The Truth About Afrigator Rankings

A recent temporary change in algorithm for blogs ranking by Africa’s most authoritative blogs aggregator attracted a lot of debate and suggestions. After complaints by bloggers, Afrigator soon made a formal announcement via their blog: Blog ranking changes

Afrigator
Afrigator

A lot of bloggers have noticed that there has been quite dramatic changes in the ranking of blogs. This is due to the fact that we had to make a change to our algorithm. We removed the links from blog posts and links from blogs from the ranking algorithm. This is only a temporary change until such time as we’ve optimized the link checking.So, while you may be on top of the world at the moment, this may change again soon. We still believe that our Afrigator ranking algorithm is a good representation of a blog’s authority, thus the change is only temporary. We’ll let you know when it is back to normal.

For now, traffic and only traffic counts!

The effect was profound. Several low-quality blogs found their way to the top of the rankings. Some new blogs with only traffic, quickly found their way to the top. Find some reactions from some bloggers across Africa, at varying times regarding the rank changes.

Diary of a gay Kenyan wrote: Afrigator blog rankings taken with a pinch of salt

I noticed this morning that I’d been propelled to the cabinet of Kenyan blogs at number 7 according to Afrgator.com (and 156 in Africa)! Surely that would make me the equivalent of Minister for Water and Irrigation in the government so thank you dear readers and followers for the time you take to drop by, we do have a good laugh.

The Trials & Tribulations of a Freshly-Arrived Denizen…of Ghana wrote: The trouble with Afrigator.com!

He goes on to do an analysis of some 5 blogs he feels do not qualify to be listed in Ghana’s top 20 blogs due to what he perceives as their poor quality and even questions what qualifies as a Ghanaian blog. He concludes on this note:

I don’t know what monitoring and evaluating afrigator is doing about blogs under countries, but it strikes me that it might need to do some significant revision of what “constitutes” Ghanaian blogs. Is it merely a cut-and-paste job from papers about Ghana, or entries about Ghana or on Ghanaian life?

Gitui.com whilst explaining why bloggers should follow other blogs, also made a mention of the Afrigator ranking changes:

Lets face it, a blog without visitors is dead. And with everyone out there wanting a piece of the pie its important to keep up to date with trends, news and anything that will affect your blogs. Today I found out that Afrigator has changed their blog ranking algorithm, and so my blog will be affected. If the news were to be quite devastating, I can still have a chance to redeem my blog before its too late.

Stii Pretorious is the Chief Technical Officer (CTO) at Afrigator. Therefore, his words in this matters carry a lot of weight. Whilst reacting to concerns expressed by users, he wrote:

Just a note on the linking. Our reason behind this was that if you write good posts and your peers link to you, it is an indication of authority. If you get craploads of traffic of which most is bounced visits, it does not mean that you’re very authoritative and it means the system can easily be fooled into thinking you’re popular. We’re trying to avoid that by taking a number of things into account to get the best mix.

A detailed break-down ofΒ  what’s in a Afrigator ranking is available in his personal blog.

In summary, Afrigator uses the following indicators to achieve a relatively balanced ranking system:

  1. Unique visits: the number of unique visitors to a blog, within a particular period of time
  2. Page views: the number of pages visitors check on each particular blog
  3. Link from other blogs: the number of blogs linking to a particular blog are counted
  4. Links from blog posts: only links from blogs aggregated by Afrigator are counted. This makes this particular indicator somewhat skewed.

An average score of the 4 indicators is eventually used to determine the top African blogs.

Frequency of posts: In recent times, Afrigator has also added a 5th point. Only blogs that have posted within a 30 day period are ranked. Thus, blogs that have not published any content in the last 30 days will not appear in the rankings.

The Afrigator team is rightfully open to feedback from users and seems to appreciate suggestions offered to them directly.

Any lessons learnt from the ranking dancing?

  1. There should be multiple factors considered in ranking what constitutes a top blog.
  2. Do not offend an army of bloggers. Should they decide to go town on your matter, it can attract a lot of attention and probably backlash or praises from the public, depending on what the matter is.

At this time, my personal blog has gone back to where it had always been: Number 2 in Ghana, and hovers among the top 20 in Africa. Fair enough. πŸ™‚

Previous articleMTN partners with IMImobile in a major deal to bring mobile and online content to 103 million users
Next articleEffective Blogging: The Importance of Blog Traffic Statistics
Oluniyi D. Ajao
Oluniyi D. Ajao is an Internet Entrepreneur and Tech Enthusiast based in South Africa. Follow him on twitter @niyyie for more tech updates.

