Hitwise Intelligence - Bill Tancer - North America

Groupon down 50% LivingSocial up 27%

2011-08-26T20:34:55Z

With all the debate and speculation regarding Groupon and its planned IPO, I thought it would be a great time to check back in with our previous analysis of traffic to Groupon and its nearest competitor Living Social.

First, to be clear, the above chart measures web-based traffic to both domains and does not include mobile or app specific traffic.

Regardless of these exclusions, the drop-off in Groupon traffic this summer has been significant nearly 50% since its peak in the second week of June 2011 compared to last week. During the same time, Living Social has achieved 27% growth in visits to its site. Overall visits to a custom category of Daily Deal & Aggregator sites were down 25% for the same time.

So why is there a narrowing of the gap between the two market leaders in group coupons? Perhaps it is simply a case of increased number of competitors and deal fatigue among consumers or simply not enough of the right deals. PriceGrabber® released results from its Local Deals Survey in June, stating that 44% of respondents said they use or search daily deal Websites. However, 52% expressed feeling overwhelmed by the number of bargain-boasting emails they receive on a daily basis.

While consumer fatigue may be one factor another key consideration for these sites is to focus on the attracting new and preferred audience segments via the inbox. Currently the audience segments for both Groupon and Living Social are very similar so it will be interesting to see how both sites and category perform heading into the holiday season.

Google Plus – Innovators and Early Adopters

2011-08-12T18:08:08Z

On June 28, 2011, Google+ launched with a limited, invitation-only beta. Now, nearly a month and half later and a less restrictive invitation policy, what does the data have to say about who’s using the nascent network?

The late Everett Rogers, in his book, Diffusion of Innovation, discusses the sequential adoption of new ideas or innovations, from innovators and early adopters to late majority and laggards. Using the Experian Hitwise sample of over 10mm Internet users in the U.S., combined with our New Mosaic segmentation system launched this summer, its possible to visualize the adoption of Google+.

Careful analysis of the Mosaic segments since launch; reveal that in just over six weeks, we’ve moved from innovators to early adopters to early mainstream users visiting the new social network.

Here’s a snapshot of top Mosaic types (by representation) that have been frequenting Google+.

From the graph above we can see that within three weeks of launch, the Mosaic type Colleges and Cafés (O53) peaks at a representation index of over 400 (members of this segment are more that four times more likely to have visited Google+ during the week ending July 16, 2011 then the Internet population overall).

We describe this Mosaic type as young singles and recent college graduates living in college communities. Here’s a word cloud of the key descriptors for Colleges and Cafés:

Its not uncommon for innovators to trial new services online and in some cases abandon those services when they lose interest. As of the writing of this post, for the week ending August 6, 2011, Colleges and Cafés have dropped to an index of 73 and now make-up only 0.3% of visits to Google+.

However, in this case, when innovators move on, several early adopter segments take their place. One of the predominant early adopter segments in the chart above is A03, Kids and Cabernet, described as “Prosperous, middle-aged married couples living child-focused lives in affluent suburbs.”

Here’s a word cloud of A03 descriptors:

Currently, Kids and Cabernet index at 268 in their visits to Google+, and make-up 2.9% of visits to the site. Interestingly, when we look at this segment’s stats for visits to Facebook, they index at a mere 68 and make up less than 0.7% of visits to the leading social network.

While Facebook maintains its position as social networking market leader, a number of New Mosaic segments under-index for visits to the top social networking site. As Google+ matures, watching the migration of these segments will provide an indication of whether Google+ will grow into a Facebook competitor or find its place as a niche social network for specific demographics.

Dating Sites by Credit Score - A Valentine's Day Special

2011-02-14T15:53:50Z

This post is for all of our single readers. While it might be a little late to start your Valentine's Day date search (we know that dating site visits typically surge the first week of the New Year), I thought I'd give you a novel new criteria to evaluate online dating sites. Why not add good credit to your dating criteria checklist.

We have over 1,100 sites in our Lifestyle Dating category. Choosing whether you go for a major like Match.com or eHarmony versus Niche dating sites can be daunting. Using our new segmenting capability at Hitwise, I filtered all sites withing our dating category for sites visited with the highest possible credit score, those in the VantageScore "A" category.

