Steve DiPietro is amazed at how frequently he listens to prospective clients parroting clickthrough percentages, Web traffic statistics and conversion ratios with great enthusiasm but little-to-no understanding of their value to their organizations. Increasing a conversion rate from 12 to 15 percent can become a goal unto itself as marketers immersed in number crunching can lose sight of the fact that sales aren’t also growing.
Making Sense of Metrics
ALGORITHM: A set of mathematical equations or rules that a search engine uses to rank the content contained within its index in response to a particular search query.
ANALYTICS: Technology that helps analyze the performance of a website or online marketing campaign.
BENCHMARK REPORT: A report used to market where a website falls on a search engine’s results page for a list of keywords. Subsequent search engine position reports are compared with that.
CHARGEBACK:An incomplete sales transaction that results in an affiliate commission deduction. For example: merchandise is purchased and then returned.
CLICK & BYE: The process in which an affiliate loses a visitor to the merchant’s site once they click on a merchant’s banner or text link.
CLICKTHROUGH: The process of activating a link, usually on an online advertisement connecting to the advertiser’s website or landing page.
CLICKTHROUGH RATE (CTR): The percentage of those clicking on links out of the total number who see the links. For example: If 20 people do a Web search and 10 of those 20 people all choose one particular link, that link has a 50 percent clickthrough rate.
CONVERSION RATE: The percentage of clicks that result in a commissionable activity such as a sale or lead.
CONVERSION REPORTING: A measurement for tracking conversions and lead generation from search engine queries. It identifies the originating search engine, keywords, specific landing pages entered and the related conversion for each.
HIT: Request from a Web server for a graphic or other element to be displayed on a Web page.
IMPRESSION: An advertising metric that indicates how many times an advertising link is displayed.
KEYWORD: The word(s) a searcher enters into a search engine’s search box. Also the term that the marketer hopes users will search on to find a particular page.
PAGE VIEW: This occurs each time a visitor views a Web page, irrespective of how many hits are generated. Page views are comprised of files.
RANK: How well a particular Web page or website is listed in a search engine’s results.
UNIQUE VISITORS: Individuals who visited a site during the report period – usually 30 days. If someone visits more than once, they are counted only the first time they visit.
“It’s sad and somewhat surprising that after all this time there is a pervasive lack of understanding … of how these numbers correlate with how to make money,” says DiPietro,who works with clients large and small as the president of the Marlton,N.J.-based DiPietro Marketing Group.
Many marketers continue to rely on basic campaign performance data as the primary or even sole metric for measuring success, according to DiPietro. People often get caught up in the measurability of online campaigns and miss the ultimate corporative objective of a marketing campaign – to increase profitability.
Despite many marketers’ incomplete understanding of how buying keywords affects the bottom line, search marketing spending continues to grow rapidly. According to a survey conducted by the Search Engine Marketing Professional Organization (SEMPO), advertisers in the U.S. and Canada spent $5.75 billion on search engine marketing in 2005, up 44 percent from 2005. Search engine marketing spending in North America is projected to reach $11 billion per year by 2010.
Some marketers whose careers started in the brick-and-mortar world have seemingly become spellbound by the top-level data for measuring marketing campaigns and forget their “old-school” fundamental tenets about increasing sales and stockholder value, according to DiPietro. Finding methods of doubling the conversion rate of a keyword campaign is admirable, but who cares if sales don’t grow? Estimating the value of a keyword purchase by focusing on clickthrough rates or increasing traffic to the website is an easy way to justify spending, but may be totally meaningless, DiPietro says.
The clickthrough ratio is analogous to the batting average in baseball – it is easy to compute and understand, and therefore is the most relied-upon statistic. However, during the past few decades, baseball executives such as the Oakland A’s Billy Beane, who probe deeper into statistics, have learned that other metrics – such as on-base percentage – are more directly related to achieving the objective (scoring more runs). The A’s have managed to succeed while spending considerably less than competitors, and many fellow baseball executives now are looking beyond the batting average. Similarly, marketers who identify the metrics that more closely correlate to their specific goals can increase their success.
Getting customers to your website is an important first step in increasing revenue, but determining the return on the investment requires analyzing what happens after they arrive at your doorstep. “You must have an action attached to [increasing traffic] or the campaign is useless,” says Douglas Brooks, vice president of consulting firm Marketing Management Analytics.
Before embarking on a campaign, marketers must define the objective – be it increasing leads, sales or brand recognition – and apply the appropriate metric, according to Brooks. The most appropriate metric may depend on whether the company is focused on e-commerce sales or if sales staff is usually involved in any transaction. Different yardsticks are appropriate for companies that use their website as a direct sales channel than for companies who are focused on generating leads that are converted off-line, he says.
Companies that rely on sales personnel should look at the volume of leads a campaign generates, according to Jerry Moyer, manager of analytics at interactive agency Refinery. Moyer says he tells his media clients – many of whom continue to focus on clickthrough rates – that tracking leads is a more effective barometer of campaign performance.
Campaigns that drive traffic to a website that cannot identify where visitors came from may be over- or underestimating their effectiveness, according to Moyer. By using first-party cookies and analyzing all of the activities that occur over time, advertisers can better understand the value of the leads generated.
Using cookies enables marketers to identify the unique visitors, according to Andrew Hanlon, who owns advertising agency Hanlon Creative. Cookies enable companies to track how many times a visitor was exposed to messaging during an entire campaign, as well as counting the total number of interactions on a website before visitors enter personal information and become a lead. “Unique visitors is the most raw level of success; you have to consider how many [leads resulted],” Hanlon says.