26 Comments

  1. One point I disagreed the most with is the fact that the blogs that link to you must also be on Afrigator. To me this is an unfair strategy to sideline blogs that have links from sites outside Afrigator. I just think it is a way of getting more blogs unto Afrigator, to me this is not the right way to get numbers. Why cant they just go the Google way???

    • Edward, you miss a critical point. Afrigator is African and by extension, its rankings should be “African”. If you want a global blog ranking, simply head to technorati.com

      I understand that the restriction on blogs covered by Afrigator makes results from “links from blog posts” a bit skewed but I agree with the principle of limiting the links to African blogs only. That is the authentic way to reflect an African ranking system – in my view.

      • Lol. Now I see why I rank so low an Afrigator :-(. All, and I mean all my link backs are from OUTSIDE of Africa. Not a single one of them is African. That’s not all that fair to me. If the subject matter I’m writing about is not that popular in Africa, I doubt if I’m going to get any blogs in the same niche let alone link back to me.

  2. I cannot help but chuckle when I read posts on the Afrigator rankings. Call me naive but I simply fail to grasp the concept of being overly concerned with where one places on a ranking of blogs across Africa. Are bloggers out there blogging for the sole purpose of being ranked highly by Afrigator? Is a high ranking an affirmation that what one is blogging about has some sort of impact factor? Some weeks ago EKB pointed out that Ato Kwamena Dadzie’s blog was on top of the Afrigator ranking but now appears to have disappeared. Yet we all know that Ato’s blog is probably the most well-known and widely read blog in Ghana! So do Afrigator rankings actually matter?

    • Afrigator rankings matter – to some extent. It tells of the authority your blog commands and also helps boost the ego of the blogger(s) behind that blog.

      Ato’s rapid rise was during the period I am referring to here. The disappearance could be that he could not face the true rank of his blog when the folks at Afrigator went back to the original algorithm they used in ranking African blogs considering that he had bragged a bit. πŸ˜‰

      The folks at Afrigator can add some more “sugar” to what I have stated.

  3. David, great post – really well written and I think captures the essence of what’s been going on with the rankings of late. I applaud you for this!

    I can certainly see by the comments that there are mixed reactions to the rankings and to be honest – there always will be. I must say that we know the ranking is not fool-proof and it probably could be done differently but we do have a reason for this.

    If you look at other aggregators who rank websites (be it blog or other sites) they generally rank these sites by traffic alone. While this is certainly an important factor we felt that when it comes to blogs there are other factors that need to be considered which is why we introduced all those other variables into the equation.

    If someone links to your blog or something you’ve written it essentially means that you’ve written something that has had an impact – even if it’s just for that person. Now the problem with other sites is that these things aren’t factored into rankings.

    While there certainly are other ways to do this, we feel that this is the best way to rank blogs in Africa. The ranking algorithm has adapted and changed considerably to when we started and I can assure you it will adapt and change in the future as well. We are continuously trying to be as relevant as possible and I hope that with your guys feedback and assistance we can achieve this.

    One thing I’d like to say to Edward is that we don’t count links outside of Afrigator for two reasons. The first being that we want to track links within Africa as per David’s comment and the second is a scale issue. A site like Technorati has taken the Google approach but we unfortunately don’t yet have the capability to aggregate the entire Internet. We will get there at some point and then we’ll certainly try and include links outside of Afrigator.

    Hope this helps?

  4. Great posts and interesting comments. The Ato K.D. scenario is an interesting one. I have one question: what is the relative weight of each of the above criteria in Afrigator’s ranking system? Many people will agree with me that Ato K.D.’s blog deserves to be ranked higher that it’s current position. That still leaves me indecisive as to how fair the system is.

    • Blog ranking is scientific and not emotional. You have to work your way to do top by excelling in all the listed criteria.

      From what I see of Afrigator, they give all the criteria equal points.

      What is the current ranking of Ato K.D’s blog? I can’t seem to find it in there at all. πŸ˜‰

      • I have found the info about Ato KD’s current ranking. It has dropped to the 43rd position in Ghana:

        Blog Rank:

        Ato Kwamena Dadzie ranks 2460 in Africa and 43 in Ghana. According to site visitors it ranks 49, and 34 according to page views:

        According to blogroll links it ranks 11165 and 11165 according to the amount of links within Afrigator blog posts.

        From the above information, Ato KD’s blog rank very well with site visitors and page views, but rank very poorly with blogroll links and links from blog posts. The average is what you see as his final ranking.