Without any preconceived notions (ok, perhaps a few) I pulled the following table of sites based on segmentation percentage (or those sites within the Lifestyle - Dating category that had the highest percentage of the highest credit).

A couple of surprises here. Dating nudists have good credit, who knew? The dominance of ethnic-centric dating sites seems surprising. But perhaps the most interesting is #10 on the list. Sugar Daddy for Me seems like an odd top-credit dating site (that is until you look at the income breakdown and VantageScore distribution and realize that the site has a healthy component of "sugar daddies" and those looking to land a "sugar daddy".)

And if you're wondering which site had the worst VantageScore (or highest visits from the "F" category), that honor belongs to the website for the mobile application Crush or Flush. You've been warned.

We've Got Credit, Yes We Do, We've Got Credit...

2011-02-04T16:52:29Z

There are a few things I have yet to accurately predict with Internet data, namely political races and sporting events. Politics is tricky given the existence of confounding variables in visits to political sites. Sporting events, as skill based competition, have little relationship to search volume or Internet visits. That being said, there's no reason why we can't have a little fun with the data.

On our Experian Marketing Forward Blog we have some great posts comparing the two geographic markets of Green Bay and Pittsburgh (here and here). With Experian Hitwise, using the official team sites for the Packers and Steelers (assuming visitors to each site is a good proxy for a team fan) we can analyze the difference between the two.

At Experian Hitwise, we have the ability to report, in aggregate on the VantageScore bins for visitors to sites and categories (VantageScore is a grade-based credit scoring system, A = best credit, F= worst credit).

Here's the comparison of visitors to the Steelers' and Packers' websites over the last four weeks:

While the latest Vegas odds are still showing Green Bay favored by three. The spread in VantageScore for visitors to the two sites shows a much stronger gap. While this might not help predict Sunday's winner, I'm guessing that Packer fans have been able to finance better last-minute LCD TV purchases for the game.

Super Bowl, Search and Social

2011-02-04T14:56:49Z

I love this time of year. Anticipation builds around the match-up, Green Bay vs. Pittsburgh, Danica Patrick plus Jillian Michaels on the eagerly anticipated Go Daddy commercial, and whether the Super Bowl could cause a heart attack mid-game.

But beyond Packers vs. Steelers (I'll leave the Patrick vs. Michaels match-up for a post game analysis) there's another interesting match-up for my online data peeps; is social gaining on search as a navigational tool. First lets look at a chart of "Super Bowl" searches over the last three years.

First, we should note that over the last three years, U.S. searchers have realized that the Super Bowl is spelled as two words, not "superbowl." Also by looking at this chart you might guess that, by search volume, there is less interest for online Super Bowl content. A separate analysis of visits to the official Super Bowl site however, indicate that we're seeing a greater pregame surge in visits than in previous years. So why is traffic share on the term "Super Bowl" declining this year?

The chart above compares upstream traffic to Superbowl.com for this week versus the week before Super Bowl last year, you can see that traffic from all three search engines to the official game site have dropped, while traffic from Facebook has more than doubled.

Is this a sign that, as a navigational tool, social is gaining on search?

On a separate note, one of the reasons that I love working at Experian Marketing Services is the sheer volume of data that we have at our disposal, both online and offline. Check out these posts here and here from my esteemed colleague Bill Schneider on the depth of our market data comparing Pittsburgh to Green Bay.

How Dr. Phil Ruined My Diet Search Chart

2011-01-31T19:36:34Z

Dr. Phil, how could you? There are a few search term charts that I hold sacred, "prom dresses" and "diets" are my top two, in fact, I devoted a whole chapter of Click to "diet" searches and our New Year's resolutions. So this year as I prepared to report for a sixth straight year that January 1st was the peak for visits to online dieting sites, I was shocked to see that New Year's day was surpassed by the 5th of January!

I may not agree with everything the good doctor (Ph.D, not MD) says or does around his hit talk show, but I can tell you this; Dr. Phil is a diet promotion machine. Here's a brief postmortem on the January 1st diet spike.

