#TopvBlog2015 Analysis Part 2

You may be aware that the full results of the Top vBlog 2015 voting by Eric Siebert at vSphere-land.com are out (sponsored by Infinio).  Part 2 is an analysis of the Top 25 vBlogs from publicly accessible data.  There were some surprising figures.  Part 1 was based upon my Top vBlog results.

UPDATE! Some people felt that the financial aspect of the analysis had no bearing upon the Top vBlog List.  Agreed, I merely added it so that those new to the Top vBlog List are aware that gross income of some magnitude can be generated from advertising.  Most bloggers take their advertising revenue and reinvest it in their home lab infrastructure and cover blog operational costs.  Some bloggers do not participate in advertising.  It is a personal choice and their business.  If you do not want your revenue to be public, leave BuySellAds.com and use some other private mechanism, which will increase your operational complexity and net income.  In the interest of transparency, I added myself to the list at #37 with my Ad earnings as well, which I run privately.

The other point of contention was Quantity versus Quality.  That is, the value of a single Deep Dive versus multiple Configuration Guides, KB updates, etc.  I personally think they all have merit, as long as you are passionate about what you post, do it with integrity and provide a unique and original perspective.  Definitely the effort that goes into writing a feature or design Deep Dive and the knowledge derived from it has greater benefit than a straight How-To.

From a Top vBlog scoring perspective, the question is how do you identify and measure Quality automatically?  If there was a method (intelligent heuristics of some sort) then you could weight a Deep Dive with more value compared to all other post types (20 to 1 was suggested).  Right now the Top vBlog list is a popularity contest, it would be great if it could be automated and some additional metrics used to recognise the effort and merit of individual vBloggers within the vCommunity.  No more, no less.  END OF UPDATE.

I created the following metrics with information available from the Internet:

1. Number of posts during 2014 – sourced from Andreas Lesslhumer at running-system.com
2. Views per month – collected from Buysellads.com
3. Published Author – search of Amazon.com
4. UPDATE! VCDX – from the VCDX Directory, suggested by Andrea Mauro
5. Number of Twitter followers
6. VMworld Speaker – from memory, the VMworld session portal is now setup for 2015 only
7. Cost per month of a 125×125 banner ad (or nearest possible) – from Buysellads.com, does not account for DIY ads
8. Total gross earning potential of the blog per month – based upon listings in Buysellads.com, does not account for DIY ads.  Frank Denneman pointed out Buysellads.com takes a 25% cut.
9. DROPPED -> Year started – most of the Top vBlogs do not have an easy to use “Archive” function to locate the start year

The highest values are highlighted in GREEN and the lowest values are in RED.  Figures have been rounded down for ease of consumption.

Note: the Views/Month values from BuySellAds.com have been reported by some Bloggers as being inaccurate.

What are the characteristics of a Top 25 vBlogger?

• Minimum of 5,000 blog views per month
• Minimum of 6 posts per year
• Minimum of 1,600 followers on Twitter
• May have published a book on virtualisation (13/25 are)
• May be a VCDX (10/25 are)
• May be a regular VMworld Presenter (16/25 are)
• May advertise to earn income (22/25 do)

My Thoughts

Currently we have a popularity contest, the more people that vote for you the higher you rank.  This is a privately run competition that was Eric Siebert’s brain-child and he has carried it forward since 2008.  He has done a fantastic job of developing and running the competition.  The problem Eric is competing with is scale.  What required a small amount of effort in previous years is now weeks of manual work and judging by Eric’s blog posts and Twitter feed, it is not getting any easier.  Remember, Eric has a day job and he does this for love not money.  The same goes for all of the vBloggers on the list.

The fact is, the Top vBlog List is an industry resource.  A higher ranking does have an impact on your professional prospects and your earning potential as a professional, blogger and evangelist.  It deserves additional effort and resources from the vCommunity.

When I look at the table above, two metrics jump out at me: Number of Blog Posts per Year, Number of Views per Month.  I think the scoring mechanism needs to include these metrics during the voting process.  For example, ESX Virtualization (Vladan Seget) should definitely be in the Top 5; Vladan is a content producing machine.

My suggestions of what I think should happen:

1. Eric needs volunteer developer assistance from the vCommunity.  He owns it and sets policy, but leads a small team of developers to simplify and automate things.  If a vendor could contribute resources, even better.
2. Weight the Popularity Scoring with Post Count and View Count metrics.
3. Consider including the Year Started, Authorship, VMworld Presenter and # Twitter followers in the calculation.
4. Automate the process, move from lists and manual editing to a Web-based self-service portal.
5. New vBloggers can use a Register workflow and be added to the list automatically with some approval process.
6. Each vBlogger can update their user profile with their latest Annual Blog Post Count and Monthly View Count (numerical with JPG evidence).  This was one of the great things that Chris Colotti did with the VCDX Directory, he fixed the exact same issue (ie. manual lists -> self service portal).
7. The voting mechanism could list the unranked “High Metric” blogs first so that these people get a chance at making the Top 50/100.  From using the voting form in previous years, there is a “whack-a-mole” feel to selecting all 10 blogs.  You know the 3 or 4 you want to vote for but you have to randomly pick another 6 or 7 to meet the voting form rules.
8. Allow voters to select and rank 10 or less blogs?  The same goes for category voting, allow voters to skip it altogether if not interested?
9. A public cloud provider (eg. vCloud Air, AWS, Rackspace) could donate PaaS or SaaS for the hosting.  It does not require a lot of resources.
10. The Top vBlog annual voting process could have the following schedule:
    • Call for final vBlogger registration
    • Open the voting engine for two weeks of voting
    • Voting closes
    • Run the Voting Ranking workflow (takes seconds)
    • Schedule the Google Hangout for announcing the Top 25 vBlogs
    • Run the Report export to publish the official report blog post (takes seconds)
    • vLaunchpad automatically displays the final results for 12 months (takes seconds)
    • Prizes are awarded and sent to the address listed in the vBlogger profile
    • Eric gets weeks of his life back

The irony of this situation is quite amusing.  We have the top automation guys ranked on a list that is manually managed.

Possible Logical Design

Using my suggestions above as requirements, here is a logical design diagram of what it could look like:

A Possible Future

Create multiple industry lists and offer it as TopBlog-as-a-Service (TBaaS):

• Top uBlog (Unix)
• Top dBlog (Database)
• Top mBlog (Microsoft)
• Top wBlog (Web Developer), etc.

If you have the skills and willingness to assist, reach out to Eric Siebert and offer to help.  The data is already publicly available to get started.  It just needs to be designed, built, tested, soft-launched and placed behind vLaunchpad at vSphere-land.com.