  5. Blog ranking should definitely not be a subjective matter. The discussion, I gather, is about fashioning out a fair ranking system for African blogs, and that leaves my questions valid. A blog that draws lots of visitors, who are active, on a consistent basis deserves to be right up there. In Ato K.D.’s case, I wonder how poorly he’s performing in the other departments, resulting in the 49th position. Or maybe you should give him some link love? I have no doubt that Afrigator will improve with time. Fingers crossed.

  6. The nature of ato kd’s blog is that most of his readers do not keep blogs and so will not link to him.

    On the other hand david’s blog is very technical and will have techie type readers who will link back.

    While ato kd does some sort of baiting to get comments on his story, david posts entries that link-bait.

    Either way, the only thing that matters is that each blog stays within the acceptable ranking rules.

    So long as someone doesnt set up multiple blogs that link between each other, there should be no problem

    • One thing I love about Afrigator’s algorithm is that, it is very hard to rig. It is hard for anyone, to manipulate ALL the factors simultaneously and artificially rise to the top. Traffic can be bought but how do you buy links from over 4,000 African blogs?

      The art of blogging includes linking. Blogging goes beyond just posting articles. A good blogger links to other bloggers, and recieve links from other blogs as well. By all means, Afrigator must consider all the important factors in the blogging business, to offer a credible ranking list.

      Thanks FAF, for the simple illustration.

  7. The problem I have with rankings and algorithms is that they are skewed and output what the author of the algorithm desires.
    Stii mentions that a huge part of authority is when peers link to you.
    Although I see that in certain cases this might be true, but in most it is not.

    I write for people not my peers. I write for those who do not know not for those who already know.

    Take for example a food blog. Where the majority of the readers and subscribers and even linkers would probably not be peers, ie chefs and other cooks, but mostly you and I and for the most part our wives.
    BEcause professional cooks do not link to this blog, does it mean that it is a poor blog. You see the delema.

    The other thing is that the internet is global, many blogs have readers and subscribers from around the glob which Afrigator does not track.

    A better qualification would be a Afrigator Ranking within Afrigator.

    By blog is ranked by different methods they do not agree with each other. Because they all have a different skew, they rank from a different view point and put a different emphasis on different areas.

    One should take such rankings with a piece of salt. The only true ranking is your readers.

    • The folks at Afrigator seem to be always open to new ideas that can make their rankings better. I disagree with “pinch of salt” because I am of the view that it should be possible to somewhat scientifically calibrate your blogging success.

      Afrigator is “a” ranking and not “the” ranking. As different ranking websites have different methods of calculation, it is important to put each and every of them into the proper context.

      Amatomu for example, ranks only by traffic and is restricted strictly to South African blogs only. What do we now say about that?

      I agree 100% that the best ranking is the impact your blog has on its readers but the more readers you have a positive impact on, the better. All Afrigator is seeking to do, is publicise an idea of how many readers you are having an impact on.

      Feel free to reply me again.

  8. Great post!

    What would bloggers do without their rankings…? I wonder what all this means in the final analysis. Ranking is yet to translate into anything meaningful to bloggers, particular those on the side of the globe — well, beyond the ego trip of course– since Afrigator and co started publishing figures.

    There is all lot of factors that go into blogging, and the relevance of a blog, that I think the emergence of, and over-dependence on, ranking metrics may make mockery of the whole experience all together.

    David, looking at the first comment you made in response to Edward, don’t you think it’s a bit of fuzzy logic to state that links from “non-Africa” blogs is irrelevant in an African-centered blog ranking system?

    I thought we now live in a global village — thanks to the emergence of the Internet, social media and computers — and the gap between Africa and the outside whole is closing, yet it does not matter to have non-African links factored into ranking alogorithm? This does not add up much in my head. Besides, what makes a blog “African”?

    Cheers!

    • πŸ™‚ Did you read the response from the MD of Afrigator, Justin Hartman?
      Did you read my response to Abena where I started with:

      Afrigator rankings matter – to some extent. It tells of the authority your blog commands and also helps boost the ego of the blogger(s) behind that blog.

      Now, let’s not be in a hurry to deceive ourselves about the global village concept. There is a wide gap between Africa and the rest of the world. To think we are really connected to the world in the way the westerners have in mind is a fantasy. For now, let Africa grow out of the depth of famine, poverty, poor infrastructure etc. Then, perhaps…

      Also, even if Africa was really part of any global village, what is the harm in celebrating a regional block within that global village?

