From the Editor

Thanks to the hard work of Kent German, Bonnie Cha, and Nicole Lee, CNET's reviews of mobile handsets are unparalleled: we are the place on the Web to get in-depth evaluations of nearly every handset offered by the major carriers.

But we're not satisfied because we know you're not.

Judging a phone by its features, design, and quality is just half the story when determining whether it is worth purchasing. The other half of the equation is the quality of service provided by the carrier--and that's a factor we have not yet provided. After failing to find a solution for many years, we've finally arrived at one that I expect will rattle the market.

We've partnered with a company called Root Wireless, which is led by some cellular-industry veterans, to provide consumers with detailed, real-world coverage of major U.S. cities.

(Credit: Root Wireless)

With Root, we're now collecting data in eight major metropolitan markets: New York, Los Angeles, San Francisco, Seattle, Chicago, Washington, Dallas, and California's Orange County. Once enough data is sampled, the information will be available on CNET for cell phone buyers to determine the quality of service around their home, on their way to work, and at the office.

The information, being collected for the four major wireless carriers--AT&T, Sprint, T-Mobile, and Verizon Wireless--covers three key service concerns:

1. Average signal strength and the number of signal bars for the selected carrier in a specified area
2. Data connectivity and throughput
3. Network issues such as drop calls or failed data connections

This is no small effort. It entails installing software on a variety of phones and collecting real data on those key metrics. The information--which includes time of day, terrain, and number of users in an area--is transmitted to a database and made available via a Web interface. Users can drill down to see how the big carriers perform in their neighborhood or zoom out to view the coverage on their commute route.

We're doing this because we know there's a huge need. We recently conducted an online poll of our users in which 26 percent of you said you had returned a cell phone due to poor service, and nearly half of you feel that there is not enough information available on service quality.

It may be premature to write about this now, but I couldn't wait. I've seen the working prototypes, and I know that this is going to be big hit for our users. Rather than rely on anecdotal evidence from friends or user reviews, or even from CNET's editors, you will have empirical data on which to base your purchasing decisions.

We have collected data in those eight markets so far, and we'll be beta-testing for the remainder of 2009, with a general release as early as January 2010. And don't fret, if your city wasn't listed in the first batch of eight because we'll be adding many more areas in the next year.

At CNET, we take great pride in the quality and thoroughness of our reviews. We know they play an important role in helping millions of consumers to determine which tech products to buy and which to shun.

But we're also realists. We know that a CNET review is just one part of the research that goes into picking out a new phone, laptop, or TV--especially in this economy. The other sources you're likely to turn to include other review sites, manufacturer information such as specs, reviews from other users, and advice from friends.

For years, CNET has had you covered on all those points except the part about friends. Well, we're aiming to change that.

Imagine that you're looking for a new laptop and you ask your friends what they own, what their experience has been with that manufacturer, and whether they would buy the same machine again. Chances are the responses will be useful, to a point. It's likely your friends have not purchased a notebook in a while, so they have little insight into the current crop. And their experience with breakage and customer service--whether it's good or bad--is valid but anecdotal.

This is where CNET can help, if you're willing to stretch your definition of "friend." Each day, hundreds of thousands of like-minded people click around our sites doing research and deciding what to buy. If you could see what these people are viewing most often, wouldn't you find that to be valuable information? It would be like walking into a Best Buy or Wal-Mart and seeing a throng of people gathered around one or two laptops on display. Of course you would join the pack to see what the buzz is about: is it a low price, hot new hardware, or both?

The online equivalent to seeing what people are checking out in a store is CNET's Business Intelligence group, which crunches a "Matrix"-worthy amount of data about activity across CNET's many sites. Their reports are typically used by the sales team to demonstrate the power of the CNET brand and its influence on buyers. Now, however, we are taking the first steps to making that information available to you as one of the tools we provide to meet our core mission: connecting buyers and sellers.

CNET tracks what it calls "considered users," defined as visitors who clicked on a product review, like this one of a new Sony Vaio, and/or on a pricing link on the review page or elsewhere on CNET. In industry parlance, the people reading product reviews, checking out specs, viewing a video, and so on, are described as being "up funnel" because they are thinking about buying something but are still in research mode. "Down funnel" users are those who have clicked on a pricing or merchant link, because they are that much closer to making a purchase. In either case, they are all "considered users" because they are considering a product.

