ImpactWatch

With the Beijing games’ conclusion in August, the Olympics Demo we’ve been running has a lot to show for itself. Over the course of the summer, we picked up thousands of news stories ranging from athletic scandals to the cuisine of China, medals sweeps to Tibet supporters.

Overall, as noted in our previous posts, the Olympic games held a much higher favorability rating than that of host China. Nearly all articles reporting on sporting results were much more correlated with the Olympic Games or other countries, while China’s stories consisted largely of pollution, Tibet, and human rights.

As the games began, the “Phelps Fever” swept the press. With the flexibility of the ImpactWatch platform, I was able to add attributes to tag all articles related to Michael Phelps. Taking data from August, here is a graph I generated showing the topics breakdown:

It is incredible to see that out of the articles published about athletes, Phelps-related stories comprised of nearly 1/5 of the total news about athletes. I had expected it to be slightly higher, but this is still an incredible percentage.

Our demo worked out just as we had intended. Aggregating articles about the Olympics and arranging them in intelligent and meaningful ways yielded interesting results, as well as provided an outlet to showcase the features of the ImpactWatch platform. As this demo was my first IW experience, I am convinced that the platform is a valuable tool for monitoring media (our clients agree!).

We’re throwing some ideas around within the IW team for a smaller-scale demo in the future, and so far we’ve come up with UFOs, assorted B-list celebrities… have a suggestion? Drop it in the comments, we’d love to hear it.

As the Olympic games have kicked off over in Beijing, I thought it was time to update the topics and issues portion of the demo to reflect current events. We’ve removed some of the topics and issues that are no longer in the news (such as the torch run) and added some others that have come to light since the 8th. So far the new additions are medal race, journalistic freedom, and security. Have any ideas for other issues or topics we should be tracking for the report on the games’ duration? Let us know in the comments.

With only three days left before things kick off in Beijing, our ImpactWatch demo has made itself clear: within the media, China is being portrayed at a far less favorable rate than that of the actual Olympic Games. Even with widespread buzz addressing athletic performance enhancers, coverage of China’s human rights record, air pollution, algae blooms, and authoritarian domestic policies have been pushed to the forefront of western media. While glittering articles about athletes’ preparations boost the favorability of the Olympics, China is struggling to find its own crutch topic under the strain of relentless criticism and unfavorable press.

The graph below illustrates what I’m talking about:

 Even from this raw graph of daily average favorability, the difference is quite noticeable. When I visualize by month to show overall trends, things get interesting:

These trend lines highlight the discrepancy between the two topics. While the Olympic Games’ average favorability has had its ups and downs, its average favorability has both remained positive and increased in positivity from May to August.

China has not fared as well in its coverage; while enjoying two relatively neutral months (May and June), its favorability rating began to falter in July (-.13) and August (-.29). Quite the opposite of the Games’ coverage, China’s trend line indicates both a negative rating and an increase in negativity from May to August.

As the Games’ opening ceremony nears, a new chapter in our ImpactWatch demo begins. With Beijing as a stage, the world — and our demo — is all eyes…anything could happen.

I have spent this summer working for the Bivings Group as an intern. The main focus of my internship has been working on ImpactWatch. One feature that strikes me as particularly interesting is the use of human editors to rate the sentiment of articles.

Many media measurement companies now use automated systems to rate the sentiment of news articles. While this system may be able to read and rate the articles faster than a human, we feel that that the complete automation of the system causes the information to lose value. The Bivings Group will continue to monitor the automated sentiment analysis technology and will switch to it if and when it is warranted.

We have found that while great strides have been made in the field of sentiment analysis, things like sarcasm and colloquial speech can often throw the system off. Another weakness of automated sentiment analysis programs is that they tend to mishandle trends. If a company has made a mistake, but is being praised for their efforts to correct that mistake, automated systems will often rate that as negative, when it should be positive. Sentiment analysis software also seems to run into problems when it comes to rating the sentiment of different entities in an article.

It is for all these reasons that the Bivings Group has chosen to stay with human editors for the sentiment rating in ImpactWatch. We are confident at this point that the extra time it may take to rank the articles is worth it because we know that we are giving our clients the most accurate data possible.