This year, of the top 10 searches containing "diet" the number one search was "17 day diet" (variations on the term were also #6 and #9) followed by "hcg diet" and "atkins diet." Last year's #1 diet search "Jesus diet" (really I'm not making this up) didn't make the top 100.

Here's a chart on traffic to the top three searched diets:

Each of those massive spikes in visits to www.the17daydiet.com correspond to the diet being featured on the show. Interestingly, as far as the #1 place that searchers visited after typing "17 day diet" into a search engine, they didn't go to the diet's official site, or even Dr. Phil's site they went to Diets in Review.com (31.9%).

So diet book publishers take note, the new Oprah of weight loss book promotion is none other than her prodigy, Phil McGraw.

Dr. Phil, I'll let this one slide, but so help me, if you start promoting prom dresses on your show...

LivingSocial Closing the Gap

2011-01-27T18:46:11Z

Let's face it, $6 billion is a lot of money, especially considering the hyper-innovation we're seeing in the social buying space.

In a previous post less then two months ago we analyzed the difference between Groupon and LivingSocial visitors. At the time, visits to Groupon outnumber those to LivingSocial by a factor of 10-1. In that same post we commented that LivingSocial was a site to watch given its strong representation of early adopter visitors.

Fast forward to last week. LivingSocial has closed the gap, now with over half the visits of group coupon leader Groupon.

Sure last week's surge was due to LivingSocial's door-busting $10 for a $20 Amazon gift certificate deal, and as the New York Time's reported last week, the Amazon offer wasn't a traditional group coupon. In this case LivingSocial purchased certificates from Amazon at an undisclosed cost and then resold them through their service.

Despite some claims of fraud and issues with customer support, LivingSocials 80% traffic surge last week (and Groupon's 20% decline in the same time period) proves that the race for dominance in the group coupon space is far from over. With LivingSocial still exhibiting a healthy component of early adopters, Google launching a competitive offer and rumors that Facebook might enter the fray, I think I would have taken the $6 billion.

LivingSocial and the Adoption Curve

2010-12-02T19:58:15Z

The buzz continues to build around group coupon/social buying phenomenon. First Google's rumored offering of $6 billion for sector leader Groupon, and now, the Washington Post is reporting that Amazon is poised to make a $150 million dollar investment in competitor LivingSocial. According to the Post, this investment will put LivingSocial's valuation at $1 billion. Of course the question we're all asking is whether social buying is the next big play in our 2.0 world or is it tulipesque hysteria.

First a chart, the market share of visits to Groupon and its two nearest competitors, LivingSocial and Buy with Me:

Two observations. First, it appeared in Spring of this year, that Living Social's visit growth would make it a viable threat to Groupon. As the chart denotes, that growth-spurt was very short lived. Second, if we look at the graph's end-point, Living Social's market-share is almost an order of magnitude lower than Groupon's. So one-tenth of $6 billion dollars is less than Living Social's projected valuation. Of course traffic to valuation is a gross over-simplification. Perhaps Living Social's audience composition would validate a higher value.

Here's a breakdown by Mosaic type comparing the audience of Groupon to Living Social:

The main difference between the two audiences, is that Groupon has clearly "crossed the chasm" in Geoffrey Moore speak, with strong showings in types like Mosaic type C02 - Prime Middle America. Probably the most interesting difference is the percentage of Young Cosmopolitans - type H01 that are frequenting LivingSocial. From prior analysis we know that Young Cosmos are one the strongest early adopter segments. Despite the very unimpressive traffic numbers, perhaps Amazon is recognizing the potential for LivingSocial to move beyond its current early adopter phase.

VantageScore Provides New Insight Into Online Real Estate Visits

2010-09-22T22:52:09Z

News yesterday that housing starts jumped 10.5% to a four month high provided some hope that the National Bureau of Economic Research's pronouncement that the recession was so 2009, may in fact, have some validity.

We've been following the online component to real estate industry activity, our Business - Real Estate category. In the chart below, you can see that as of August 2010, visits to the Real Estate Category are down 20% compared to highs in 2008 and 2009. In fact, reviewing visits to the category over several years we can see that last year's visit pattern was markedly different that the typical pattern we see in 2008's data indicating that 2009 was clearly different than previous years. Regardless of visit patterns, the online housing market remains challenged.

