Web Analytics in Action: Alienware Case Study

This week, I’ll be taking a look at web analytics in action on a platform much bigger than my own humble blog. While I’ve been able to share personal experiences setting up and analyzing some free GA reports, I’d like to examine how a large scale e-commerce company uses web analytics to assess its site visitors and, ultimately, drive online sales. The company I’ll be profiling is Alienware, a manufacturer of uniquely-designed PC computers aimed at online gaming enthusiasts. Over the past decade, Alienware has built a solid and profitable customer base through responding to trends identified through web analytics efforts. Alienware’s success ultimately led to its acquisition by Dell in 2006 (Hachman, 2006). Unlike my previous analytics posts to date that have focused on Google Analytics’ free tools, we’ll be looking at tools Alienware used to track its customers through the paid services offered by the analytics tool Omniture. For the sake of simplicity, I’ll focus on a few key areas where Alienware’s web analytics translated into real business results. I’ll also dedicate a short section to my additional thoughts on ways Alienware could further employ web analytics efforts to increase its online sales.

Example 1: Star Wars Promotion

Alienware uses Omniture’s SiteCatalyst tool to view real-time customer interactions on its website, including clicks and purchases about each specific product it sells. This tool was useful during a 2005 promotion in conjunction with Lucas Films in which Alienware created and sold Star Wars-branded desktop PC models, giving customers the option to choose between a black “Dark Side” model or a white “Rebel Alliance” model (“Star Wars,” 2005). Through real-time web analytics monitoring – likely set up through some kind of specialized report to watch this specific promotion – Alienware noticed that its dark side models were outselling the light side models 15 to one. Because of the early detection of this trend, Alienware was able to seamlessly prepare its manufacturing, supply chain, and marketing teams to react to the lopsided orders for these products. (Omniture, n.d.)

Example 2: Black Friday and Cyber Monday

Also because of its web analytics efforts in November of the same year, Alienware had a hugely successful Black Friday through Cyber Monday weekend. The company set up alerts to watch for products that were selling above or below expectations throughout the week. By keeping an eye on sales in real time, Alienware was able to make tweaks to its site including which products were being featured (hot-selling products were featured more prominently) and which price points to use (those which received better response from customers were used more widely). Alienware saw a weekend sales increase of 105 percent as a result of its careful attention to customer interactions on its site. November 2005 also went down in history as Alienware’s best month to date, setting record-breaking site traffic, conversion rates, and revenue. (Omniture, n.d.)

Example 3: E-mail Marketing

Alienware used Omniture’s partnership with DARTmail to track the effectiveness of its e-mail campaigns. If you’re looking for a comparison, in the Google Analytics world, this is something that could be monitored somewhat through campaign tagging and tracking. The biggest added feature provided by DARTmail is the ability to identify each specific e-mail address associated with a click. Alienware certainly used this feature to help close sales through follow up campaigns. After an initial e-mail campaign promoting two high-end notebook computers, Alienware identified 36,000 customers who clicked on a link in the e-mail but never ultimately made a purchase of either computer, based on data collected through web analytics. A follow-up campaign was sent to these customers with a special offer as an incentive to close the sale on the computers offered in the initial campaign. This follow-up campaign resulted in $160,000 in revenue – a $4.40 revenue value for every e-mail sent. (Omniture, n.d.)

Additional Suggestions

Clearly, Alienware’s use of web analytics tools in Omniture’s suite provided the company with measurable success, ultimately leading to the company’s acquisition by computer giant Dell. Whether Alienware still retains some independence in its operations now is unclear. But assuming the company is still in control of its own website and analytics therein, I see a few areas where improvements could be made to further increase online sales.

First, I couldn’t find evidence of Alienware’s use of analytics to optimize its site design on an ongoing basis. Obviously, real-time response to customer behavior is important during high-volume, fast-turnaround periods like Cyber Monday and high profile product releases. But during all the other times of year, Alienware could stand to benefit greatly from using its web analytics capabilities to take a look at customers’ behaviors in a long-term capacity. Are there any pages where customers are dropping off at a high rate? Is there an ideal path through the site that Alienware would like its customers to take, and are they taking it? Are there ways to suggest add-on products to increase incremental revenue? It could be that some simple tweaks to overall site navigation could help Alienware increase its day-to-day sales, above and beyond the steps it is taking to react to big promotions in real-time.

Also, I found no reports of Alienware using filters to segment its audience and perform highly-specialized analysis of groups that might be of special interest. For example, Alienware could take a look at site visitors who find the page through a keyword search involving gaming computers. This keyword would indicate a visitor was within a highly-targeted segment to which Alienware should be paying special attention. Other segments could include visitors who are referred from sites that focus on the gaming audience and, if the web analytics software program provides it, demographic segments that meet typical gamer profiles according to age, gender, etc.  Site analysis based on these filters could help Alienware see whether its intended audience is responding to promotions, site changes, and targeted campaigns as expected. It may also show opportunities to take advantage of potentially lucrative gaps in current behavior.

Overall, I think Alienware did a great job of maximizing its profits as a result of web analysis of its current customers. Even though most of the data is a few years old and is centered around a paid analytical tool, it still provides great insight into the capabilities of this type of analysis when it’s done well.

References

Hachman, M. (2006, March 22). It’s official: Dell beams up Alienware. PC Mag. Retrieved from http://www.pcmag.com/article2/0,2817,1941376,00.asp

Omniture. (n.d.). Retail and commerce sites. Industry guide.  Retrieved from http://www.omniture.com/downloads/06_IndustryGuide_Retail.pdf

Star Wars: Revelations – Fan film sweeps net, Alienware ‘Star Wars Edition’ PCs. (2005, April 21). Enforcers HQ. Retrieved from http://www.omniture.com/downloads/06_IndustryGuide_Retail.pdf

Google Analytics is starting to make my head spin!

