There is a problem: what do you do if you want to monetize your site and advertising companies you use rely on third party statistics that are not accurate? Alexa is known for being inaccurate (more to forecast trends than actual stats). What does a blogger/writer do if Alexa is showing inaccurate stats and the advertising company you use relies on Alexa and other third party services? Good question.

Since I use Text Link Ads on this site I decided to ask Patrick Gavin, President of Text Link Ads, how the company deals with issues like this. Seriously, I manage my site and the more information I have the better I can do it.

Tyme: Alexa obviously has accuracy issues – all of them do. Anything with a toolbar can be gamed. Technorati has their issues too. For example, if a site has a high number of inbound links because the blogger requested people to write a review of their site, which many feel is another way of “gaming” the system. Does TLA and Review me have other methods in place to ensure that sites are judged fairly? If so, can you share a little bit about how sites are judged/priced?

Patrick: We agree Alexa is not a perfect solution for gauging a website’s traffic. A perfect example is a blog devoted to Apple users whose user base will be heavy FireFox users not using the Alexa toolbar. Because of this the Alexa ranking for this blog will be disproportionately high compared to say a blog about Microsoft. We understand this so Alexa ranking is not the only criteria we use when pricing our publisher’s websites for our products.

We use a combination of: The website’s theme, location of the ads, number of ads to be sold, Alexa ranking, Technorati rank, and a measure of the website’s link popularity. We have an algorithm that prices our inventory based on these many criteria. The results are never perfect but by taking in six data points it can help even out some of the disparities that happen with Alexa data alone.

Six different data points – that makes me feel better but I can see where more metrics will have to be used to better screen sites initially because those things can be gamed. Getting your site accepted might be a challenge but honestly, it should be. If you’ve already been accepted, how do you make sure your site is getting the best advertising deals?

Text-Link Ads [linkbuildingblog.com no longer active] has a blog with a wealth of information on how to optimize your site and increase traffic. Subscribe to the feed. After going through the blog, do the one thing most people don’t do. Take some responsibility in the monetization of your site and work with the advertising company to get the most out of it. That’s the main point of this entry. Most people slap code on the sidebar and then complain on how it doesn’t perform. Or worse, move the ads around for a month and because there isn’t instant change become frustrated. I see my relationship with TLA like a partnership and I expect them to do their part and I have a responsibility as well. I actually have to manage my site and work with them for optimal success. If you are using a company that you can’t interact with, well…why would you do that?

I had a question, I asked, I received an answer. When I start optimizing this site I will (after looking over the blog) most likely have some more. I know one thing, I’m not going to throw code up blindly and pray it works.

Rick and I will be doing a podcast answering any remaining questions. Keep ‘em coming!

Why isn’t the blog used more to educate people? It seems like it’s mainly used for announcements.

It’s true that we use our blog mainly for announcements that are applicable across FeedBurner’s wide audience. But your point is well taken and there is a lot more we are doing on that front, and will continue to do. Check out recent posts from Dick, me, last fall’s look inside TechCrunch’s numbers, Brent’s bylined articles in Adotas, the hundreds of interviews we’ve done over the past few years with media publications as well as independent publishers, our Publishers Tips site, or last spring’s podcast market overview as a few examples of educational posts. In addition to our own blog, the publisher team (that’s my team) typically leaves 30-50 comments per week on _other_ blogs so that we’re answering questions people ask in near real time. That way, we don’t force people to come to us to get educated about how we work… if they ask, we’ll answer.

I want to know what happened with the Pageflakes thing. How was it resolved? What measures were put in place to stop it from happening again?

This question is referring to a bug where, in certain cases, they were over-reporting the number of Pageflakes subscribers to a given feed. In that case, we contacted Pageflakes, they acknowledged the bug, let us know when it was fixed, and we resumed reporting the Pageflakes data. While we obviously have no control over how web-based aggregators collect and/or report their data, our algorithms are often able to detect anomalies in the data we receive that allows us to proactively reach out to those services to verify the information we’re seeing. When we establish that things are out of whack, we disclose the issue (e.g. this example) and wait until things are resolved before resuming inclusion of their data in the overall reports.

Is it possible to export data and if not will it ever be?

Yes, click “export” from the “Analyze” tab. Data’s available in Excel or CSV format.

What are their privacy policies? The company I work for is starting a blog in Feb. The very first thing they would question is who has access to the stats and what precautions can be made to ensure privacy. I read on a blog that by not checking one thing stats could be pulled through the API.

Our privacy policy is here. By default, your data is private, and no one other than the account holder has access to the data. If you choose to make your data public (which you can do only by explicitly setting the Awareness API service to “public” – it is set to private by default), then anyone who follows the documented API guidelines can retrieve shared data (like circulation, hits, etc.). A good example of how this information can be used well is the podcast directory here. To reiterate, this information is *private* by default, and is only public if you choose to make it so.