Since its launch in 1999, our ImpactWatch media monitoring platform has been focused on serving the enterprise market.  The product has been aimed at organizations that receive a high volume of coverage from both traditional and online media, require a large number of employees to access the data ImpactWatch collects and need high-end  reporting and analysis of coverage.  Due to its robust feature set and flexibility, ImpactWatch remains a great solution for these types of enterprise clients.

But in showing ImpactWatch over the years, I’ve run into a lot of people who love ImpactWatch’s interface and base tool set, but who don’t need all of the advanced features we’ve built into ImpactWatch over the years. They were looking for an elegant and affordable way to manage and report out on their media coverage as opposed to an enterprise level media tracking platform.

In an effort to provide these folks with a product they can use, we are in the process of rolling out two new ImpactWatch product lines that will enable the IW to serve the needs of organizations of all sizes.  Specifically:

1. ImpactWatch Basic - Aimed at groups that get a limited amount of coverage (approximately 1-50 clips a day) and only need for a handful of people to access data.
2. ImpactWatch Professional - For organizations with a higher volume of coverage (50-100 clips a day) and who need a few dozen people to access the system.
3. ImpactWatch Enterprise - This is our current system, which is aimed at organizations with a high volume of daily coverage (100+ clips), a large number of users and that need access to some of ImpactWatch’s high end reporting features.

I know this is vague, but we are in the process of a big development push and aren’t ready to talk about all the details yet.  Check the ImpactWatch blog for updates on our progress and look for us to roll out the new product lines later in the summer.

We’re excited.

Only sixteen days after my last post about our ImpactWatch demo, the media has begun drifting away from Sichuan’s earthquake in light of high-tech espionage, Paul Hamm’s broken hand, and a record-breaking year for track and field. With the Olympic trials season in full swing, positive coverage on athletes and the games itself is steadily rising. I created an issues breakdown visualization using ImpactWatch, which reveals that just as quickly as the media surrounding the earthquake sprang up, most of it seems to be over:

While the stories surrounding the earthquake have subsided quicker than I had anticipated, it certainly set the stage for interesting coverage of the 2008 Olympic season. Stay alert, more updates to come.

In the last post, I did a micro-study of Kinoki Detox Foot Pads and the way they are being portrayed in the realm of YouTube. In addition to analyzing the “views” count on the website, I adjusted the quantity by minutes to give a better representation of the exposure YouTube is actually giving the Kinoki brand.

This time around, I will examine the user feedback to the videos that were part of the study. If I had been the one who had actually posted these videos, I would be able to utilize YouTube Insight, a powerful built-in tool that turns an ordinary YouTube video into something that resembles an online focus group (there will be more on this tool in the future). Seeing as how I am not interested in posting my own videos about Kinoki Detox Foot Pads, I will have to rely on the inexact method of comment analysis and “feel it out”.

First, a red flag. On the positive videos, there were only a couple of comments that were all positive and framed as customer testimonials reiterating their “refreshing feelings” and how using them had changed the users’ lifestyles for the better. After three or four comments, the commenting feature had been disabled, thereby locking in a few positive comments and keeping naysayers from commenting on the wall. That, and the videos themselves were infomercials in the guise of “health update” news reports. Rather suspect if you ask me.
On the videos that allowed commenting, the tone was overwhelmingly negative. A couple of comments would be users who have actually used the product and would claim that the Kinoki Detox Foot Pads actually made them feel better, to which users would instantly become outraged and prove them wrong with their elite high school chemistry knowledge.

In review, the coverage of Kinoki Detox Foot Pads on YouTube is quite negative, both in videos and commentary. If an undecided consumer were to consult YouTube to make a more informed decision, they would more than likely decide not to purchase these foot pads — maybe changing your diet is the way to go after all. I still swear by my twice-daily regiments of snake oil.

According to Alexa’s listing of most visited websites (sorted by country), YouTube is ranked as #4 in the United States, #6 in the United Kingdom, #4 in Japan, #5 in South Korea and #2 in Germany. It has the power to launch overnight public-relations Cinderella stories such as the Blendtec miracle (yes already, it will blend) as well as broadcast to millions a reputation-killing moment even more swiftly. These occurrences, once posted to YouTube, are available indefinitely.