This week I spent several more hours learning about some of the advanced Google Analytic reporting tools that I have never been exposed to in the past. I have heard of goals and funnels before, but I’ve never investigated how to apply them or why I should even be considering them in the first place. Last week I discussed basic analytics reports and how they can be used. These reports are great for judging basic interest and user behavior on a website, but they don’t really give you a perfect picture of how your site is accomplishing your actual business objectives. This week, I’ll take a look at GA goals, funnels, filters, and profiles – a few more advanced tools that, as some experts say, are what you really SHOULD be tracking (Lofgren, n.d.). The tricky part, as I found through my research, is the actual implementation of these metrics. It turns out that more complicated analytical reporting means more complicated behind-the-scenes set up.

To be completely honest, before this week, I felt confident that I could take on a lot of web analytics functions on my own. But after watching a series of Google-produced videos about setting up profiles, filters, goals, and funnels, I have a new appreciation for the necessity and function of a true analytics professional. Although I couldn’t find any analytics experts being quoted online about how much time a person must invest in analytics for their education to be useful, I would assert that an amateur in the web analytics arena must make this subject more than a secondary focus in the workplace to provide truly meaningful results to his/her company.  That is, at my day job I NEVER would have been able to spare the hours I spent learning some more in-depth (and helpful!) GA tools.

Since I’m sure many readers are marketers on a small team, with a small budget, or for their own small business, I’ll save you some time by breaking down my newly-acquired wisdom about some of Google’s more advanced analytics tools and why they really are worth the time it takes to learn about them. If you like what you see here, I highly recommend finding a way to commit the time necessary to learning more about how to implement these tools for your specific business needs. Just keep in mind that a significant time commitment is required to fully understand these tools and to implement them effectively!

Goals

Goals are the best way to see if your site is meeting its business objectives. GA goals reports allow you to monitor certain activities or interaction levels that might be important to the success of your business based on one of four key metrics – URL, time on site, pages per visit, or event. I really like goals as a way to get an easy gauge of success on your site. Some of the objectives I might have for this blog could include some of the following:

1.     Having readers spend a certain amount of time on my page that would indicate they are interested and actually reading at least one article. I’d say a minimum of three minutes.

2.     Having readers click onto more than one story on my page, indicating that they’re interested in reading beyond the original article they find.

3.     Having readers sign up for e-mail updates when I publish new content.

All of these objectives can be clearly measured and monitored over time by setting up GA goals as follows:

1.     Set up a goal that will be achieved through visitors spending more than three minutes on my site.

2.     Set up a goal that will be achieved through visitors reading more than two pages per visit.

3.     Set up a goal that will be achieved when visitors arrive at the confirmation page after signing up for e-mail updates. (Note: To be completely honest, this e-mail update page is actually fictional, but it’s at least a good example of how this kind of goal could be easily tracked.)

Funnels

A function of GA’s goal reporting, funnels allow you to see whether your visitors are progressing toward a destination URL through a specific path. For example, if your goal is to get your site visitors to sign up for your mailing list, you’d want them to first click on the registration form, then persist through to submit their registration and arrive at a sign up confirmation screen. Your goal would be the final destination URL, the confirmation screen. But you could also set up a funnel that starts by tracking visitors on the registration form page, then ends at the goal URL, the confirmation screen. By seeing whether there is a larger than expected drop off between the two pages, you can determine if you need to make any adjustments to the registration form. You can also test improvements over time in an effort to increase the conversion rate, thereby increasing the total number of visitors achieving your goal of actually registering and arriving at the confirmation page.

Filters

GA filters allow you to look at your entire set of analytics reports for certain subsets of visitors. Depending on your specific business objectives, you may want to know monitor many dimensions of the behavior of a segment of your audience. For example, on this blog, there are a few different groups I’d like to know more about:

·      Visitors who aren’t me – I click on my blog from time to time to make sure things are posting appropriately. Since my site traffic is still pretty low, just a handful of my own visits could be skewing my overall analytics reports because I’m not using my blog in the same way a normal visitor would.

·      Visitors coming from referrals or organic traffic – I’m planning to do some better blog promotion in the future, specifically through social media sharing and improved SEO. I’d like to know how my visitors from these efforts are behaving, so I can filter out other audiences to see how this specific group is using my site.

The great thing about filters is that the possibilities are virtually limitless. You can slice up your audience however you want or need to. The bad thing about filters is that unless you are using the over-simplified, pre-packaged filters, you’ll need to get very technical, very fast! Filters are where GA starts getting quite a bit trickier, but the ability to get a full set of analytical data on a specific segment of your audience could be well worth it.

Profiles

Another GA element that goes hand-in-hand with filters is profiles. As you develop different filters for your site, you will apply them to profiles, and you’ll likely want to set up multiple profiles to do this. Because you can’t retrieve data that has been filtered out once a filter is applied, it is recommended to keep a version of your analytics data, a “profile,” that doesn’t have any filters applied to it. You’ll also want to keep each filtered audience in its own separate profile. Additionally, Google allows for multiple filters to be applied to each profile, so you could develop different ways of overlapping filters on multiple profiles to create very complex profiles. Again, here’s where GA gets quite complicated for the typical, non-technical user.

If I had a budget that would allow me to hire an analytics professional, I would do it in a heartbeat. Google goals and filters have shown me the edge of my confidence in my own comfort level. By doing my due diligence in learning about the capabilities of GA tools, however, at the very least I’ve enabled myself to speak intelligently when asking my IT department for help in setting up these tools.

References:

Lofgren, L. (n.d.) 4 Google Analytics goal types that are critical to your business. KISSmetrics. Retrieved from http://blog.kissmetrics.com/critical-goal-types/