"It's the most measurable medium in the world," they say; and it's true. In fact, there are so many different ways to measure the effectiveness of display advertising campaigns that marketers are bombarded with metrics -- everything from video pauses to video replays to custom interactions such as scrolling down the text.
Many marketers just give up trying to understand what these kinds of metrics actually mean for their business objectives. Instead, they resort to the basics (i.e., the click). Click-through rate (CTR) is not only easy to understand, but also easy to explain to your boss. That's especially crucial when you need to show some results for all the money you've invested in advertising.
The problem with this simplistic approach is that while clicks are intuitive and generate valuable traffic, they require active participation from the user and the link between CTR and effectiveness is not actually grounded in research. In fact, research shows that online advertising is effective even at low CTR.
Indeed, comScore maintains that the primary value of online ads is the exposure itself and not necessarily the act of clicking. In a study titled "How Online Advertising Works: Whither The Click?" comScore has shown that two-thirds of internet users do not click on any display ads over the course of a month and that only 16 percent of internet users account for 80 percent of all clicks. Furthermore, clickers tend to be younger and less affluent than non-clickers. The comScore research confirmed that there is latency effect and branding value to online advertising, driving users to the advertiser's website even without clicking.
The research by comScore also indicates that display advertising has an effect on user behavior even at low CTR. In the research, which included 139 display campaigns from seven verticals, comScore recorded substantial effects on traffic, sales and branding despite low CTR. The campaigns yielded a 46 percent lift in advertiser website visits over a four week period. During the same period, exposed users were 38 percent more likely to conduct an advertiser-related branded-keyword search, and 27 percent more likely to make a purchase online. Furthermore, exposed users were 17 percent more likely to make a purchase at the advertiser's retail store.
So if clicks are not the quick and easy campaign measurement for online campaigns, what can you tell your boss? Here are five metrics that are simple yet comprehensive enough to effectively measure your campaign performance.
The research by comScore shows that exposure to an ad is one of the major contributors to increase in sales, site visits, and branded keyword searches. Therefore, reaching as many relevant people as possible should be a prime goal for every campaign. So, how do you measure how many online users you have actually reached?
Until recently, reach was typically measured with a cookie-based approach -- ad servers counted the number of unique cookies placed on the browsers of users exposed to the campaign. The problem is that if a user deletes the cookie, a new cookie is placed in the browser, and the user is counted again.
According to comScore, 33 percent of users delete their third-party cookies (cookies used by ad servers) during a one-month period and that each user is counted an average of 5.1 times during that same period. The prevalence of cookie deletion means that simple cookie measures of reach largely overestimate the correct number of unique users.
Therefore, measuring unique users requires a metric that is adjusted for cookie deletion. Ad servers offer "adjusted unique" metrics that compensate for cookie deletion and prevent the over-counting of users exposed to your campaign.
Campaign size and reach in the United States
So how many impressions do you need to reach 10 million users at least once based on these metrics? In the U.S., it would be about 100 million impressions, on average. In Germany, France, the U.K., and Spain it would take about 120 million impressions.
What is engagement, and how can we measure it? To measure the proportion of impressions that consumers engaged with and for how long, we use a metric called "dwell." The metric measures the proportion of rich media impressions that were intentionally engaged by mouse-touch, interaction or click as well as the duration of the engagement. Unintentional dwell (lasting less than one second) is excluded.
This measure of engagement provides an estimate of the share of impressions that were, with high likelihood, seen by users, as a user's natural tendency is to follow the mouse cursor movement with their eyes. Dwell measures the proportion of impressions that had a meaningful mouse-touch, lasting more than one second. While there are undoubtedly users who see an ad without a mouse-touch, dwell allows us to gauge the number of users that are very likely to have seen the ad.
Dwell rate and duration
Research by MediaMind, Microsoft Advertising, and comScore shows that engagement does have an effect on brand metrics. Results of the joint study indicate that users who were exposed to campaigns with high dwell are three times more likely to search for brand-related keywords as users who were exposed to campaigns with low dwell. Moreover, campaigns with high engagement boosted advertisers' site traffic by 69 percent and improved brand engagement --increasing page views and time spent on the brand's site.
Lift in site visits
Site traffic is another gauge of advertising effectiveness, reflecting users (not all of whom clicked) who were enticed by your banner to seek more details about your product or service.
For example, one gaming advertiser we analyzed wanted to measure the full effect of a campaign that achieved a relatively low CTR. When we analyzed site visits, we saw that the campaign actually had a significant effect on the number of website visitors. Before the campaign, this advertiser had 6,000 visitors on average every day. When the campaign was running, the number of average daily visits increased to more than 10,000 -- an increase of 60 percent.
Note: This analysis is effective for campaigns that run online exclusively, as attribution of site visits would be a very difficult task for a multi-channel campaign.
Search and display overlap
Our search-display overlap research revealed that one of every five users that arrived at an advertiser's site was exposed to a display ad.
Measuring search activity can be a good way to assess if a display campaign is performing. For example, if users are searching for the tagline from your campaign, product name, or other branded keyword, you can be confident those terms are resonating with users. If there is no increase in branded keyword searches during the campaign, it may be time to revisit your ads.
Conversions and ROI
ROI is how every other investment in a company is measured, so why not advertising? Unfortunately, if you're not selling all your products online, ROI can be very hard to calculate. The conundrum is spillover from online advertising to offline brick and mortar shopping. In these cases, the investment is online, but the return is attributed to other channels.
Online-only retailers who can capture the full effect of online advertising should by all means track ROI. To calculate ROI, sum-up all dollars coming in from sales attributed to your campaign and divide by campaign cost.
What if you only collect leads or applications and do not sell anything online? Then you can either calculate a "notional value" for each conversion (leads or applications), or track your investment on advertising by calculating the average cost per conversion.
Measuring the most measureable medium should not be convoluted. Clicks are simple and intuitive, but are proving to be ineffective for measuring real results.
These five simple metrics offer a more telling alternative that your boss can appreciate. Depending on your campaign objectives, reach, engagement, lift in site visits and search activity, conversions, and ROI cover all the angles you need for confident campaign measurement.
On Twitter? Follow iMedia Connection at @iMediaTweet.
"Business chart collection 1" image via Shutterstock.