Topsy Pro Review – Web Journal – Mid August 2013

I first heard about, and got intrigued with, Topsy’s in depth geo-location capabilities in late June and added some text about it to my Social Media for the Arts class at Rutgers.  That led to a conversation with Topsy Pro PR and engineers last week and a 2 week trial of Topsy Pro which I’m halfway done with.  I don’t want to write reviews on platforms that I haven’t spent a least a few hours or more doing concentrated work on, and Topsy Pro isn’t a complicated platform but might have aspects to it that can’t be immediately discovered without a reason to discover them – so I figured there was no hurry to write a review, lets just play with the platform.

topsy_geo_analysis

Topsy Pro Alerts, triggering and  Reporting -

After a week of time, I think I’m in a position to write about Topsy Pro.

Topsy Pro is an extension of the Topsy search engine (free) and the pro version provides analytics which the free version doesn’t try to do  as much of – but which I quite like, anyway.   Rather than share the press releases all over again, since the story about Topsy Pro’s latest API offerings have already been shared – here’s a link to an article that provides all the information.   I will focus my commentary on Topsy Pro and leave the API stuff for programmers who want to use Topsy data (going back 3 years, mostly Twitter data and any website or social media link, photo or file that is shared via a tweet).

First, Topsy Pro is a search engine – in fact, most social media analytics platforms use search engine technologies in one form or another and almost all of them have an index as well as crawlers.  But rather than obscure that connection as many Social Analytics have tended to do (which I highlighted in my Social Media Analytics book – esp in chapter 10) to Search Engines like Google, Topsy clearly extols itself as the Search Engine that has all the Tweets (and whatever was included in the tweet) indexed and you can generate results with little delay on the full Twitter firehose (and anything contained in it).

Below, I have cited some recent stories where a search engine and analytics around Twitter and it’s metadata might be useful (and in fact, where it may have been used, for example in the Obama 2012 campaign).

Recent Data that supports the use of Twitter as a Predictor of Success – and supports the use of Topsy Pro

Twitter as predictive tool for elections - http://www.theatlantic.com/technology/archive/2013/08/a-new-study-says-twitter-can-predict-us-elections/278612/.  The post cites a recent  Indiana University study saying there is  a statistically significant relationship between Twitter data and US election results. Specifically, the study found a correlation between the number of times a candidate for the House of Representatives was mentioned on Twitter in the months before an election and his or her performance in that election. The more a candidate is mentioned on Twitter, the better.

Twitter and TV Tune in Behaviorhttp://blog.360i.com/social-marketing/nielsen-confirms-what-we-all-knew-tweeting-leads-to-tune-in  “….– a spike in tweets will cause a statistically significant spike in viewership. But what’s most interesting about this study is that it confirms the cyclical effect that social chatter has with regards to ratings: Twitter drives ratings, which lead to a higher volume of tweets, which in turn drives increased viewership, and thus even more tweets.”

For this kind of information Topsy Pro is useful because it can pull the Twitter information and what’s in it very quickly get the data .  If you’re an entertainment network or a large retailer, then Topsy Pro is good platform to have in-house because there’s a lot of interesting behaviors you can uncover.  And as more and more online media is appearing over offline media, often the consumption of that media is reflected in the Twitter stream - so if you’re studying the ecosystem (similar to what I am showing in the Twitter Drives Ratings image above) then Topsy Pro can probably capture the real-time information and go back the three years, or whatever interval you need, very quickly.

So if that’s your use case, or it’s something along those lines, and you need to do reporting based on it – then what Topsy Pro offers and continues to deepen will probably suggest a regular monthly/yearly subscription.

Offline_Sales_Impact_-_PTw_engagement_drives_greater_in_store_sales_12

Then again, there’s also evidence that online tweets to predict offline buying behavior in another study from Vision Critical (see the  stats from Vision Critical showing that 40% of social media users have purchased an item after favoriting or sharing it on Twitter, Facebook or Pinterest).  Again, if this is something you want to watch in Twitter – than Topsy Pro is your place to do it.

 

Source: http://www.youtube.com/watch?v=z_5x9mYSz5Y&feature=youtu.be

Topsy Pro Review

 

Figured I would try to use Topsy Pro on a few topics that I am  interested in see what kind of results I get along with what kind of things can be done with the data coming out of Topsy Pro, above.  Figured a video is worth about 10,000 words or more so I spent 11 minutes to take you around Topsy Pro.

My feelings about Topsy Pro are that it’s an interesting platform if you’re working mainly on Awareness or Buzz related metrics and you willing live with simple keywords, hashtags and Boolean Queries.  The Twitter Predictive results for US Elections are an interesting example of how the Twitter data, pulled quickly via Topsy Pro can go beyond awareness and with some algorithms lead into predictive analytics.

On the other hand readers of my recent ClickZ article know how I feel about about Boolean Queries and Keywords - while these common constructs are useful at times, they’re also incredibly limiting and simply aren’t ready for the next paradigm shift needed around social intelligence – but I admit that it may have been all we had to work with up till now.

Therefore, I don’t blame Topsy or any search engine for depending on keywords or Topsy Pro for enabling Boolean Queries, I only want to say that, to the extent what you want to know or study is constrained to a keyword or term, these tools can be helpful, but to the extent you want to study the  psychographic and behavioral “aura” of what you really want to know about (for whatever that might be),  what is beyond those keywords or queries, any keyword based tool will have problems- even Google, though it’s getting a lot better with Google Now, so I’m told (but there again, Google is gathering contextual data to add to the search results based on your past and present actions and behavior).

It’s almost as if we’re asking keyword based tools to do things that they can not do without a lot of extra work.  I have suspected that what people say and do are largely out of scope for most keyword based approaches – they simply can’t capture and process the necessary data, as I alluded at ClickZ last month. It also helps to know what you want to do with the information you’re going to pull out of the tool – any tool.  When you can frame your questions or asks in terms of what the platform tool can produce for you, then you probably have the right platform to work with.  If not, keep looking.

Web Journal

Speaking in Bogotá Colombia in late October.

That’s all for now.

0 replies

Leave a Reply

Want to join the discussion?
Feel free to contribute!

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>