In previous posts, I looked at the emergence of universal search as well as the types of universal search results. Now let’s take a look at how consumers interact with the universal search results page and how they click.

The data tell an interesting story, showing that clicks generally decline on pages with universal search results. Not a surprising result. Most often the intention of a universal search result page is to deliver the desired information directly without requiring an additional click. For instance, if you are looking for a map and directions and they are automatically presented, there is no need to click. If the real-time stock quote is listed for you, there is no need to click. The “Click Performance Index” in this chart indicates the likelihood of a click-through occurring on a page as compared to the average click-through rate on Google. For example, an index of 101 for “No Universal” below indicates that a searcher is 1 percent more likely than average to click on any search result if there are no universal results present on the search results page.

In looking at the chart above, we see that click-through rates are generally lower on pages with universal search results, especially Maps/Stocks/Weather. From my perspective, the most interesting aspect of this data is the relationship between paid click rates and overall click rates in the video and image areas. Image and video results will often have commercial value (i.e. an image of the new iPhone), but the video or image presence on the result page will sometimes discourage clicks away from Google. This is great quandary for the search industry. The best consumer experience often presents the answer right on the result page – no click required! As I mentioned in my last post, get your images, videos, and local store information integrated into maps for easy discovery by the engines. While it might not yield clicks, it will get you “exposure” with the same effect.

Last month I looked at the overall penetration of universal search - in this post I’ll break down the various types of universal results.

In January, I examined search results during a single week on Google. During that time, 17% of all results were universal search results, and 58% of those who searched Google saw a universal result at least one time. By a large margin, most saw video and news, followed by images and maps/stocks/weather. Maps constitute the majority of that final bucket with stocks and weather in the low single digits.

The “multiple” column indicates that 15% of the audience was exposed to two or more different types of universal results on one results page. This figure may not rapidly increase if search engines move slowly so as to maintain an optimal consumer experience. However, if it does increase one can imagine a truly multi-media experience on the search engine results page.

Looking at the share of universal search results by type, video and news clearly dominate. Is this driven by consumers? By marketers? It's actually largely driven by a combination of consumers and Google itself. Video results on Google are mostly links to YouTube, where Google has a tremendous inventory and knowledge of the content and a huge base of consumer usage. However, as marketers create more relevant material, that new content will likely start to surface on the results page and change this dynamic.

Next: Click Performance for Universal Search Results

Earlier this year on this blog, I promised to share data about Universal Search. Below is the industry’s first look at the data, shown for the first time publicly at Search Engine Strategies NYC last week on the “Orion Panel” – an exciting session with John Battelle from Federated Media, Lyndsey Menzies of Big Mouth Media, and Jack Menzel from Google.

comScore’s findings are based on Google search queries observed through comScore qSearch during one week in January 2008. During this period we observed over 220 million Google searches containing a universal result, out of 1.2 billion searches overall:

1. 17% of the queries had a universal result
2. 16% of total Google clicks were sourced from a page where a universal result was present
3. 14% of paid clicks were sourced from result pages where a universal result was present

Assuming the pattern of 17% holds, we’re talking about approximately one billion monthly universal search queries on Google alone. This is not an insignificant number and shows that universal / mash-up / blended search results are clearly now in play. It will be interesting to track how this grows (or not) over time.

For the search marketing industry, the shift from 17% total universal search results to 14% paid clicks is a vital stat. In other research situations, a move of 3 percentile points is hardly worth mentioning. In the world of search – where decimal points of change move tens of millions in commerce – it’s a big deal. It means that the presence of a universal results yield fewer clicks and will create more competition among search marketers. However, these changes should improve the referral quality among those consumer that do click.

In my opinion, major search engines like Google are adapting their result page to best meet the needs of the consumer – even if it results in challenges for or criticism from marketers. If the engines lose the consumer, they will lose their business. It will be incumbent upon the marketers to quickly adapt and create videos and images that are relevant to the consumer and make them available to the search engines. Not an easy task given the legal and technical hurdles many major brands face, but those who do it quickly will have a competitive advantage in the search arena.

NEXT: Penetration by type of universal result

I was recently interviewed by Eric Enge at Stone Temple Consulting about the search industry. We cover a lot of issues focused around the evolution of search as a marketing vehicle. You can read the full interview at http://www.stonetemple.com/articles/interview-james-lamberti.shtml

The topic du jour for Search in 2007 was clearly universal search: the inclusion of different types of search results such as news stories, maps, local information, images and video alongside the web search results you are used to seeing.

We recently discussed the implications of universal search – AKA “mash-up” or “blended” search – on a panel at SES Chicago. What we found is that search marketers are excited about the prospect of more creative options and better use of the entire page, but also have concerns about the changes that will take place in marketing practices and measurement.

Those of us in the search marketing industry are eagerly anticipating the arrival of creative tools like images and video as additions to the staid list of text links we have today. And as search ads get more visually arresting and persuasive, we’re hopeful that marketing efforts will also be more engaging and hence more successful. Even agency creatives – sadly missing in the first 10 years of search marketing – will want to play in this new interactive environment.