How does one quantify the amount of successful or damaging exposure YouTube is causing them? Here are some measurements of a micro-case study crafted specifically for this blog post: an analysis of YouTube exposure of the television infomercial product, Kinoki™ Detox Foot Pads. Note: The Bivings Group and the ImpactWatch service are in no way affiliated with Kinoki™ Detox Food Pads.

I manually aggregated data by watching YouTube videos related to the “miraculous detox system”. You can see the raw data that I collected here in an excel spreadsheet. Using the resulting data, I created some visualizations within ImpactWatch:

The first graph shows a raw view of the types of videos people watch related to the Kinoki™ Detox Foot Pads. This is strictly measured in views, which is valid because YouTube only counts views once from each unique IP address, and only if an overwhelmingly large portion of the video was viewed. However, this unit of measurement is often misleading and does not give an accurate representation of the actual exposure Kinoki foot pads have received. That’s where a little spreadsheet manipulation and the second graph comes into play.

In the second graph, a new unit was analyzed. Taking the number of views for each video and multiplying them by their length in minutes, an adjusted unit that more accurately represents total “exposure” was created. This graph provides a much better metric for gauging the float-or-sink status of the product. Although in both graphs negative exposure was dominating (I wonder why?), the adjusted quantity of “YouTube minutes” shows that it wasn’t as bad as just a raw views count might have initially demonstrated.

In the next part of this two-part case study, comments for the videos and other forms of responses will be analyzed.

After last week’s debacle with Barack Obama’s MySpace page, lots of bloggers have been discussing whether or not political campaigns should take control over MySpace profiles and other voter-created websites and online groups. It’s no surprise that bloggers have been covering this issue, but I wondered to what degree the story had reached mainstream media (MSM) outlets.

I did a quick Google search for articles about Barack Obama from May 2 to May 4. I limited my search to print media outlets (newspapers and magazines) and major online news sources (such as CNN.com or MSNBC.com). I did not include blogs, international sources, or TV/radio outlets in my search. To further narrow my search, I only considered articles that were directly about Barack Obama–his name was either in the headline, or he was a substantial part of the story. I did not include political overviews or articles where he was mentioned in brief.

I found 51 articles over the three days, 10 of which were about Barack’s MySpace page. I marked particularly negative articles with an asterisk.

Political blogging growing like a vine*
Barack Obama Could Lose Some ‘Friends’ In MySpace Debacle*
This Should Make Bloggers Happy *
Learn from Obama: Hire a Professional *
Campaign takes control of ‘official’ MySpace site
Between Barack and MySpace
I’m Sorry, Barack, We’re Through*
Obama Campaign Asks: Is it MySpace or Yours?
Obama Takes MySpace Page from Backer
Obama to MySpace: That’s MyPage

These 10 articles were spread over 8 different publications and make up 20% of Barack-specific coverage during May 2-4. I was pretty surprised that such a variety of MSM outlets, some of them very significant publications, picked up on this story. I was also surprised that MSM outlets were so critical of Barack’s efforts at regaining control over his MySpace page. I expected that since most newspapers and magazines maintain top-down strategies in most of their media coverage, they would be supportive of the transition to the “official”, rather than voter-controlled strategy.

To see the articles I included in my survey, please take a look at my datasheet.

Our media monitoring platform, ImpactWatch, would be perfect for tracking an issue such as this one. In order to complete this analysis, I had to manually search through Google results. If I had used ImpactWatch, this search would have been automated, allowing me to pre-determine which news sources I would be tracking. In addition, ImpactWatch would have given me a mechanism for sorting, graphing, and ranking the results according to subject, source, author, and bias (positive, negative, neutral).

For more information about ImpactWatch , check out the website and our current free demo, which tracks coverage of the PS3 and Nintendo Wii.

I was against promoting Ajit’s music on the blog. But then his cover of Gnarl Barkley’s Crazy got over 10,000 views on YouTube, so I realized we needed to jump on the bandwagon and support our own. So here’s our Senior Director in charge of Impactwatch, Ajit Verghese, singing Crazy:

Listen to Ajit. Buy ImpactWatch. Love the Bivings Group.