We all recognize that search is a powerful and critical part of every media plan. But search media is fundamentally different from other types of media. One of the key differences is that the majority of search impressions are free and this is radically different from broadcast, print or online display media. In each of those formats, advertisers pay on a per impression basis and we typically look at costs from a CPM standpoint (cost per thousand impressions). We are in effect paying to get our ad in front of a guaranteed number of consumers.
Search, however, is different. Let’s say that we are an auto manufacturer, such as General Motors, and we have multiple brands and each of those brands has its own website with a distinct URL or Web address. We can bid on the term “SUV” as part of an integrated search strategy serving up links to multiple brands and local dealer websites. If a shopper goes to Google and types in the word “SUV” we can list different brands and dealers all on the same page in the space reserved for paid search. For Google this is the three sites listed on the top of the page and the five to eight sites listed in the right hand column of the page. If a user clicks on any of these links the advertiser will get charged for that “click.” All the other paid search listings or impressions on that page that aren’t clicked on are in effect free. So a natural question is–Do these “free” search impressions matter? Do they help my brand? And if they do, can I quantify their value?
Google provides many of its advertisers with whitepapers and survey data that shows without a doubt paid search impressions improve many key branding metrics, especially when the advertiser is top in natural and paid search results. Google even asserts that presence in paid search decreases branding metrics for your direct competitors! Is there any other data available to support these claims? We will come back to these questions in a minute, but first we need to understand a key measurement difference between display media and search media.
Online display has one big measurement advantage over paid search channel and that’s the ability to track “view through” performance. This is the idea that any user who sees an ad served by a major ad server is cookied at the impression level. That means if you see a display impression (you don’t need to click on it or interact with it) the advertiser will be looking at your behavior after that impression and report on the results. This creates “view through conversions” which can account for 90%+ of all conversions for many large advertisers’ display campaigns. And view through data is a key input to any attribution model that is designed to accurately share ROI credit across multiple digital touch points.
Paid search activities, however, have no equivalent “view through” metric–only clicks are tracked for search. So, for large direct marketing advertisers the search impressions are largely ignored because they are difficult to directly measure. And Google and Bing don’t have any plans to allow advertisers to cookie search impressions any time soon.
Now, back to our questions about the value of search impressions. Just because search impressions are difficult to directly measure doesn’t mean we can’t employ some creative modeling solutions to gauge their impact. There are several statistical techniques we can use to measures correlations between search impressions and other online behavioral data.
Looking at a large combined data set of search activity from multiple business verticals we have uncovered some key insights that do support the assertion that increased search impressions have a demonstrable value on overall business performance. Let’s take a closer look at some aggregate data.
We have participated in several paid search “own the page” strategies. For certain keywords we occupied 75% of the available paid search impressions on the page with multiple products for several brands that drove to distinct URLs. During a multi-month period we saw a 250% increase in search impressions. Yet during a the first month of these combined initiatives, overall site visits dropped by nearly 10% yet paid search traffic rose substantially at 80%+.
Most importantly, these brands saw a major shift in the overall engagement score for traffic sourced from paid search. This unusual result stuck in our minds and initiated a longer term study. We examined a number of onsite activities that were previously shown to be highly correlated to sales. As expected, search impressions have a strong correlation with paid search visits (.6) but they also have a strong correlation with site traffic (.5), and more importantly engagement score (.7) and natural search traffic (.36). After accounting for seasonal fluctuations there was a significant relationship between the level of paid search impressions and consumers visiting the site. This led us to the passive paid search conversion rate.
For every two people who clicked a paid search link, there was one passive converter who saw the link and later ended up at the site via another method (most often natural search). While only a fraction of the value of paid search clicks, this indirect value is a significant driver of revenue that should not be ignored.
So, what should you do with this new information?
1. Work with your search and analytics teams to estimate the revenue contributed by search impressions.
2. Determine the passive paid search conversion rate for your brand and key products
3. Account for the passive search conversion rate in your ROI models and include these values in your optimization efforts
4. Look outside the box the next time you run a new search strategy by expanding to larger generic terms–you may find that your ROI increases in channels you may not have expected.
Putting PR and Social Media efforts aside, I know of no other opportunity for “free” media impressions (and as everyone knows good PR and Social Media programs cost money). It will take some time and effort to effectively deliver on this search strategy, but even with an improving economy can you really afford to not take advantage of this? I wouldn’t want to have to explain to my boss why I was missing the boat on “free” media impressions.
Steve Kerho is the SVP, Analytics, Marketing Optimization at Organic.