Much has been made of late about rising gas prices, with recent national averages surging from about $3.00 per gallon to $3.60 per gallon in just the past 3 or 4 months. Price at the pump has always been a particular sensitivity for the American consumer because it’s a price many pay every week or two that can account for a significant portion of people’s discretionary spending.

We have seen evidence in recent years of the negative impact rising gas prices appear to have on e-commerce spending, which is mostly discretionary in nature. If we look at year-over-year retail e-commerce spending growth rates, we can see that the exceptionally strong growth we had been seeing up until mid-2007 began to drop off considerably over the next year, coinciding with a run-up in retail gas prices from about $3.00 per gallon to over $4.00 per gallon.

With the worst of the economic downturn seemingly behind us (although the unemployment rate remains stubbornly high), we have begun to see a consistently improving outlook for the retail e-commerce sector with growth rates back into the double-digits and sustained over the past four months. However, with gas prices similarly on the rise, there is a legitimate question as to whether these growth rates will continue.

How Do Gas Prices Correspond to E-Commerce Growth Rates?
There has been debate over the years about the effect of rising gas prices on e-commerce spending. One theory says that as gas prices rise, consumers will be more likely to forgo car trips to the mall and opt instead to shop online. While this effect might exist to some small degree, the more prevalent effect of rising gas prices is the constraint on total discretionary spending. Put another way, the $30 or $40 a month more the average American is spending on gas plays a much larger role in their retail spending habits than the fact that they can save a dollar or two in gas by forgoing a trip to the mall.

To get a better understanding of just how important gas prices might be in determining the levels of retail e-commerce spending, I wanted to see how much of a correlation existed between those two variables. I began my analysis by running a scatter plot of Retail Gas Prices vs. E-Commerce Spending Growth Rates (for the months of January 2005 through February 2011) expecting to see a visible inverse relationship between the variables. In other words, when gas prices went up, I expected to see growth rates decline. The data instead showed that the correlation overall was very low(r-squared = 0.0083), meaning there was no immediately apparent relationship between the variables. What could explain this?

On examining the graph, I realized that many of the data points that seemed to influence the regression were the ones on the lower left hand side – when low gas prices coincided with low and negative e-commerce growth rates. These data points all occurred during the recent recession when consumer demand plummeted, bringing both consumer spending and gas prices down with it. It became clear that other variables beyond gas prices would need to be considered in this analysis to help isolate the effect of gas prices on e-commerce.

So I introduced a couple of new variables into the analysis to help account for the effects of the recession: Unemployment Rate and S&P 500 Index. Unemployment Rate would help account for the most tangible and immediate effects of a recession, while the S&P 500 Index would provide a reasonable approximation for household wealth and its effect on consumer willingness to spend.

Regression Results
I conducted the regression using monthly data from January 2005 through February 2011 with (Y) U.S. E-Commerce Growth Rates as the dependent variable and (A) U.S. Unemployment Rate, (B) S&P 500 Index and (C) U.S. Retail Gas Prices as the independent variables. Below is a sample of the data used in the analysis:

The results of the regression, which generated a reasonably strong r-squared value of .64, provided the following model for prediction:

Y = 0.05522 - 0.01881*A + 0.00032*B - 0.06351*C

This regression indicates, as one might expect, that there is a positive relationship between the S&P 500 Index and E-Commerce Growth Rate, and an inverse relationship between Unemployment Rate and E-Commerce Growth Rate.

But most important – and the reason for conducting this investigation in the first place – is that it clearly illustrates the strong inverse relationship between Gas Prices and E-Commerce Growth Rate. For every 10 cents gas prices rise, one would expect e-commerce growth rates to decline about 0.6 percentage points.

Putting these findings into the context of the current economic environment, with February 2011 data showing a U.S. unemployment rate of 8.9, a closing S&P Index of 1327, and U.S. Retail Gas Prices of $3.21, our model predicts an E-Commerce Growth Rate of the following:

Y = 0.05522 - 0.01881(8.9) + 0.00032(1327) - 0.06351($3.21) = 10.7%

The prediction of 10.7% is not too far off from the actual February growth rate of 12.0%. But the point isn’t to have a perfectly predictive model; rather, it is to gain a reasonable understanding of what to expect with e-commerce spending as gas prices rise.

With March now in our rear view mirror, we have seen average gas prices accelerate to around $3.60 per gallon, nearly $0.40 higher than the February average. If we hold the other variables constant and raise the gas price in our model to $3.60, we get a predicted E-Commerce Growth Rate for March of 8.2% – or 2.5 percentage points lower than February. Now, that might be an extreme view, as average monthly gas prices for March will probably be in the ballpark of $3.55. But even with that input into the model, we get a predicted E-Commerce Growth Rate of 8.6%, still more than 2 percentage points lower than February.