      My definition: A blog is “African” when it is written by an African, or [at least, substantially] about Africa, or by blogger[s] in Africa.

  9. Hey guys – me again! This is turning into a FANTASTIC debate and one I’m enjoying considerably. In particular I loved the point about a Tech blogger being able to attract link backs more so than someone who writes about food. This is extremely relevant and something we have to consider.

    I was chatting to Stii now and one thing we could look at doing is using comment count on posts as a basis for the algorithm. The problem with using comment count is that it changes all the time so we have to think of a good way to do something like this.

    For example, let’s look at this post. Not sure many people have linked directly to this post but with 19 comments already that has to count for something. The point with this post is that it’s spurred a great debate and therefor should be weighted accordingly.

    Again, I’m not sure of the feasibility of it but it’s certainly worth exploring. There is 2 more things I’d like to add:

    1) Ranking, but its very nature, isn’t black and white – it’s more grey. There isn’t a right way to do it and from our perspective as long as everyone is ranked by the same methods then surely it has to be fair?

    2) This is an important ranking and that’s why we want to get it 100% right. We’re not just tracking Afrigator stats, we do track all the stats from all the readers and subscribers from all over the world – it’s just the links that are restricted to Afrigator users. There simply isn’t another type of ranking system like Afrigator’s which is why we feel it is actually authoritative. Sure, it needs more work and more thought but no one is able to rank blogs from Africa.

    • Hi Justin, I have a lot respect for what you and your team are trying to do, which is really difficult to accomplish, scientifically.

      I wish there is some form of analytical test to use to statistically access the validity and reliability of the possible algorithms that can be constructed for this ranking business. Obviously, the more the variables used the better and more valid the ranking becomes. But without some measure of fidelity testing, it’s like shooting in the dark.

      Also, one major limitation of rankings is that is blogger-dependent. It is a fact that bloggers are most likely to read and comment on other blogs. Of course the use of links is also flawed for this reason. A non-blogger can’t link back, but may read, and most likely will not leave any comment. With these limitations, it is often misleading to generalize ranking figures without making clear the various caveats associated with the numbers.

      @ David:
      I hope you know where I stand regarding celebrating African social media πŸ™‚

      I feel that it does more harm than good to be have a ranking system that disregard the ‘Diaspora-nature’ of Africans and their diverse audiences.

      • Diaspora what? My friend pack your luggage and come claim your own square mile in Africa. There are things you enjoy where you’re domiciled. Enjoy those. If you want to enjoy what we do in Africa, move back home. Until then…

        Seriously, I see your point. A LOT of diasporan Africans are online, proudly flying the African flag and deserve a slice of the Gator pie.The folks at Afrigator are reading.

  10. “Several low-quality blogs found their way to the top of the rankings. Some new blogs with only traffic, quickly found their way to the top.”

    I think these statements collate far too many issues and reduce blogging to being part of a clique.

    When you say ‘low-quality’ here you imply that those that had been downgraded were ‘high-quality’. But high quality according to who and on the basis of what parameters. Clearly what your article proves, and the responses of your commentators suggest is that ranking is a very relative concept.

    In your view then, if a blog is new but has one million readers who do not have blogs to link to then it is of a lower quality than the one that has one hundred readers who link to each and everyone of its posts. That regardless of the fact that the former has original content and the other publishes links. This is as flawed as the tendency to judge the popularity of a blog on the basis of comments.

    On the other hand, popularity does not, of necessity speak to the quality of a blog. Quality can only be judged by the consumer, bearing in mind that certain blogs, as other products, are successful because they serve niche markets.

    Purpose is also key. I started blogging because I wanted to be a writer. Excerpts from my blog have appeared in a literary journal in Kenya, translated by an Italian journal, republished by a Nigerian magazine and taught in an American university. My blog jump started my commercial writing career and is now also syndicated and paid for by a kenyan internet provider.

    I just checked my afrigator rank and i am in the 1,000s or something. Do I consider my blog successful? Of course, because without running ads or anything, my blog serves as my only calling card and has earned me money, career growth, speaking engagements and travel opportunities than anything I ever did before I started blogging.

    Rankings are great, but remember that links don’t drive traffic to porn sites, their market knows where to find them. Traffic is great, function is greater.

    P.

    • I truly love the debate this blog post has generated.
      @Potash, by “low quality”, I meant low quality. Blogs that simply did cut-and-paste cannot be of high quality. Can they? E K Bensah offered a list, related to Ghana. You could read through his list as well.
      Thanks for sharing your thoughts.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.