I thought FeedBurner had a content network? Can’t find it from the front page.

Since different people mean different things by “content network”, I’m not sure how to answer this. We have FeedBurner networks available to publishers who want to aggregate their content – currently in beta, details are here.

As a blogger am I able to opt out of my site’s data being used for advertising purposes? Data collection?

By default, FeedBurner users are *not* in the ad network, so their data is not used for any advertising purposes. Since the primary reason bloggers use FeedBurner is to collect data about their feeds and sites, I’m not sure how to address a desire to opt out of data collection, so perhaps we need to do some follow-up on this.

How secure is the FeedBurner service for bloggers?

If by secure the questioner is asking about data security, see question/answer above about our privacy policy. We lock our doors at night, and all of our laptops have really secure passwords. If the question pertains to something specific or less sarcastic, let me know and I’ll be happy to address. Beyond that, I’m not sure what to say in terms of security, since the goal of this aspect of the service is to provide publicly available data in the form of RSS feeds.

For commercial use if we decide we do not want to use the service any longer what happens with the data collected from our visitors?

You may export your statistics (via the export feature described above) and delete your account. We do not keep any data once an account is removed from the service. Let me take a minute to address another aspect of leaving FeedBurner that wasn’t asked here but is often asked: specifically, what happens to your subscribers once you choose to leave FeedBurner. They’re your subscribers, we don’t want them after you leave. So we’ve gone to great lengths to ensure that your subscribers will follow you after you decide to leave (visit the link for more details).

Will performance be an issue if my site receives a large spike in traffic like Digg?

One of the many reasons publishers choose to use FeedBurner is to offload all bandwidth relating to their feed traffic to us. We manage more than 500,000 feeds, including thousands of feeds for 7 of the 10 top newspapers in the country, 3/4 of the Technorati 100 most popular blogs, and many thousands of high-traffic feeds. With more than a quarter billion feed requests per day, we maintain a high availability server infrastructure, and MySQL has used us as a case study (see this to show how MySQL can be used to power high demand environments like ours). We consistently keep our server farms at fractions of their total capacity, and our connectivity is structured in a way that were we ever to need to spike well above the hundreds of megabits of sustained connectivity we currently use, we could do so without any impact on overall performance.

Which is a long way of saying, “No.”

Why does the FeedBurner count increase dramatically after I am Dugg, then goes down a couple of days later? Did I lose all those visitors?

This is primarily an artefact of prior versions of Firefox (1.5.x and prior), where requests to the feed are labeled the same whether coming from the browser or the feedreader (LiveBookmarks). As a result, we would have to infer from the traffic patterns (i.e., polling intervals, IP addresses, identical user agents) to determine whether the access to the feed was a one-off (i.e., browser) or a repetitive request (i.e., the feed reader). Because of the high concentration of traffic that a site receives when Dugg, those inferences are often impossible to get 100% correct – so some one-off accesses end up looking like repeated requests, and our algorithms count them as subscribers instead of browser views. Recently, the Mozilla team added a unique user agent for the feed reader in Firefox 2.0 (per FeedBurner’s request), which means that these anomalies should be far fewer from now on.

What do they think needs improvement? Can they talk about future plans? They also seem like they are all over the place. What do they DO? Is it feeds, advertising, what is it?

Wow. Let’s see: what needs improvement? Everything. We want the web audience at large to be more comfortable with the notion of subscribing to media feeds. We want publishers to get more value from their feed content by adding more functionality. We want to see creative developers build more cool stuff using our open APIs. We want to sell more ads, so our publishers can make more money. The radiators in our office are maddeningly spotty, so we never know when to dress warm, so that’d be good to fix while we’re improving things.

As far as futures, we’d like to see the Bears win the Super Bowl. A Cubs/Sox World Series would be totally cool, too. On a more serious note, we’ve recently stated that further integration between our feed and site stats is in the near future, as well as a platform for measuring all aspects of content consumption: in a feed, on a site, in a widget, or via some new thing that doesn’t have a buzzword attached to it. However content is consumed, we’ll measure (and monetize) it. Expect to see a lot from us on this in 2007.

Regarding the ‘all over the place’ comment, our goal is simply this: FeedBurner aims to be the world’s foremost provider of online media distribution and audience engagement services. We want to help content publishers, be they independent bloggers or podcasters, blog networks, or commercial publishers – to distribute their media further, engage and build their audience and ultimately, make money. We think we’re in the best position to understand, measure and enhance the distribution of content – via feeds, resyndication, widgets, and who knows what comes next – and then, for publishers who want to capitalize on that understanding, we offer a monetization platform with our advertising network.