Cautious Outlook for Retail E-Commerce?
In summary, I think it’s fair to say that rapidly rising retail gas prices are cause for concern for the retail e-commerce sector. As disposable income shrinks, so too does discretionary spending as manifested in online retail. All other factors being equal, we should anticipate a 2-percentage point decline in e-commerce spending growth rates vs. what we would have otherwise seen. Hopefully this effect is offset slightly by a healthy stock market and a declining unemployment rate in March, but it still means there’s potential to return to single-digit e-commerce growth rates following four consecutive months of healthy double-digit growth rates.

We’ll have a better sense of how these dynamics play out when we release our Q1 2011 online spending figures in a couple weeks. For now, we can only hope that spiking gas prices is a temporary effect of the conflicts in the Middle East. But with summer just around the corner, chances are gas prices will remain high in the near term and may continue to climb even higher. Online retailers must keep a close eye on these factors because they will likely play a significant role in the ongoing health of the retail e-commerce sector.

This post was originally published at SearchEngineWatch on April 4, 2011.

This past week I decided to download the newly released Firefox 4 web browser, which featured several noticeable improvements over previous versions. As someone who lives and breathes search every day, my eyes were immediately draw to the new look of the search bar in the top right hand corner of the browser. While it defaulted to Google search, there was a dropdown that enabled me to access several different search experiences.

I began to wonder about the relationship that the search browser has with consumer search behavior. Does your affinity for a particular browser make you more or less likely to engage with certain search engines? Do the default search options in browsers make a difference? I thought it would be worth investigating whether or not any particular affinity existed between browsers and search engine usage.

(I will state at the outset that we must remember that correlation does not equal causation in these cases, so just because someone chooses a browser does not mean it necessarily caused them to be more likely to engage with a specific search engine. It is also very possible that someone with a natural affinity for the Google brand, for example, might be more likely to be both a Chrome user and Google search engine user. In this case, the brand affinity is the cause and the browser and search engine selection would both be the effects.)

Browser Wars Heat Up
For a long time, Internet Explorer was the dominant web browser but over the years Firefox emerged as a strong second player. In late 2008 Google got into the browser game with Chrome, which has gained traction and proved to be popular as well with a segment of Internet users. Chrome was notable for its speed, providing more competition for the established players. More recently, both Mozilla (with Firefox 4) and Microsoft (with IE9) introduced new versions of their browsers that represent strong upgrades for anyone who used the previous versions.

Now obviously search engines have a vested interest in being the default engine and/or toolbar on your browser. Even though many users will customize their browsers to fit their search needs outside of the defaults (e.g. a Google toolbar on Internet Explorer), or just download and use the browser that is already aligned with their favorites (Google Chrome), not every Internet user is so discerning. And this may perhaps explain why we see the affinity we do for search engine usage with different browsers, which the numbers below illustrate:

Google Chrome
Google Chrome has the most obvious affinity for Google searches. 87.1% of the searches that take place on Chrome are on Google. This represents a significantly higher percentage than the 64.9% share Google has in the Total U.S. Internet universe. Yahoo! and Microsoft combined account for just 12% of search share on Chrome (8.3% and 3.6% respectively). Although it stands to reason that many of the users that have downloaded Chrome are probably already loyal Google users, do not discount those that previously used other browsers (and in turn other search engines), and took a chance on Chrome and enjoyed their experience enough to keep using it.

Mozilla Firefox
Mozilla Firefox is an open source application that partners with Google for their default homepage and toolbar, but due to its extended reach these days, tends to be standard software present on most machines when purchased. The fact that PC users have a choice perhaps somewhat lessens the Google bias from the dominant Chrome numbers, but those default settings would appear to still play an important role with 73.4% of the searches on Firefox take place on Google. Yahoo! (16.4%) and Microsoft (7.1%) have inched their way back up towards their overall combined Total U.S. Internet share, but there is still a significant delta between those numbers and their total universe share.

Microsoft Internet Explorer
Internet Explorer has been the leading browser in the marketplace for many years, with Firefox and Chrome beginning to catch up. With Microsoft still controlling the lion’s share of computer operating systems, their packaging of Bing with IE would seem to benefit its cause. The value of this reach is apparent when we see that the Google search share on IE drops considerably below their overall market share, with only 55.8% search share. Bing and Yahoo! finally catch up a bit, hitting 18.7% and 20.7%, respectively. Considering that more than 60% of the Core Searches in the U.S. in February were performed on an Internet Explorer browser, those seemingly small slices of the total pie assume much greater importance than you might think.

