What Are The Odds That Your Blog (Or Website) Will Fail?

Written On Day: 181

Ever since I created this blog I have wanted to know what my odds of success are. Out of 100 people that have a blog how many fail? After searching around the best I could find was 99%. You see it everywhere, 99% of all blogs fail.

This is an interesting number but the problem is that there is never any data to back it up. The number comes from nowhere. Maybe it is just intuition – the people that have been around a long time notice that most bloggers come and then they go. To them it seems like 99% of them disappear.

Intuition gone wrong is the cause of the current 50% divorce rate – intuition is not good enough. I would like to see some data to back up the claim. I can find out the chances of killing myself skiing, so why can’t I find out the chances of success with a blog?

Blogs Die Without A Trace

There is a good reason why there are no actual statistics for becoming a successful blogger. The problem is that blogs do not leave a body when they die – they disappear. They are like aliens in a computer game – once you shoot them with your photon laser they disappear. No trace remains, no clues that they were ever there, no dead bodies to count.

To find out my chances of dieing on a ski hill is easy. All I have to do is to go to the ski hill parking lot and wait till the end of the day. On a piece of paper I draw a line down the center. On the left side top part I write ‘Alive’ and on the top right side I write ‘Dead’. At the end of the day I make a mark on the left side of the page for everybody that leaves the hill with their skies and for the skiers that leave in a body bag I make a mark on the right side. By doing this everyday for the entire season I will have nice set of statistics. By adding up the marks on the left side and the the marks on the right side I can easily figure out how many people out of a hundred die skiing.

Unfortunately this kind of data gathering is hard to do with blogs. There is no way to count dead blogs. Looking around the internet there are only two types of blogs: successful blogs and those trying to be successful. The blogs that have died leave no trace, so there is no way to count them, there is no way to gather marks for the right side of the paper. All you see are the living blogs.

Problogger to the Rescue

Problogger is a very popular and successful blog. The owner, Darren Rowse, has been doling out advice on how to write a successful blog since 2005. His audience is thousands of eager bloggers trying to figure out how to make their own blog more successful.

Everyday he writes a few paragraphs of advice and tips on how to improve a blog. Most of the visitors are beginning bloggers – bloggers who just started their blog and want to learn how to make it better. These advice seeking bloggers consume his wisdom in the magnitude of approximately 15,000 visitors a day.

With this kind of traffic it is no surprise that each of his posts have lots of comments. People leave comments for many reasons: 1) To give praise. 2) To show disapproval. 3) To increase their own traffic. By leaving a comment the person also leaves a link back to their own website in the hopes that somebody will click on it and visit their site. This is great for them, but it is also great for us because they leave a foot print – a permanent record of their existence.

Problogger has an archive section on which is stored all the posts and comments since 2005. Anybody can go and look at the posts from 2005. The posts are there and so are the comments. BINGO!

We can go back in time. We can go to 2005 and have a look at the comments for each post. Each comment has a link back to the commenter’s website. In 2005 the commenter was a beginning blogger seeking Darren’s advice on how to make his blog more successful, full of hopes and dreams. Regardless of what his dreams were in 2005, now in 2009 we can find out if (s)he succeeded or not.

A Website That Estimates Traffic to a Website

Lets imagine you are sitting at the bar and next to you is a beautiful and sexy girl listening to you as you eagerly describe your blog. Her eyes are wide open with wonder as you talk about your blog. You are an honest guy: you tell her how much work it is and how slow it is going, how you have been trying to get more visitors but it is a ongoing struggle. She smiles and touches your knee – she seems to like you. Being an optimist you start to think about what you will make her for breakfast in the morning.

But then disaster strikes. In walks a fast talking, Fabio haired, blue eyed, muscles bulging from his tight shirt lady’s man. He has it all – he rarely leave the bar alone. He sees the girl next to you and makes his approach to her with a complete disregard of your feelings. He runs a hand through his bangs slowly, making sure the girl notices his camel back biceps. After introducing himself and making a lame joke he begins to talk to her about his blog. He claims that it has 10,000 visitors a day and that he makes $100/hour. The girl is impressed – you are screwed. Looks like it is going to be another night of checking your pathetic blog stats alone.

But all is not lost yet. Fortunately you never go to the bar without your laptop. You pull it out and go to the website webtraffic24.com. It is a website that provides an estimate of how many visitors a website gets. You simply type in a URL and click a button. webtraffic24.com will then check the page rank, alexa rank, backlinks, etc of the website and using an algorithm return an estimate of the amount of traffic the website receives. It is not perfect, it is just an estimate. But an estimate is usually all you need – you need to know whether a website gets around 10 visitors a day or 10,000.

After typing in the stupid brute’s blog address you get the results – 20 visitors a day. Just as you suspected – another bullshitter making false claims about their blog to pick up girls. Proudly you show the girl the results. Seeing the number she slaps the brute and turns back to you – ‘tell me more about your blog’.



Combining Problogger’s Archives with Webtraffic24.com to Count Dead Blogs

Using the URL left behind by the commentors in the problogger archive with the webtraffic24 tool we can count dead blogs. All we need to do is to go back to the old comments, for example in 2006, and using webtraffic24 see how the blogs are doing now. To simplify I have broken up the amount of daily visitors a website receives into 3 main states:

  • Dead (0-500 visitors/day)
  • Serious Injury (500-2000 visitors/day)
  • Alive and Well (2000+ visitors/day)

Webtraffic24 is not 100% accurate but it is good enough – it can separate the dead from the living.