Rick and I will be doing a podcast answering any remaining questions. Keep ‘em coming!

Why isn’t the blog used more to educate people? It seems like it’s mainly used for announcements.

It’s true that we use our blog mainly for announcements that are applicable across FeedBurner’s wide audience. But your point is well taken and there is a lot more we are doing on that front, and will continue to do. Check out recent posts from Dick, me, last fall’s look inside TechCrunch’s numbers, Brent’s bylined articles in Adotas, the hundreds of interviews we’ve done over the past few years with media publications as well as independent publishers, our Publishers Tips site, or last spring’s podcast market overview as a few examples of educational posts. In addition to our own blog, the publisher team (that’s my team) typically leaves 30-50 comments per week on _other_ blogs so that we’re answering questions people ask in near real time. That way, we don’t force people to come to us to get educated about how we work… if they ask, we’ll answer.

I want to know what happened with the Pageflakes thing. How was it resolved? What measures were put in place to stop it from happening again?

This question is referring to a bug where, in certain cases, they were over-reporting the number of Pageflakes subscribers to a given feed. In that case, we contacted Pageflakes, they acknowledged the bug, let us know when it was fixed, and we resumed reporting the Pageflakes data. While we obviously have no control over how web-based aggregators collect and/or report their data, our algorithms are often able to detect anomalies in the data we receive that allows us to proactively reach out to those services to verify the information we’re seeing. When we establish that things are out of whack, we disclose the issue (e.g. this example) and wait until things are resolved before resuming inclusion of their data in the overall reports.

Is it possible to export data and if not will it ever be?

Yes, click “export” from the “Analyze” tab. Data’s available in Excel or CSV format.

What are their privacy policies? The company I work for is starting a blog in Feb. The very first thing they would question is who has access to the stats and what precautions can be made to ensure privacy. I read on a blog that by not checking one thing stats could be pulled through the API.

Our privacy policy is here. By default, your data is private, and no one other than the account holder has access to the data. If you choose to make your data public (which you can do only by explicitly setting the Awareness API service to “public” – it is set to private by default), then anyone who follows the documented API guidelines can retrieve shared data (like circulation, hits, etc.). A good example of how this information can be used well is the podcast directory here. To reiterate, this information is *private* by default, and is only public if you choose to make it so.

I thought FeedBurner had a content network? Can’t find it from the front page.

Since different people mean different things by “content network”, I’m not sure how to answer this. We have FeedBurner networks available to publishers who want to aggregate their content – currently in beta, details are here.

As a blogger am I able to opt out of my site’s data being used for advertising purposes? Data collection?

By default, FeedBurner users are *not* in the ad network, so their data is not used for any advertising purposes. Since the primary reason bloggers use FeedBurner is to collect data about their feeds and sites, I’m not sure how to address a desire to opt out of data collection, so perhaps we need to do some follow-up on this.

How secure is the FeedBurner service for bloggers?

If by secure the questioner is asking about data security, see question/answer above about our privacy policy. We lock our doors at night, and all of our laptops have really secure passwords. If the question pertains to something specific or less sarcastic, let me know and I’ll be happy to address. Beyond that, I’m not sure what to say in terms of security, since the goal of this aspect of the service is to provide publicly available data in the form of RSS feeds.

For commercial use if we decide we do not want to use the service any longer what happens with the data collected from our visitors?

You may export your statistics (via the export feature described above) and delete your account. We do not keep any data once an account is removed from the service. Let me take a minute to address another aspect of leaving FeedBurner that wasn’t asked here but is often asked: specifically, what happens to your subscribers once you choose to leave FeedBurner. They’re your subscribers, we don’t want them after you leave. So we’ve gone to great lengths to ensure that your subscribers will follow you after you decide to leave (visit the link for more details).

Will performance be an issue if my site receives a large spike in traffic like Digg?

One of the many reasons publishers choose to use FeedBurner is to offload all bandwidth relating to their feed traffic to us. We manage more than 500,000 feeds, including thousands of feeds for 7 of the 10 top newspapers in the country, 3/4 of the Technorati 100 most popular blogs, and many thousands of high-traffic feeds. With more than a quarter billion feed requests per day, we maintain a high availability server infrastructure, and MySQL has used us as a case study (see this to show how MySQL can be used to power high demand environments like ours). We consistently keep our server farms at fractions of their total capacity, and our connectivity is structured in a way that were we ever to need to spike well above the hundreds of megabits of sustained connectivity we currently use, we could do so without any impact on overall performance.

Which is a long way of saying, “No.”

Why does the FeedBurner count increase dramatically after I am Dugg, then goes down a couple of days later? Did I lose all those visitors?