Conclusions
Search experience, relevancy of results, and let’s face it, behavioral inertia, are still the main drivers behind which search engines we use every day. If you love Google, you’ll find a way to use it anywhere. If you are a Yahoo! mail user, you probably have a Yahoo! toolbar downloaded and you might use that for some of your searches. If you are an Xbox Live member, maybe you ended up downloading a Bing toolbar and that becomes your search engine of choice.

Even though such factors can strongly contribute to your search activity, do not overlook the influence the browser has on your search behavior. Browser choice and search engine usage are interrelated, so it’s little surprise these companies are racing to improve their browsers in the hopes of gaining market share, which may help deliver incremental revenue such as increased search activity on their engines. Only time will tell how these browser battles will play out, but whatever happens we can be sure it will have some effect on the complexion of the search landscape.

Recently, comScore released data that showed online traffic in India experiencing a surge in the first few weeks of the 2011 Cricket World Cup. In the last few weeks of the World Cup, visitation to top cricket sites continued to see a sustained peak in traffic as the final matches brought the tournament to a close. In India, the March 30 semifinals match featuring India vs. Pakistan drew the largest online audience for an individual day, with nearly 3 million visitors coming to Yahoo! Cricket alone.

Traffic in the Knockout Rounds Sustains Increase in Volume
Coming off the traffic spike brought about by the March 12 match featuring India vs. South Africa, cricket sites in India saw a sustained rate of visitation throughout the knockout tournament rounds, which began on the week of March 20.

Leading site ESPNCricInfo.com continued to attract the highest number of weekly visitors from India during these rounds, with 6.3 million visiting the site during the week of semifinals (week ending March 20). In the final week (week ending April 3), ESPNCricInfo.com reached 6.6 million visitors, up 8 percent versus 5 weeks prior.

Other top cricket sites experienced even more dramatic increases. Yahoo! Cricket’s weekly visitation saw a 33-percent jump from 5 weeks prior, while CricBuzz.com gained 36 percent. ESPNStar.com saw a sizeable 75-percent increase to 4.3 million visitors, while CricTime.com saw the most dramatic percentage increase at 86 percent.

Highly-Anticipated Matches Show Highest Visitation in the Final Rounds
An analysis of daily traffic coming from India during the knockout rounds shows the highest volume of visitors occurring on the days of highly-anticipated matches involving India. During the quarterfinal rounds, the March 24 Australia vs. India match drew 1.8 million visitors to ESPNCricInfo.com, 1.7 million visitors to Yahoo! Cricket, and 1.3 million visitors to ESPNStar.com.

The final match between India and Sri Lanka on April 2 also drew significant visitation for these sites, although the greatest volume of daily traffic occurred during the March 30 India vs. Pakistan semifinal match. With 1.8 million visitors, Yahoo! Cricket actually surpassed ESPNCricInfo.com’s audience of 1.7 million visitors that day. ESPNStar.com, CricTime.com, and CricBuzz.com also drew their highest daily visitor totals on this day.

The surge in Internet traffic to cricket sites during the Cricket World Cup reflects not just how wildly popular the sport continues to be among participating countries, but also how digital technology supplements the traditionally TV-dominated coverage of this and other major sporting events. With the continued surge in Internet adoption in India, along with the proliferation of mobile devices, the Indian population has more options than ever to stay connected and current to the important events of the day.

Here at comScore, we’re always interested in discovering new applications for our data by combining it with other available sources. And what better way to unleash this innovation than by enlisting the help of some of America’s best and brightest. This past weekend, comScore hosted a social media case study at Carnegie Mellon University along with the Heinz College School of Information Systems and Management, with the goal of developing a new analytics product targeted to the motion picture industry to forecast box office performance and evaluate marketing decisions based on digital analytic inputs.

The emergence of social media offers the potential to better predict box office performance, and the value this can create is enormous. Hollywood movies gross approximately $30 billion per year, so it begins with a very large market. The average production budget for a major studio movie often tops $70 million and another $35-40 million is typically spent on marketing. Each year a few runaway hits often compensate for the majority of movies that fail to break even. However, once a movie has been completed (whether it’s blockbuster or a bust), production becomes a sunk cost and the role of the movie marketer is to maximize yield on ticket sales. Another way to think about it is that a movie marketer’s job is to manage risk and employ strategies to minimize the downside on movies unlikely to be a strong box office draw and to maximize the ticket sales of the ones likely to be breakout hits.

The competition featured 21 groups with 1 to 5 participants on each team. Teams came from many of Carnegie Mellon’s graduate school programs including the Heinz College School of Information Systems and Management (ISM), Tepper School of Business (MBA), computer science masters program, and others participated. The competition began with each team submitting a one-page overview of their approach to building a predictive analytics product. The teams’ ideas blended comScore’s digital consumer behavioral data with externally available data sources, such as open APIs or using web page scraping techniques.