What our data shows is that the number of considered users looking at notebooks climbed 31 percent from July 2007 to July 2008 and--despite Depression 2.0--the number of these users climbed another 25 percent in the past 12 months (sorry, the data guardians at CNET won't let me publish the actual totals but I'm working on them). That's healthy growth that could be the result of several factors. As the site editor, I'll list the two I think are most plausible: our superb laptop editors, led by Dan Ackerman, are cranking out more reviews than ever before, and our SEO team has made sure their reviews rank extremely high on Google and other search engines. Even if people aren't buying as much in this economy, they are certainly researching for when they are ready.

The macro numbers are interesting, but the more intriguing stuff can be found in the micro picture. Within the notebook category, the two-year trend lines reveal how the major manufacturers have fared versus their competition. For example, in July 2007, Dell products or prices were viewed by 40 percent of all considered users on CNET, but that number dipped to 25 percent last summer and it has continued to sink in 2009 to less than 20 percent.

Who benefited from Dell's tumble? Hewlett-Packard began at 24 percent in July 2007, slipped to 20 percent by the spring of 2008, and now sits at 30 percent, giving it the largest share of considered users of any notebook maker on CNET.

The big winner has been Asus, which Ackerman calls "masters of the Netbook." And right now, Netbooks are hot on CNET. After unveiling its first Eee PC in late 2007, Asus has climbed from 2 percent in July 2007 to 13 percent a year later to 20 percent in July 2009.

Apple tends to hover at about 10 percent, but its share can spike to 15 percent or 20 percent when new models are reviewed. Other manufacturers also see such spikes around new products, but they aren't as pronounced. (See the above chart, which tracks the major manufacturers and calls out newly published product reviews that resulted in traffic spikes.)

U.S. notebook shipments by manufacturer

Of course, this data is only part of the picture, and one obvious question is how closely CNET's numbers track actual sales figures. For those numbers I'll turn to IDC, which issues quarterly shipments for the major notebook makers, and compare four notable names:

HP: IDC shows HP hovering at around 25 percent for the past couple of years, with a jump to about 30 percent in the first quarter of 2009. CNET also shows HP tracking at around 25 percent until October of last year, when it topped 30 percent (where it has remained since).

Dell: IDC shows Dell peaking at nearly 28 percent during the last three months of 2007 and then steadily drifting down to 23 percent in the first quarter of this year. CNET shows Dell at about 30 percent in the final quarter of 2007 before sliding to between 24 percent and 26 percent during the first three months of 2009. Since then Dell has slid further to its current 19 percent.

Apple: IDC has Apple ranging from 7 percent to 11 percent for the past couple of years. CNET also shows Apple in a similar band, with the notable exception of some sharp monthly spikes around the publication of new product reviews (which typically occur within a day or two of a new notebook being released on the market).

Asus: According to IDC, Asus has risen from almost nothing a couple of years ago to 3.5 percent in the first quarter. CNET's monthly numbers for the first three months show Asus running between 7 percent and 10 percent, and as of July it was up to 20 percent. (It's my guess that CNET is a leading indicator when it comes to the popularity of Netbooks and that Asus will show a jump in IDC's second-quarter numbers--perhaps not as dramatic as the 16 percent to 20 percent seen on CNET, but a nice gain nonetheless.)

As you can see, the trends observed by CNET and IDC are similar. The key difference is that CNET is tracking people while they research and shop, and IDC is following actual shipments. CNET is also compiling stats each month, versus quarterly for IDC. The two reports together are very complementary, but CNET is tracking the consideration phase, so we theoretically can spot trends before they manifest in actual sales, and sooner.

As a result, we think you will find the CNET numbers to be another piece of the puzzle when trying to determine which notebook to buy. Just like a single product review, or user opinion, or even a friend's recommendation, they aren't the whole or the final answer. But they are useful for seeing in a very timely way which notebook manufacturers are getting most of the attention on our site and how they are faring each month. (CNET also produces lists of specific products that users are interested in, but this new data focuses on manufacturers to provide another view into the current market.)

For now, our foray into this new area is modest--just this blog post and the above charts. In the near future, we will be injecting this data into the site in various ways. For example, the product-filtering tool on the notebook category door and elsewhere will show how certain manufacturers are trending on CNET (the manufacturer choice is the most commonly used filter in this tool, so we know readers are concerned about who is making their hardware).

After notebooks, we'll be collecting the data for the other key categories on CNET, particularly TVs and mobile phones, and coming up with novel ways to incorporate the information into the site. For example, we can see which phones are popular with people who live in your area.

As a potential buyer or follower of the tech industry, please leave a comment below on the type of data you would find useful and how you would like to see that information presented on the site: in monthly blog posts, trend lines on the categories doors, or as part of our product-sorting tools?