For example, Designer Linens Outlet implemented first-party cookies and saw revenue from returning customers increase by 45 percent and shopping cart conversions increase by 20 percent, according to Web analytics firm WebTrends, which managed the campaign.
Measuring the quality of leads is as important as the clickthrough ratios or total Web traffic generated by a campaign, according to Hanlon. He says many of his Hatboro, Pa., agency’s clients ($20 million to $1 billion in sales) “rarely know what they are asking for” when trying to gauge the impact of campaigns on sales.
He stresses to clients the importance of tracking leads throughout the entire sales process. “The client has to be able to act on the data – what happens with the lead after it is collected,” he says. The ability of keywords to generate leads varies widely, says Hanlon. Marketers should use metrics that create quality leads versus those that merely drive traffic.
If branding is the goal, then measuring increases in traffic can be appropriate since many keywords generate low-quality leads, Hanlon says. Companies looking to reinforce messaging through multiple media should consider several online metrics, according to Jason Palmer, vice president of product strategy at WebTrends.
Some campaigns are incorrectly viewed as ineffective because of low conversion rates, according to consultant Hanlon. Landing pages that were not designed to entice visitors to delve deeper into a website could turn away potential leads, so their effectiveness must also be evaluated. Landing pages should have interesting content such as blogs or unique offers to encourage clickthroughs, says Hanlon. Companies should measure conversion ratios after visitors hit a landing page, and if they are shown to be “dead ends,” they should revise the landing pages to add more content, he says.
Software companies including WebTrends and Salesforce.com are developing applications that zero in on landing-page performance. For example, Webtrends Dynamic Search evaluates the effectiveness of the landing page and keyword in matching specific company objectives.
Tweaking the content of a landing page can increase the percentage of clicks converted to leads by as much as a factor of 10, according to Kraig Swensrud, senior director of product marketing at Salesforce.com. Tracking and improving landing-page conversions is equivalent to increasing money spent on Google AdWords, he says. “Everything is interconnected – as soon as you have visibility on [landing-page] conversion rate, you can impact change,” says Swensrud.
FROM CLICKS TO SALES
The best metrics link gains in Web traffic or clickthrough percentage to the overall business objectives – increasing sales, profitability and effect on the stock price. Consultant DiPietro recommends the break-even sales analysis is applied to off-line marketing should be applied online. Companies should calculate how many sales – based on the profit margin per average sale – would have to be generated to determine whether or not a campaign is a good investment.
“Whether it’s participating in a trade show, setting up an affiliate program or a PPC campaign, work it back to break-even sales,” DiPietro says.
Connecting the dots between Web analytics and sales data has been largely a manual process for DiPietro, who spends more hours than he would like handcrafting spreadsheets to complete his analysis.
Web analytics firms such as WebTrends, Omniture and WebSideStory are addressing this software void with applications and services that can link Web and sales data to simplify calculating the return on investment. These applications can incorporate Web data such as traffic analysis, email marketing and search marketing performance with customer relationship management sales data.
Accurately gauging the value of a campaign to a company’s bottom line, tracking a visit as it becomes a lead and until the sales cycle is completed is what should be measured, according to WebTrends’ Corey Gault. Web data should be combined with higher-level key performance indicators (KPIs) such as cost per visit, cost per lead and cost per sale, he says. “KPIs can also be combinations of various metrics, such as revenue dollar per marketing dollar spent, or percent of orders from repeat purchasers,” says Gault.
Measuring the lifetime value of online branding campaigns is challenging for companies that also sell off-line, as the ability to automate the process ends at the desktop. Refinery’s Moyer says customer surveys are an efficient method to link online with offline impressions. The surveys incorporate data collected by contacting customers about their behaviors before and after campaigns, and factor in both online and off-line (broadcast, print, outdoor) impressions. This enables companies to calculate how the campaign contributed to the overall sales effort, he says.
Metrics should factor in all of the times a company interacts with the customer, not only the most recent, which can skew performance data, according WebTrend’s Gault. “Many marketing analytics solutions credit the conversion to the last campaign touched, effectively undervaluing all the programs that initiated awareness and consideration.”
Vendors are also re-engineering their products so that sales data can automatically be integrated with Web analytics to complete the campaign-to-revenue analysis.
“The ability to tie marketing metrics with sales metrics is one of the biggest problems that customers have,” according to Salesforce.com’s Swensrud. To address the difficulty in understanding the impact of keyword purchases on sales, the company introduced Salesforce for Google Adwords late in 2006. The software, which is sold as a service, traces the leads generated by keyword purchases and follows them through the sales process to determine their return on investment.
Although comparing current campaign- to-revenue performance with historical data is informative, marketers should create a baseline of return on investment so that they understand the relative value of each type of online campaign.
The cost per thousand of a keyword campaign may seem relatively low when compared with cost of an email marketing campaign. However, determining the return on investment of each can justify what appear to be higher costs per customer contact, according to Marketing Management Analytics’ Brooks. He says calculating the individual return on investment for each type of online campaign enables an apples-to-apples comparison.
For example, the lifetime value of a customer acquired through keyword buys might be a fraction of that of someone originally contacted via email. After factoring in revenue, marketers can better decide the best marketing mix for their collective media expenditures.
The volume of statistics contained in monthly Web analytics reports can make it a challenge to interpret the metrics that matter most. The bottom line: Don’t forget about the bottom line.
JOHN GARTNER is a Portland, Ore.- based freelance writer who contributes to Wired News, Inc., MarketingShift, and is the editor of Matter-mag.com.