The variety of student backgrounds made for very interesting combination of submissions featuring very different approaches to the assignment. Computer science students outlined sentiment analysis algorithms for social media streams, relevance and influence ranking systems for social media targeted marketing, and television and radio voice sentiment analysis. Other students walked us through multivariable logistic regressions to isolate important predictive factors during different stages of the movie (pre-release, open weekend, on-going). Many submissions focused on the importance of social media as a predictive input, and a few had particularly clever solutions that also incorporated offline inputs and historical signals, including movie performance by genre, actor, director, etc. Other groups incorporated various macroeconomic indicators, which can provide strong signals of general consumer behavior. Despite these very different approaches to the project, one consistency across all groups was an extraordinarily high level of creativity.

The entries were judged by a committee from comScore and CMU: Linda Abraham (Chief Marketing Officer), Dean Logan (Senior Director, Cross Media), Jeff Weinstein (Director, Research and Development) and CMU professors Ari Lightman (Practice Professor, Digital Media and Marketing) and Michael Smith (Associate Professor of Information Technology and Marketing). We selected the 5 best initial proposals and invited them to perform an in-depth pitch of their idea the following afternoon. The panel was universally impressed with the technical, analytical, design, and marketing creativity of the final five groups.

The winning product was a multivariable logistic regression model from Asad Sheth and Patrick Clary from the Tepper School of Business. We were impressed by their detailed and inventive approach to data collection, which included Youtube trailer visitation from the comScore panel, sentiment analysis from Twitter, economic indicators, weighting based on social media influence, and prior sentiment correlation to box office outcome. Their writing was clear and concise and their presentation skills were fantastic. They dove deep into the model training method as well as outlined market positioning, tiered pricing strategy, and valuation steps. Their technical backgrounds (both had undergraduate degrees in computer science) were well balanced with their overall business acumen, which enabled them to deliver a memorable and convincing pitch. Way to go guys, and congratulations on being selected as the winning group from amongst a crowded and talented field!

Unless you’ve been living in a media-free cave for the past month, you are well aware of the Royal Wedding between Prince William and Kate Middleton that took place today. These nuptials mark the first major royal wedding since the dawn of the Internet era, with the last wedding of this magnitude between Prince Charles and Diana captivating the world back in 1981, almost three decades before YouTube, Facebook, and Twitter were a part of our everyday lexicon.

In addition to being a primary topic in broadcast, print and online news coverage, the Royal Wedding has taken center stage across social media outlets as well. Using comScore’s Social Analytix™ service (powered by Radian6*), we followed global activity across social media destinations - including social networks, blogs, discussion boards and video/image sharing sites - over the course of the past month.

An analysis of social media mentions around the Royal Wedding revealed that the topic is exploding across different social media venues. Social media mentions around the topic eclipsed 1.5 million citations globally since the beginning of April. On April 27, there were more than 200,000 mentions of ‘Royal Wedding’, up 1,215% from the first of the month, as the approaching event fueled a flurry of discussion. As with most weddings, the bride captured more attention than her groom, with social media mentions of ‘Kate Middleton’ outpacing mentions of ‘Prince William’ by a decent margin. On April 27, there more than 64,000 global social media citations for ‘Kate Middleton’ compared to nearly 47,000 for ‘Prince William’.

When looking at a comparison of the U.S. and U.K. markets by volume of social media activity around the Royal Wedding the United States commanded the lead by a wide margin. Nearly more than 1 million social media mentions on the Royal Wedding occurred in the U.S. from April 1 to April 27 compared to more than 256,000 mentions in the U.K. (It should be noted, of course, that the U.S. has about 5 times the population of the U.K., and only about 4 times the number of social media mentions. That said, the U.S. is surprisingly close to the U.K. on a per person basis.)

With people tuning into the event across the world – whether it’s through live TV broadcast, DVR recording, YouTube video, Twitter feed or one of the many other ways we now consume information – it will truly be an example of the breadth of media consumption options we now have available across the globe.

*Radian6 covers over 150 million sources each day, including the broadest available coverage of: Blogs, Mainstream online news, such CNN.com, Video and Photo sharing sites, like Flickr and YouTube, Micromedia, including the full Twitter firehose and Friendfeed, Forums and Discussion Boards, Blog comments, Facebook public discussion forums. Radian6 uses a combination of RSS and other data feeds in addition to proprietary crawlers in order to index the entire social web. If data is available publicly, we’ll capture it. (That also means that data the community trusts to stay behind a wall, will; we don’t crawl or capture sites like private portions of Facebook, private communities, or anything like that.) We capture over eight million on-topic posts and comments each day, filter out the spam and irrelevant data, and deliver results to you in near real time.