Of course every metric is crap, every ranking is crap and every top 10 list doesn’t deserve anybody’s attention.

Are they really bad or just bad when you don’t make it into the ranking?

I found a very useful online tool for all of you, twitter lovers. The service is called twitter analyzer and it is some sort of Web analytics software, but for twitter. Honestly, I don’t know if there is some sort of service similar to Twitter Analyzer so you’ll tell me if it’s not the best tool out there, but so far from what I’ve seen, it is pretty powerful.

My favorite feature is by far the “How many of your followers are online” which gives you, as you expect, how many of your followers are online at the moment. Pretty useful if you want to share a link or one of your post and want to maximize exposure. Of course, that’s only one of the many features of twitter analyzer. (more…)

Would you read the newspaper if there wasn’t any titles? Would you read the newspaper if there wasn’t any sections? would you read the newspaper if it was a single block of text? Would you read the newspaper if the presentation was different everyday? I don’t think so. Then, tell me why so many bloggers don’t structure their posts? (more…)

I am not going to make friends with that one for sure. As most of you know, I published an article last week about how you could artificially inflate or fake your Feedburner subscriber count and strongly advised you not to do it for obvious reasons. Now, do you think I would let you guys get away with it that easy? Of course not!

Wake up!

By exposing how easy it was to fake the subscriber count, I guess I wanted to say a little something to every advertiser out there: “Wake up!”. If you think this doesn’t apply to you because you’re not an advertiser you are wrong because the minute you comment on other blogs, you advertise your website. Advertising doesn’t mean paying money for an ad spot: to me it means having your brand displayed elsewhere. What I am about to say does apply to paid advertisement, but I’ll explain why it does apply to you also.

The main reason I think people are faking their RSS Subscriber count is to sell advertisements or reviews for more money. You know, some people will ask 25 or 30$ a month for an ad on their website simply because the feedburner chicklet shows 600 subscribers: it’s a common thing. Of course, you can check out some Alexa stats to make sure that number is real, but ShoeMoney has an Alexa Rank of 101,291 which makes me think Alexa is mostly crap. I also have another blog getting about 50 times more traffic than this blog with a crappy Alexa Rank as well. If we can’t really rely on Alexa Rank  – and I’ve talked about it in the past – it all comes down to the Feedburner subscriber count, right? Well it pisses me off because a lot of bloggers fake it and there is NO, absolutely NO return on investment when advertising on their blogs. It’s important for every advertiser out there to know what the real numbers are and I’m about to show you how (mostly).

I hope advertisers will read this because it will help them maximize their return on investment (ROI). The same applies to you, the average blogger with not so much money to invest in advertisement. I want everybody to make the wisest choice when it comes to advertisement and trusting the Feedburner chicklet is probably the worst decision of all.

I don’t pay for advertising, how does this apply to me?

You comment on blogs, you advertise. Whether or not you pay for it, every single time your link appear on a blog, it’s called advertising. Now, why does it matter? When you comment on blogs showing 600 subscribers, you’re actually making that blog a little more powerful because you’re advertisers more reasons to think the blog is active with real readers.

Can’t we use comments as a metric?

Unfortunately no, a lot of popular blogs out there don’t receive a lot of comments even if they have an impressive REAL number of subscribers. Take John Cow for example, he doesn’t receive 50 comments on every post, but he has a pretty good subscriber count and pretty good traffic. It’s a good indication, but we can’t use it as a very good metric.

Google Reader is the key

Not many people know that with Google Reader, you can get the number of subscribers subscribed via Google Reader for a particular feed! It doesn’t show you the real numbers because it only displays the number of subscribers subscribed using Google Reader, but this can give us a pretty good estimate. Where is that feature? It’s hidden a little bit, but when you select a particular feed in your Google reader, you will see a little “Show details” link and this link is the key! As an example, here’s are feed details for The Net Fool:

How can this help me get the real numbers?

Let’s do some maths here and compare the Google Reader count with the Feedburner chicklet on some blogs I have in my Google Reader:

I tried to include some well known sub 1000 blogs as well as some bigger blogs like Dosh Dosh, ShoeMoney and John Cow. There’s only 7 blogs in the list as I only took some blogs from my Google Reader and didn’t do an extensive search for feeds. While some might say it’s not that much and we can’t really get any statistics out of that, just check your Google Reader and you’ll notice it’s around 25% for a lot of blogs.  So you guys can basically do your own investigation, but you get the big picture and you know where I’m going with this.

What’s a “normal” ratio?

You’re the judge on that one, I don’t want to go out and say “Anybody under 10% is a cheater!”, I just want to raise a flag. I strongly encourage you to go through your list of feeds in Google Reader and check who looks suspicious. I unsubscribed to A LOT of feeds, especially in the MMO niche because of that and trust me, I know who’s faking it out there. I won’t point the finger at anyone as you’re all able to figure it out on your own.

I also want to point out that anybody with 300 Feedburner Subscribers and 15 Google Reader Subscribers is extremely suspicious. But as I said, you’re the judge.