One of the big challenges of making the jump from online activity to consumer sentiment and intent to buy or sell real estate is that an online visit to real estate could indicate the beginning of a real estate transaction or just an exercise in window shopping.

Today, we announced the release of our latest demographic tool, credit scores with VantageScore. Now we're able to view the aggregate credit scores of visitors to websites and industry categories (note that VantageScore scoring differs from FICO scores in that VantageScores range from 501-990 and are grouped by letter grade).

Here's the distribution of visits to the Real Estate Category over the last four weeks by VantageScore:

The following graph, shows the change in distribution of VantageScores over the last three months.

Looking at credit score data, Mid-June showed a little glimmer of hope, an inflection point for the highest ranked credit bin A(901-990). If we assume that those with higher credit scores are more likely to be purchasers versus window-shoppers, (the opposite of lower scored visitors) the graph above may indicate the return of serious buyers and sellers to online real estate sites. Stay tuned for more analysis.

Facebook - The Most Searched On Brand in the U.S.

2010-04-01T20:14:28Z

As I've discussed before, examining search term data across all engines, while providing excellent tactical insight for search programs, also can provide invaluable insight for brand marketers. We're now able to provide a new search metric beyond volume (search volume for a specific term over all term volume) and breadth (# of search term variations over all search term variations), our new metric which I'll call breadth volume, sums the volume for all variations on a root term.

Here's a table for the top brand searches (U.S. for week ending 3/27/2010) by breadth volume:

So in this example we see that when we combine the volume of searches for all queries that contained the root "Facebook" over the last week (there were 10,673 unique terms), those searches accounted for 2.8% of all U.S. searches.

Exporting our data into Excel, we can chart the changes in a brands "breadth volume" over time, and in this case compared to a competitive brand.

Stay tuned for some other exciting ways we can leverage our new breadth volume metric.

Haiti Earthquake Relief - News Sites are Major Traffic Driver

2010-01-15T22:34:48Z

With the devastating earthquake in Haiti on Tuesday, you would expect that traffic to charitable websites would register an increase. The chart below shows daily visits to the top 325 sites in our Community - Humanitarian websites.

While search continues to be the #1 driver of traffic to charitable websites, news websites registered the greatest increase as a source of traffic.

And which news sites were the top contributors of traffic on the days following the tragedy; Drudge Report 3.6%, ABCnews.com 0.2% and MSNBC 0.14%.

Bing - Rising Success (rate)

2010-01-14T15:59:21Z

Yesterday we published our monthly search volume statistics, and as expected Google continues to dominate the space with 72.3% of all U.S. searches executed, while Yahoo! Search and Bing shares were at 14.8% and 8.9% respectively.

A far more interesting stat is the positive movement that we've seen in Bing's success rate. At Experian Hitwise, we measure success rate as the percentage of executed searches that result in a visit to a site other than a main search domain.

Here's a time series of success rates for Google, Yahoo! Search and Bing:

In October 2009, BIng's success rate was well below the competition with a success rate just north of 70%. That rate has been steadily increasing, and now Bing is showing a success rate over 75%. There is one caveat when reading these numbers; search success rate is often influenced by the complexity of the search. Traditionally, portal search engines tend to have more simple navigational and brand related queries in their top searches which are easier to resolve then searches for information.

When examining the top 100 search terms from Google, Yahoo! Search and Bing, we find that nearly all terms are brand and/or navigational in nature for each engine. When we look at the percentage of searches that these top 100 terms account for on each engine, we find that Google is at 9.5%, Yahoo! Search is 15.1% and Bing is at 16.5%.

If we take navigational percentages into account, Bing would probably still trail Yahoo! Search in success rate, however, given the rapid improvement in this metric over the last three months, future success rate trends should prove interesting.

Facebook Hits #1 on Christmas and New Year's Day

2010-01-04T18:45:54Z

The main Google page has been ranked the most visited site on a daily basis for U.S. Internet users. That statement has held true for 364 days of 2009. Google's dominance of "all category" visits was disrupted when Facebook visits surged on Christmas Day. Facebook was able, albeit by a slighter margin, to recapture the #1 position on Friday, New Year's Day. Here's a chart showing market share of visits (U.S.) to the Two domains over the last two months.

