I created this dashboard to analyze my posts on Google+. You can see how your shares, +1 and comments are growing or going down. You can see your most popular shares and sort them by comments or shares or +1s. And you can just see how your last 10 posts did.

My questions are: how useful is it? Would it be something you would be interested in? Should we launch it? What features would it need to become useful for you?
Susan O'Dea's profile photoJeff Sullivan's profile photoMartijn van Beek's profile photoDaniel Schwabe's profile photo
Thanks +Arvid Bux for resharing... as for the concept, I like the idea, but it would also be interesting to count the number of 'replies made to your posts' and 'mentions of your name' as well...if only from a vanity perspective!

This would allow for filtering of people who post a lot, but they never get any comments back in, and never engage in conversations - could be given a 'rating' similar to the Twitter Value concept used.

Another comment would be that the terms 'Followers' and 'Following' should be changed - this is Twitter Terminology, and not relevant for G+. From your 'Statistics' on the screen shot it shows you are 'Following' no-one... the only way people will see your posts is if you 'Circle' them and share information that way...but how do you get 22670 'Followers' with 0 that you are following...? ... but then this doesn't match your profile...
It would be great if the simple statistics the website show now, would be correct, before launching new stuff.
The problem of 2012-04-11 has never been corrected. And at the moment I don't follow anybody according to http://socialstatistics.com
(All my data are public.)

But I am interested in the statistics, as long as they can be calculated in a normal way. So no "Klout-like" actions.
Something that I'm interested in is is the lifetime of a post. I am sometimes amazed at how long after making a post I still get reactions to it. Sometimes months later.
+Jochem van Drimmelen true. We seen the screen shots on it. GA integration will come but when. That is the big question. Until then, the community is the most important group of developers.
Any feature which even slightly dilute the celebrity worship culture and functionality imposed on a thriving G+ community in early September would be a positive step. A glance at the results of your Most Popular Posts reveals the domination of a very few pop culture celebrities and Google-promoted pseudo-celebrities, turning G+ into an artificially exclusive and effectively closed environment, rather than an open and inclusive community able to select its own leaders and content. Do the Most Popular Posts represent the best contributors of the best content out of the 150+ million participants on a daily basis? No way. There would be far more new users represented, and far more outside of the Top 100 or Top 500 cliques.

So I think there are two valid answers to your question. Yes, such a feature could help many G+ members "do better" with their social and professional networking on G+. But when you look at a specific population of participants, for example photographers, and consider that out of the tens of thousands of photographers on the site the 500th photographer has has been circled by about 16,000 people compared to the top Google-promoted one being circled by nearly 2 million, it's clear that the only way to truly succeed on Google+ is to be chosen by Google, to be promoted by Google as one of the "Fun and Interesting" people on Google+, to the right of everyone's Home stream. Here's the impact on the community of photographers who have helped G+ thrive:
500th: http://www.group.as/500%2Bmost%2Bfollowed%2Bphotographers%2Bon%2Bgplus/500/ 1st: http://www.group.as/500%2Bmost%2Bfollowed%2Bphotographers%2Bon%2Bgplus/
Note the interesting and significant fall-off around places 30 to 40 on the list, where circle follower count suddenly drops from about 1 million to around 60,000 or less.

This special status and the additional contacts gifted by Google leads to more views of posts, more comments, +1s, and shares, being featured far more often in What's Hot, which also results in far greater selection in +Jarek Klimek's selection of top G+ photos, greater likelihood of being chosen to speak at the Google+ Photography Conference being organized by +Scott Kelby, being featured on various Internet shows, etc.

So that leads to the second valid answer to your question: no, tools to study post effectiveness are unnecessary, since individual action on G+ tends to be pretty worthless while even the more successful people in the community lag Google's picks by a ratio of over 100-to-1 (and getting worse by the week), and while all other site functionality and the off-site additions to the Google+ environment naturally flock to feed off the networks artificially awarded by Google, and further promote and glorify the artificially imposed state of unbalance.

This does all point to an interesting opportunity that has gone unaddressed: content discovery features which help highlight better content, beyond the few people Google has had the time to promote. Such tools would be of far more interest to the community, and would help the G+ community thrive. For example, an alternative to What's Hot, without the Google-introduced bias. Calculate the most interesting posts, but discount them by circle/follower count. Whose posts are really the most interesting on Google+? It would be fascinating to explore.

+Jari Huomo has implemented a number of interesting features such as various ways to process and display their own timeline, or the posts of others: http://www.googleplussuomi.com/photos.php?id=107459220492917008623

An interesting and fun feature might be to enable G+ community members to compare their best posts against the best of others, sort of a "Celebrity Death Match" format. A simple display of posts in two columns side by side would be an interesting enough feature that I think people might have some fun with, like this but comparing two users: http://www.googleplussuomi.com/timelinetest.php?googleid=107459220492917008623&sort=best

Additional unique value could be added by enabling users to vote, and when Top 500 celebrities and their posts come up in your site's searches the future, there's now additional value-added data to provide: "If you like this user or post, here's another one that community members have told us that you might like even more". Such a feature could help turn the community back over to the community, one post and vote at a time, helping make G+ more open to all participants.