Contests brings the ratio down

BS. While I agree contests are extremely popular and one of the requirements to get entries is to subscribe by email, I don’t think it would impact the count that much. I think most people will unsubscribe after the contest and subscribe to the feed via a reader if they like the blog. Even if people don’t unsubscribe, I’m not sure it can cause the ratio to go THAT down. The Net Fool ran a lot of contests and still has a pretty decent ratio. John Cow ran a lot of contests and has a pretty low ratio, but with 10,000 subscribers it really is a different game. We all know John Cow is getting traffic so there’s no question there.

I’m not saying ShoeMoney is cheating

Some might say, Shoe is at 5%, is he cheating? I don’t think so because if you check ShoeMoney, TechCrunch and big blogs like that the ratio is very low, mostly because the subscriptions come from so many different sources. For sub 1000 subscribers, it seems that the stats are pretty accurate. I also made some research online and it seems to be a well accepted fact that the Google Reader market Share is around 20-25% (That’s what it is on frogstr…)

Feedcompare.com

Another method Chris at chrisguthrie.net mentioned me is to actually have a look at feedcompare.com. It really is a nice website that let you have a look at how Feedburner feeds evolve into time. With it, you can actually see if there’s anything unusual on a feed if the feed in question has the “Awareness API” enabled. Mine isn’t enabled as I don’t like other websites and outside sources to have access to my feed without me knowing, but here is an example of a suspicious feed:

I’m not saying this guy is faking it, but going from 395 to 690 in two days is very suspicious…

You’re the judge

As I said, you’re the judge on this. I’m just saying you should be careful with the Feedburner count and don’t necessarily take it as a very good metric until you confirm the numbers by yourself. Google Reader is a good way to estimate the numbers and I honestly think 10-25% market share is a pretty accurate number. So, be careful when you buy advertisement, the Feedburner subscriber count shouldn’t always justify a higher price.

Prove me wrong

If you’re an honest blogger with 300 subscriptions and something like 15 Google Reader subscriptions It would be great if you could share some detailed stats. I would be more than happy to blog about it and say I was wrong!

Share this

I hope you guys will share this. Not because I want some link love, but because I want the community to know. I personally think the Google Reader subscriber count is a much better metric simply because Google Reader is the most popular reader out there and it’s a lot more harder to fake.

I was wondering if it was a good idea to publish that one and finally came to the conclusion that I had to share this so everyone can understand how easy it is to fake the feedburner subscriber’s count and win any subscriber contest, anytime…It’s ridiculous. I’m posting this more so that everyone realize that it’s important to trust the person you are competing against or you could end up losing for stupid reasons.

I’m not posting this for people to do it, go out there and win contests or whatever: It’s not worth it because it is a FAKE subscriber count and won’t give you more readers, so don’t think about more traffic. I’m posting this because I’m sure you’ve asked yourself in the past: how the hell can that website have 100, 200 or 500 subscribers? How the hell did that website beat me in that contest? How the hell did that website increased its subscriber count by 100 overnight? Well folks, it’s easy as hell and that’s why the RSS subscribers count can really be BS.

You will also find some packages on DigitalPoint like “500 subscribers for 50$”. You can do shit alone, no need to pay. In fact, you don’t even want to do it alone because there really are NO ADVANTAGES and here’s why. (more…)

What a day. I’m running a website with one of my friend and the project is going very very well.  Within three months, we managed to get a steady 500+ uniques everyday and it’s going up everyday. We almost reached 1,000 uniques today and we are quite happy about the results we are getting for not too much work! We recently (2 days ago) switched to a hostgator reseller account at 25$ a month for this project because the database is so big (100,000+ entries), performance and server load is now an issue. So we switched without any major issues and the awesome tech support team at hostgator helped us resolve some minor problems, but today, it went completely crazy!

Google Crawl rate

If you run a website updated very often, you will notice a new option in Google webmaster tools: you will be offered to accelerate the Google Bot crawl speed for your site! This is the exact option:

We’ve detected that Googlebot is limiting the rate at which it crawls pages on your site to ensure it doesn’t use too much of your server’s resources. If your server can handle additional Googlebot traffic, we recommend that you choose Faster below.

Faster A faster crawl will enable us to crawl your site quickly, but may put more load on your server.

How tempting is that? Being indexed faster, the dream of every webmaster! With that fresh new reseller account I decided to turn the faster crawl rate on.

What a bad idea

What a bad idea that was to turn the faster crawl rate on. Note that the website we run is very very heavy on resources, so what happened to my website may not happen to your website if you enable the option. It took a couple of hours before the Google Bot decided to crawl my site at top speed, but when it did, boom! No more website!

I checked my statcounter account around 5pm to notice no new visitors came to the website within the last 30 minutes. That’s really unusual when you get 500-1000 uniques a day, so I typed my domain name to see if there was anything wrong. The result:

500 – Internal server error

The evil 500 internal error! The error that tells you something bad has happened, but we don’t tell you what it is and there’s no way to find out! So I emailed HostGator and received an answer within a couple of minutes. The problem was that all 25 allowed processes were used, so no more request could come in. The rep killed the processes and guess what? 5 minutes later, same thing! Eventually, it went back up and I made sure to check the normal crawl rate.

Be careful

So be careful if you check that option. Make sure your server can handle the evil Google bot!