Use Web Analytics to Maximise Search Marketing Results

By: Dana McGraw, Bruce Clay, Inc., May 2006

Search Engine Marketing (SEM) has emerged as one of the most effective marketing strategies available. Today, one out of every two online purchases originates from a search site (DoubleClick 2005). Search spending is on the rise, with total revenues reaching $5.75 billion in 2005 and projected to reach $11 billion by 2010.

Search marketing works because people pre-qualify themselves with their keyword queries in the search box. SEM is powerful for generating desired actions because it captures people at their point of interest.

Search marketing generates sales leads, newsletter subscriptions, site registrations, online/offline sales and even branding. It can drive substantial profits for your business, but you need to know the value of your marketing activities to link marketing objectives to business objectives. The only way to do that is through web analytics.

Marketers need to deliver the right ad to the right audience in order to produce revenue. It should be easy to display a relevant ad with persuasive copy, matching consumers with the products and services they seek. But it's not that simple.

Help Is on the Way

It is difficult for today's marketers to answer all the business and technical questions posed by executive management. Marketers are faced with issues such as:

• Distinguishing between profits from paid and organic search
• Measuring visitor behavior beyond a click-through
• Plugging leaks in the conversion funnel

Maximising your success with search marketing is a challenge in today's ever-changing search environment. Here's how web analytics can help uncover solutions for the above issues.

Paid Vs. Natural Search

A few years ago, it was reported that most searchers didn't know the difference between organic and sponsored links. Perhaps today more people are aware that these are different. Nevertheless, it is important to be able to distinguish between paid and organic search traffic so you can learn the effectiveness of those separate marketing campaigns.

When traffic comes to your website from search engines, the keyword phrase used for the query is normally included in the URL query string; but there is no distinction between paid and organic clicks. That's because the referring URLs from the search engine to your website are the same regardless of whether a user clicked on a paid or organic listing.

Fortunately, there is a way to distinguish between paid and organic links. The solution is to add a unique tracking code to the destination URL with campaign and/or search term information for each paid search term. Then, voilà! Once the ad is clicked, the information for the paid search term is transferred within the URL, letting you know that this click comes from a sponsored link, not an organic one.

Your web analytics program should be able to identify the profit resulting from either sponsored or organic search marketing campaigns. This data can be used to determine your return on advertising spend (ROAS) and return on investment (ROI) from your search engine optimisation (SEO) or pay-per-click (PPC) campaigns. There is a belief that if you have matching paid and organic search terms, you can maximise your profit by reducing the paid campaigns that cannibalise your robust organic search positioning. Others believe that you are likely to get more clicks from combining SEO and PPC because you are "dominating" the listings. That is something that you can test to determine the validity.

Beyond the Click

It all begins with a click, but today's marketers are really looking for conversions. With the escalating cost of paid search advertising, it is important to measure beyond the click because every search term has a break-even point. If you keep bidding higher to stay on top without this information, you can go broke.

A successful PPC campaign is based on buying the right keywords, creating persuasive ad copy and providing a good landing page. However, this is all moot if you don't know your break-even point for keyword bid management. It is important to bid on keywords based on the profits they generate. Last year, JupiterResearch reported that 75 percent of search execs did not bid based on revenue. You don't want to make the same mistake.

Usually, when a user clicks on your paid listing in a search engine, the engine records a click in their system and you are billed. This reporting alone allows you to generate reports on click-through-rate, cost-per-click and average position.

A good web analytics program can help you generate additional reports that tie keyword clicks to downstream events like leads, purchases, page views and even revenue. This data allows you to calculate performance metrics like conversions, revenue-per-click and page views-per-visit. With such post-click activity data, you can make informed and educated decisions about which terms are converting and which ones are not.

Your cost data is in the search engine's system, and the downstream performance data is in your analytics package. The two need to be combined to measure ROI or ROAS. There are two options for doing this.

• Pull the data into an Excel spreadsheet. This works for smaller campaigns of a few hundred keywords but is not effective for large campaigns with thousands of keywords.
• Extract the data from the search engines and upload into your analytics program. With this option, you can combine keyword-serving metrics like impressions, cost and average position with revenue, leads and page views. This allows you to analyze ROI at the keyword level, making decisions based on profits.

Plugging the Leaks

Average online conversion rates range from 2 to 10 percent. This means that most online visitors do not convert. They bail out somewhere along the way. It could be right on the landing page or even just before confirming the order. (How many times have you done this - changed your mind and abandoned the shopping cart?) If you merely measure click-through-rates you cannot identify and plug the leaks in your conversion funnel.

A web analytics program helps you solve this challenge. While the clickstream data will show the click path of each and every visitor, this information alone is hard to use because visitor paths vary and you'll see thousands of different paths.

Search marketers need a higher level of path analysis called fall-out or conversion funnel reporting. Fall-out reporting allows you to select major checkpoints, identifying major leaks in your conversion funnel.

Lower Risk and Maximise Profits

Marketing decisions can be risky, but a good web analytics program lowers that risk while maximising profits. It can help you address suspected shortcomings, resulting in continuous improvement of your website and your marketing campaigns. Continuous improvement can be accomplished one step at a time by tweaking your paid search campaigns, making suggested site changes for better organic rankings, removing the obstacles in your website conversion funnel and optimising your landing pages to encourage conversions.