However, this new scenario will not come without growing pains. Today we have a huge repository of knowledge about the consumer interaction with our “old school” text-only page. Details like how and where they click, where they look on the page, and how we measure the impact will change in this new search experience. We have not seen changes like these in more than 10 years!

For example, common metrics like success rate, click rates and even our beloved PPC model will be unhinged when demonstration videos play on the result page, product images appear and a map to the closest “Best Buy” shows up. A more involved experience on the search engine result page (SERP) aims to keep consumers at the engine, not send them off as quickly as possible. Goodbye clicks, hello SERP engagement! Isn’t that the opposite of what we’ve been trying to accomplish during the last 10 years?

My bet is that the major engines will likely manage this transition slowly to protect the user experience. Ask is already embracing this movement and has seen a modest uptick in share. We’ll see if it continues and if it can be linked to the new Ask experience.

Make no mistake: universal search will eventually arrive in full force, even if it takes until 2010 or 2012. However, we’ve all seen evidence that the good old text link is a very effective way to deliver a message – text links are not going away anytime soon!

To date, there is limited data on universal search. comScore plans to apply a concerted effort to understand its impact. We will be releasing our findings throughout 2008. We’ll keep you posted.

As we head toward Searchnomics 2007 on June 27, I’d like to share with you some more thoughts on the future of search marketing.

Branding is a big issue for the search industry and could potentially change the economics of search. Brand-building ad dollars are a huge part of media spending and an attractive market for us as an industry. However, the direct response ecosystem we’ve built makes it hard to adapt to the measurement needs of brand marketers. So what can we do to demonstrate search’s effectiveness to brand marketers and capture these dollars? (Let’s leave aside for a moment the debate over the ability to brand via search – I will come back to that in a later post.)

The most classic and rigorous approach to measuring ROI is to test the effectiveness of a search campaign in driving classic branding measures like awareness, purchase intent, and likelihood to recommend. Using a classic test/control approach, this is the best way to convince a major national advertiser of the value of search in branding. Be warned – not all campaigns will succeed on these measures! That said, in the past year, 75% of search brand tests conducted by comScore showed a significant lift in at least one of the branding metrics. Message association was most often impacted (75%), but awareness and purchase intent were improved 50% of the time. There are a few common threads running through the successful campaigns: integration with other forms of media both online and offline, the “creativity” behind the keyword selection, including integration with the brand communication or creative elements in the offline media, and the size and reach of the campaign.

While major marketers will expect this classic type of testing design, there are alternatives for measuring smaller campaigns. Traditionally, we in the industry shy away from the use of reach and frequency metrics to evaluate search campaigns. For example, I frequently hear the phrase: “search is not a reach vehicle.”

My answer to that is “who are you trying to reach?” If you want to reach the total U.S. Internet population or some major demographic within it, then yes – search will not achieve reach in the same way as TV, Print, Radio or Digital Image campaigns. In fact you will be lucky to reach 1% or 2% of those broad targets using search. But if you are trying to reach, for example, the 15 million consumers who buy a PC in a quarter., search offers one of the most effective, if not the most effective, means of reaching your target audience.

Here’s a rough example for all you numbers people:

~ 15 Million PCs Bought in a Quarter
~ 7 Million People Searched on “Computers” or “Laptops”
= 47% Reach of Search Among Likely “In-Market” Computer Buyers

Can you think of a TV campaign that has that type of reach within a target segment of prospective buyers? Imagine if the industry began reporting search campaigns in this manner? You would get the attention of the CMO of a branding team in a hurry. You’re speaking her/his language and presenting search marketing efforts using data comparable to the offline world. A discussion using reach, frequency, gross rating points (GRP), and impact on awareness / purchase intent is far more likely to engage the brand marketers than keyword lists, natural and paid-click rates, or direct response ROI calculations. We have the data to speak their language – let’s use it!

Hi, my name is James Lamberti, SVP Search and Media at comScore. I’m heavily involved in our search work and often partner with clients and the industry to dive deeper into the role and value of search. On June 27th, I am delivering a keynote at Searchnomics 2007, titled “Beyond Direct Response: Search's Role in the Media Mix for Branding and Offline Sales.” This keynote is a great opportunity to show the industry, using a variety of real data points amassed over the past year, that search is much more than the direct-response vehicle that launched the industry.

One key data point at the center of this discussion is the offline sales impact of search. We have conducted research across dozens of retail and non-retail categories and find consistently that 65% to 90% of conversion occurs offline. These figures alone have a profound impact on the strategy and economics of search. For example, pure-plays like eBay and Amazon rely extensively on search to drive ecommerce. How do the economics change when multi-channel nationals like Wal-Mart and Target achieve 5x or 10x the ROI as pure-plays? Issues like the value of search in branding, its link to other media, and the opportunity to re-engage agency creatives are all a part of the next level of search marketing. Be sure to check out the Searchnomics 2007 event and check back here over the next few days to hear more thoughts about the keynote theme.