Stats Their reporting seemed near real-time and offers a nice cross-section for doing impression and click distribution analysis on a campaign-to-campaign basis. I can see the potential for customizing this information to a higher degree being another big positive if they work out the other issues outlined here first.

Final Conclusion

There is a ton of potential here, but it is quite obvious that this is very much a beta and they really need to iron a few things out. Even with the “beta” label, the network doesn’t make up for the many shortcomings I’ve experienced during the small amount of time I’ve tried to advertise through the network. Much of this could be alleviated by offering some sort of advertiser support. I’d love to see a Campfire/chat system setup similar to the one used by Slicehost.

I’m also not convinced that the PoF user base is this disinterested in ads. Facebook is free for users and advertising on Facebook can be extremely lucrative if you know how to target the right audience. It will be interesting to see how PoF ads perform over the next few months. I’m certainly not writing them off just yet.

Your Thoughts?

I’m very interested to hear the thoughts and experiences of others that have used the network. If you have any thoughts at all on the subject, please feel free to comment away.