How to Find Out the Success Rate of Blogs

To determine the success rate of blogs and websites I have done the following:

  1. From the problogger.com archived posts comment section I have taken all the URLs for the month of January 2006. There was a total of 2514 comments for that month.
  2. I had to do some editing:
    • When somebody put a link to a specific webpage I edited the URL to just the homepage. For example, if the the URL was www.mywebsite.com/my-dream-of-having-a-great-blog.html I changed it to www.mywebsite.com.
    • I excluded Darren Rowse’s comments.
    • I excluded all URL’s that were email addresses.
  3. After getting rid of all the duplicates (same person who made more then 1 comment that month) I was left with 693 distinct URLs. 693 eager bloggers and webmasters hoping to be successful one day.
  4. Using webtraffic24 I checked each URL to see where it stands today. Is is dead, injured or alive.

I gathered all the data into a single table (download table). Each URL and its statistics are recorded. With the table complete it is a simple matter of adding up the living and the dead.

website and blogs that have failed and succeeded


How Many Blogs Become Successful And How Many Die – The Results

If you were a commentor on problogger during January 2006, here are the odds of what state your blog will be in three years:

What percent websites and blog fail
Dead: 72%
Serious Injury: 12%
Alive and Well: 16%


72% chance of total failure. It is high but not as high as the 99% claimed by people who ‘just know’. Another way of looking at it is you have 28% of your blog not being dead in 3 years.

I was quite surprised by these results. They are much better then I expected. Even while working on this blog I was under the assumption that there is a 99% failure rate. I am glad that it is not that bad.

To answer my original question: 72 out of every 100 blogs die.

Of course this experiment has some flaws. The biggest one is that there is no way to tell how well these blogs where doing at the time they made the comment (January 2006). It is possible, although very unlikely, that all the Alive and Well blogs where alive and well in January 2006. The only real conclusion that can be made is that in 3 years 72% of the commentor’s blogs have died.

The second flaw is that problogger commentors do not represent the entire blogging sample. There is a lot of commentors but they do not represent the whole. It is possible that by being a problogger commentor you improve your blog and your chance of avoiding death. Maybe problogger commentors have a 72% chance of death but bloggers as a whole have 99%. Unfortunately there is no way to tell because there is no means to count dead bodies on the entire blog population. This experiment applies to all blogs only by assuming that problogger commentors are bloggers just like everybody else. But if you want the certainty of a 28% chance of avoiding death then become a problogger commentor.

More Then Dead

497 blogs are dead. Above I defined dead as having less then 500 visitors a day according to website24.com. Having anything less then 500 visitors a day for a website or blog is only virtually dead (you probably will not be making too much money) because even 3 visitors a day is still not dead. True dead is 0 visitors. 0 visitor happens when website24 cannot find the domain – which means the owner of that domain canceled the hosting and the domain is unused. Out of the 497 dead blogs 192 are truly dead – 28% of the total and 39% of the virtual dead.

Further Study

The data for this experiment was only 1 months worth of problogger comments. Of course it would be better to use a whole years worth of comment URLs. But, as you can probably imagine, that would be a lot of work. For 2006 problogger probably has over 30,000 comments. If you eliminate the duplicate URLs you might be left with at least 5000 URLs. I cannot imagine anybody wanting to collect all those URLs and then run them through website24. I would be glad to see it, but I am not going to do it.

What I might do in the future is to check how many blogs die within 6 months.  Instead of going back 3 years I will go back 6 months.


  1. July 1st, 2009 at 22:50 | #1

    Very well written and informative post. Thanks for the recognition as well, I’m glad to have been an inspiration for this post. Cheers mate and best of luck!

  2. July 2nd, 2009 at 14:53 | #2

    Great article! Your articles are easy to understand, very humourous in some way and very informative.
    Your writing style is unique. Never change it and never give up this blog.

    Great work!

  3. July 2nd, 2009 at 20:24 | #3

    Thanks Nobi. Now I just need 1937 more visitors like you so that I can resurrect this blog from the dead.

  4. July 7th, 2009 at 02:38 | #4

    Kudos to you Roman for introducing the Scientific Method to the world of blogging. Backing up your words with science is something that the blogosphere needs badly, because we make a lot of bold claims and fail to back it up. I admit, I’m guilty of this as well. So if anyone is equipped to call out other bloggers for not backing up what they claim, it’s you.

  5. July 11th, 2009 at 13:38 | #5

    Great post and great blog (just discovered it via websitebabble). Must have been tons of work to gather all the data manually. I like your statistical approach, too. However, I do have only one critical remark concerning your definition of dead, injured, alive and doing well based upon the number of visitors. There are many niche blogs which are not after maximizing number of visitors. And 500 visitors a day seems to be a very large number, anyway (at least for me). A lot of blogs (most?) are not after money either. If I would have that many visitors (I just started blogging and its just a hobby and way of learning some additional IT skills for me) I would be quite glad.

  6. roman
    July 11th, 2009 at 14:28 | #6

    Hello Randy,

    You are right, some people do not blog for money. So for them even 10 visitors a day can be considered successful. But since Problogger’s blog is about making money with your blog, it is safe to assume that most commentors on Problogger want to make money with their blog.

    Here is how I justify the three groups: dead, injured, alive: Most starting bloggers monetize their blogs by advertising. With advertising 500 visitors is not that much. If 1% of visitors click an ad and each ad pays $0.50 then all you would make a day is $25.00 (injured). At 2000 visitors a day you could make $100/day – almost enough to quit the average day job (alive). 200 visitors or less a day and your making $1 a day (dead).

  1. July 8th, 2009 at 07:41 | #1
  2. July 10th, 2009 at 16:54 | #2
  3. October 26th, 2009 at 10:10 | #3
You must be logged in to post a comment.