Why would these features interesting enhancement go G+? Psychology offers answers:

What makes people happy
"According to the selfdete rmination theory, all humans have three basic psychological needs: the need to belong or feel connected, the need to feel competent, and the need for autonomy or self determination. 'When those needs are satisfied, we're motivated. productive and happy. When they are thwarted, our motivation, productivity and happiness plummet,' Prof Ryan says.
Community members want to select their own connections, they want an equal shot at opportunities to be recognized as competent, and they want complete autonomy, not to have their actions, network growth or connections dictated to them (even indirectly by ubiquitous domination onsite and offsite by the same few dozen people in their field). Content discovery, rating and recommendation features would help users take control of their use of the site (give them the autonomy they crave) and provide them with the bonus of seeing better content.
Hello +Boris Veldhuijzen van Zanten interesting graphic and some "personal statistics" are very welcome. Apart from the metrics that are mentioned, I would find it interesting to take into account the number of "followers" one has, so it would be easier to see how many comments/mentions etc. per follower users generate.
Furthermore, it would interesting to see who is interacting most with my posts, p.e. a top 20 of people who +1, mention or share my content the most (aggregated).
+Jeffrey Sullivan if I understand you correctly, you state that the current "voting system" of which content is high quality does not suffice. I wonder how you would "calculate the most interesting posts" in a different way than this is already done (by +1's, shares and comments)? Or is the SUL your main gripe?
+Martijn van Beek Any mention of certain lists does run the risk of getting misconstrued as being "about" the list. Thanks for allowing me the opportunity to clarify the opportunity with regards to content discovery. I tried the Most Popular Post feature, and 100% of the results it provided (for the first page or two) appeared to have been made by SUL members. Popular can be defined as "most votes", but it certainly had failed to find the most interesting content. That is not a complaint about the SUL at all, simply an observation that the algorithm was biased towards posts with more exposure, reminding users that they the site allows little or no self determination, not a particularly useful, insightful, attractive or psychologically rewarding result.

For its Explore, Flickr calculates interestingness in some way related to the time intensity of response to a post, with the ratio of favorites to views being a decent predictor of high interestingness (on G+ an analogous ratio serving as one indicator of interestingness might be shares to +1s). A notable difference however is that there is not a small list of people favored for all or most selections. (In fact, many accounts are reportedly placed on a blacklist and excluded from consideration.) The brilliant part of the algorithm is that virtually any post can get on the list, provided that the response to it is unusually high for that person, and possibly for that time of day and geography. To extrapolate that to G+, some sort of baseline like the statistics above could be established and stored for each person... typical +1s, comments and shares, even adjusted for a given time of day, and unusually interesting posts could be identified relative to a particular user's baseline responses, not just absolute totals.

In fact, each person has a personal recent ratio of circle followers to responses, as described by +Damien Walker in this blog post: http://google-plus-blogs.blogspot.com/2012/01/where-has-all-google-love-gone.html. I suspect that with very little tweaking, such a numerical adjustment factor can be used to create a level baseline to measure against across users. For example, if someone's average number of recent +1s is normalized to the value "1" the value of a given new post can be measured as (personal response ratio) x [(the post's number of +1s) - (that person's recent average of +1s)] + 1. A resulting value above one is an above average post. Assuming shares are an even stronger endorsement of a post than +1s, a given user's recent personal ratio of shares to +1s can also be normalized around 1, with below average posts providing numbers below 1 and above average factors providing numbers above 1. Multiply all the factors together to get a numerical result indicating how far above or below normal the response is for a given post. Add as many factors as you want into the mix, and resulting numbers near one will always indicate an average post, with better posts resulting in higher numbers. Once you see the typical range of results, you can even add a constant in front of each factor to control and adjust their relative weight in the overall formula.

In summary, I'd propose that a simple Most Popular feature, being inherently limited, flies in the face of people's desire for an environment and tools respecting the basic human desire for self determination. On the other hand, a Most Interesting feature with an algorithm normalizing measurements across the community might actually achieve the goal of surfacing interesting content, popular relative to both a community member's own typical posts and the posts of the rest of the community. An algorithm should be able to achieve that goal well.

I like your idea of being able to see who is interacting most with our posts.
+Jeffrey Sullivan thanks for the explanation. The idea for normalized posts is very interesting. Also read Damien's blog post and had the same feeling about the descending levels of interaction (have not measured it, but from a gut feeling).
+Boris Veldhuijzen van Zanten I think it's an excellent concept and just signed up to try your dashboard. Maybe when my 'statistics' grow and look better I will place the widget on my site too.
+Martijn van Beek There are a few interesting considerations in the opportunity with Google+ to implement something similar to Flickr's "interestingness" algorithm used to pick 500 photos per day to highlight in their Explore.  Look at how popular Photo Extract has been with 10 or 12 picks per day (example: http://www.photoextract.com/plus-extract/2012/6/27).  An algorithm could pick 50-100 or more posts per day and drive a LOT of traffic for the host, but perhaps more importantly, it could provide a window into valuable daily contributions on Google+ which otherwise would have been lost in the flow.  

Such a feature would bring the value of G+ out into the light, it would  give that value greater persistence and availability, and it would make that selected content easy for Google Search to index (and help valuable content rank higher).  Anything which makes G+ look really good like that is something Google would be VERY supportive of.

Best of all, in the community it helps with user retention... rewarding both creators of quality content and readers/viewers who like to discover cool things every day.  And who doesn't?
+Jeffrey Sullivan Have you also seen http://www.circlecount.com/br/popularengagement/? They show three different measures, some which are similiar (in spirit) to what you propose. Those are very interesting proposals, but it remains to be seen whether they truly interest Google. They could very well be thinking on focusing on the "head" of the long-tail distribution (the top 50 photographers reach ˜50 million members - maybe less because people likely circle many of them, so they would be counted several times, but you get the idea) as a quick way to enlarge the network... These alternative proposals essentially aim at extracting the "value" out of the tail of the distribution, based on what would motivate individuals regardless of their ranking, but it isn't so clear if they would be really effective (from Google's point of view). I guess the only way would be to implement them and see the result, but first Google has to be convinced of this!
Add a comment...