This is primarily an artefact of prior versions of Firefox (1.5.x and prior), where requests to the feed are labeled the same whether coming from the browser or the feedreader (LiveBookmarks). As a result, we would have to infer from the traffic patterns (i.e., polling intervals, IP addresses, identical user agents) to determine whether the access to the feed was a one-off (i.e., browser) or a repetitive request (i.e., the feed reader). Because of the high concentration of traffic that a site receives when Dugg, those inferences are often impossible to get 100% correct – so some one-off accesses end up looking like repeated requests, and our algorithms count them as subscribers instead of browser views. Recently, the Mozilla team added a unique user agent for the feed reader in Firefox 2.0 (per FeedBurner’s request), which means that these anomalies should be far fewer from now on.

What do they think needs improvement? Can they talk about future plans? They also seem like they are all over the place. What do they DO? Is it feeds, advertising, what is it?

Wow. Let’s see: what needs improvement? Everything. We want the web audience at large to be more comfortable with the notion of subscribing to media feeds. We want publishers to get more value from their feed content by adding more functionality. We want to see creative developers build more cool stuff using our open APIs. We want to sell more ads, so our publishers can make more money. The radiators in our office are maddeningly spotty, so we never know when to dress warm, so that’d be good to fix while we’re improving things.

As far as futures, we’d like to see the Bears win the Super Bowl. A Cubs/Sox World Series would be totally cool, too. On a more serious note, we’ve recently stated that further integration between our feed and site stats is in the near future, as well as a platform for measuring all aspects of content consumption: in a feed, on a site, in a widget, or via some new thing that doesn’t have a buzzword attached to it. However content is consumed, we’ll measure (and monetize) it. Expect to see a lot from us on this in 2007.

Regarding the ‘all over the place’ comment, our goal is simply this: FeedBurner aims to be the world’s foremost provider of online media distribution and audience engagement services. We want to help content publishers, be they independent bloggers or podcasters, blog networks, or commercial publishers – to distribute their media further, engage and build their audience and ultimately, make money. We think we’re in the best position to understand, measure and enhance the distribution of content – via feeds, resyndication, widgets, and who knows what comes next – and then, for publishers who want to capitalize on that understanding, we offer a monetization platform with our advertising network.

The main problem is that Microsoft is good at competing but not good at caring for customers. The focus of the developer division at Microsoft is the .NET Framework and related technologies which is primarily a competitor to Java/JVM and related technologies. However when it comes to areas where there isn’t a strong, single competitor that can be focused on (e.g. RAD development, scripting languages, web application development) we tend to flounder and stagnate. Eventually I’m sure customer pressure will get us of our butts, it’s just unfortunate that we have to be forced to do these things instead of doing them right the first time around.

This is very true. I’ve written about Microsoft products in one form or another for seven years now. At one time Microsoft was a company I truly admired because they seemed to “get it”. Not any more. If there isn’t anyone to compete with, Microsoft drops the ball.

Let’s look at an example…there was a post on the Internet Explorer blog that IE7 will have tabs. It has to have tabs, this shouldn’t be news. The only reason why IE7 is going to be released prior to Longhorn is because of Firefox. Millions of people had to switch browsers before Microsoft woke up and released that we (as users) realized we could do better.

That is what saddens me about Microsoft. When I used to talk to Microsoft developers you could hear the enthusiasm in their voices. The product features showed their enthusiasm. Microsoft is only as good as their employees – that is true for any company. When I read the IE7 tab post, I realized that something is missing. Is it that the developers “get it” but are bottlenecked somehow and are unable to produce? Is it that there isn’t a true understanding of how bad IE6 is compared to Firefox? IE7 needs to be better than Firefox for me to switch back…get it? That’s how Firefox users feel. We’re happy using Firefox – Microsoft is going to have to give us a reason to come back.

Don’t get me wrong, I’m not anti-Microsoft and I think I am being fair. I was extremely excited when news on Xbox 360 was released. I was giddy with delight when I saw some of the games I will be playing on my Xbox 360. I wasn’t swayed by the fact that, spec wise, PS3 trumps Xbox 360. I realize one important thing – it should. It’s coming out six months after Xbox 360, it’s supposed to be better. It has to be better – Sony gets it.

Microsoft needs to get it.

Why is the default “no image” image a male? If I state that I am a female (you must specify a gender), why didn’t they have enough forethought to have a female default “no image” image? See what I mean about defaulting to males? Why not connect Yahoo 360 with Yahoo Avatars? There are a alot of people who wouldn’t want to put a picture up but would show their avatar.

And why doesn’t Yahoo Avatars and LaunchCast support Firefox? When I have to switch browsers to use a service, I don’t use it.

Bottom line? Try it for yourself. If you need an invite let me know and I’ll send you one.