Its interesting to note that Facebook's two #1 days both occurred on major holidays. Is there something about holidays that might cause social networking to trump search as the most popular online activity? One possible explanation might be that being home for the holidays might also mean being physically separated from one's social group, leading to greater virtual contact.

Another Interesting stat to close out 2009 is the most searched on terms across all major search engines in the U.S. for the year. Here's the list:

Top 10 Search Terms (U.S.) for 2009:

1. Facebook
2. MySpace
3. Craigslist
4. Youtube
5. Yahoo Mail
6. Google
7. Yahoo
8. eBay
9. Facebook login
10. Myspace.com
Source: Experian Hitwise

Along with its surge in visits, "Facebook" captured the #1 spot for most searched on term. In 2008 "MySpace" held the top spot, with Facebook in the #10 position. I still get a kick out of seeing "Google" in the search term list. Over the last four weeks 64% of "Google" searches were executed on Yahoo! Search and Bing, while 29% were executed on Google itself. I often wonder why anyone would Google "Google" on Google. I'm interested to hear your hypothesis.

News Corp. - If You de-Index Will They Still Come?

2009-11-23T22:03:01Z

Two weeks ago we posted on Rupert Murdoch's threat to block Google from Indexing News Corp. content. While at first it seemed as though Murdoch was merely posturing with hypotheticals, reports continue to indicate that News Corp. is seriously considering choosing Bing as the exclusive "indexer" of their news content.

While our data hasn't changed substantially since the last post on this topic, given continuing talks I think we should dig a little deeper into our search data. For brevity's sake, I'll restrict this analysis to U.S. traffic to WSJ.com and leave the analysis of other News Corp. properties and markets for another post.

As of last week, WSJ.com's referred and non-referred traffic from Google and Google News amounted to 15.3% and 11.0% respectively. Analyzing Google search terms driving traffic to the Journal, the top 100 terms accounted for over 21.6% of all Google search traffic to WSJ.com. Of that 21.6%, 13.4% were navigational or brand searches (e.g. "Wall Street Journal," "WSJ," "WSJ.com" etc...). Even if Murdoch decides to block Google, these navigational search queries will most likely remain intact.

Of the remaining 8.2%, the majority of searches were for stock quotes, and general business related searches. Most specific news related searches fill-out the long tail of search queries. While the Journal may lose traffic if it ceases to cooperate with Google the loss may be less then anticipated.

The potential loss of Google News traffic is potentially more serious. As reported here, over the three years, WSJ.com's traffic from Google News has grown from 2% to over 11%. As we see in the table below, the Journal is receiving more than double the traffic from Google News than newspaper sites overall (a custom category including national and regional papers). Bing, a potential News Corp. suitor for search exclusivity provides less than half of Google News' volume as of last week.

As newspapers continue to search for a way out of the search rip current, its hard not to root for Murdoch's maverick de-index strategy, that being said, the numbers bring us back to reality. As print continues to hemorrhage readership, could blocking your most significant traffic source be a wise choice?

News Corp. Google-less?

2009-11-10T00:04:36Z

Rupert Murdoch made an interesting statement in a Sky News interview today, when asked about news content available through search engines such as Google, Murdoch says he would consider blocking Google from indexing News Corp.'s news websites such as the Wall Street Journal. Of course it would naturally follow that I would immediately chart the amount of traffic that Google drives to Murdoch's flagship news site.

Here's the same chart from Google News to WSJ.com

In fact, on a weekly basis Google and Google news are the top traffic providers for WSJ.com account for over 25% of WSJ.com's traffic. Even more telling. According to Experian Hitwise data, over 44% of WSJ.com visitors coming from Google are "new" users who haven't visited the domain in the last 30 days.

While Mr. Murdoch makes some strong points in his Sky News interview regarding the plight of the news industry and the perils of making all content free, as clickstream data demonstrates - blocking Google could isolate the Journal from potential new online subscribers.

Thanks to Cheyne Winterton for the data hat-tip.