The researchers describe domain bias as “a user’s propensity to believe that a page if more relevant just because it comes from a particular domain.” Using data from a blind test of click activity, the study asserts that domains can influence a user’s preference 25% of the time. Even more surprising is that users would choose recognizable domains over results that were actually more relevant. The crux of search engine services is returning results relevant to a user’s query, so this behavior feels somewhat irrational. One potential rationale might be the trend of users ticking to the “safe neighborhoods” of the Internet to avoid viruses, poor content farm content or scams from Nigerian princes; people tend to trust the sites they already know and feel comfortable with.

An example given in the report compares CTR for similar ads when the domain is switched. In this instance, users preferred WebMD.com over GeneticHealth.com 25% of the time. Of note is that the experiment accounted for users randomly selecting answers by removing results from failed “honeypot tests” where click behavior was not consistent or proved erratic (about 10% of the time).

As digital marketers, we try to capture a user’s attention by a variety of tactics, though this report implies that a large portion of success on search engines is influenced by a domain name that could have been an afterthought in the online marketing plan. A big takeaway here is the importance of addressing the domain name early in the process, accounting for perceived credibility and trust with your property’s URL. Furthermore, pay-per-click advertising relies on Display URL’s to draw eyeballs and entice clicks. Since most PPC platforms require the Display URL to match the Destination URL, advertisers are stuck with their chosen domains. Ads for credible sites could benefit from this bias while new sites or 3^rd party advertisers could suffer on CTR.

The portal through which users interact is the search engine, so the question becomes whether algorithms will take into account this preference for credible domains when ranking results. The Microsoft research reveals that “search results concentrate over time on fewer domains with increasingly larger share of results pointing to the top domains.” We have seen this progression over time with 2009’s Caffeine update favoring brands on search results, and a recent algorithm update lending authority to “official sites”. Google may continue to give searchers what they prefer – credible and recognizable domains – though the result of which may lead to a future of homogenized search results.

But there is hope! New sites attempting to find a footing in search results can still take action:

• SEO best practices regarding technical elements and site architecture
• Relevantly applied microdata on content
• Elements of freshness and recent content
• Enticing PPC ads served to targeted searchers

Microsoft researchers have demonstrated the influence of domain bias in user click activity, though the fact that 25% of searchers were swayed by the URL means that 75% of searchers were not. There is still an opportunity for sites to create relevant content, apply good SEO tactics and rank for a variety of terms.

SURFACE LEVEL ANALYSIS

Reviewing the below data set based on just core data (clicks, impressions, CTR), a few outliers grab our attention:

• Campaign 1 is most effective with high clicks and CTR
• Campaigns 4 and 9 are performing great with high CTR’s
• Campaign 8 is struggling under a lower CTR

COMPARING MULTIPLE DATA POINTS

Extending our view to include additional data, the above conclusions are adjusted to incorporate new metrics and a more accurate picture is revealed.

• Campaign 1 is afforded its high click count due to a low $0.98 CPC. Further investing in this campaign with the goal of increasing clicks might prove less costly than other campaigns.
• Campaign 9 might be applauded for its high 10.06% CTR, but this must be viewed in light of relatively low impressions compared to other campaigns. As well, despite the successful clicks, users are less apt to convert with a low 0.95% conversion rate.
• While Campaign 8 has the lowest CTR and highest CPC, the conversion rate is the highest. If conversions are the ultimate goal, we might explore why the CTR is so low and try to bring down the high CPC with more relevant keywords/ads to further fuel conversions.

VISUAL DATA ANALYSIS

Reviewing multiple data points visually can also yield trending insights not apparent from raw numbers. In the example below, our graph shows that as CTR increased over time, the average CPC fell somewhat proportionally.

We might infer that our ads have been viewed as more relevant by Google as the CTR increased, thereby improving keyword Quality Score and driving down our CPC. Knowing that CTR increased is helpful information but tying this metric to cost-per-click data reveals a bigger picture of how the campaign is performing (and more importantly, the factors involved).Further analysis into other influencing factors might confirm this correlation or suggest alternative theories.

Most data points in PPC performance affect each other in a constant ebb and flow of interrelated factors. Too often a myopic view of only one metric is taken (often to reinforce a hypothesis), but widening your vantage point to encompass related factors can reveal much more about real campaign performance.

We have also seen examples where a character icon was inserted into the sitelinks. This visual manipulation is most likely not in line with AdWords regulations, though such actions are not being blocked at the moment.

Here is a quick rundown of the rules & parameters for ads with sitelinks:

• Up to 10 hyperlinks may be specified, along with display URL’s, though all links may not appear, depending on the quality of your ad compared with competing ads
• Sitelinks will only appear for the single top-ranked ad on a given user search. While the official policy is 1 sitelink per query, we have noticed instances of multiple paid sitelinks in the top position.
• There is no additional cost for ad sitelinks.
• There is no guarantee on the frequency with which your sitelinks may appear.
• Ad sitelinks can only be set at the campaign level, not the ad-group level

Organic vs Paid Sitelinks

There is an important distinction to be made between paid AdWords ads with sitelinks and organic sitelinks available outside of the auction system. Organic sitelinks become available through Google Webmaster Tools when your site meets a certain threshold of content and/or pageviews, though the parameters are not exactly known. Google automatically offers sitelink options and the webmaster only has control to deny the auto-generated sitelinks or suggest new titles for those already chosen. Choosing sitelinks outside of the auto-generated options is not possible.

Paid ads with sitelinks offer more control in the text displayed and hyperlinked, though the advertiser pays for this exposure and does not have complete command of when and how they might appear.

The Positive: Higher CTR & More Real Estate

We have noticed a substantial CTR increase on sitelink ads, even across different industries, demographics and ad versions, with a typical rate above 10% and sometimes closer to 20%. Another advantage is the valuable real estate the sitelink takes up above organic results. Any additional space that your ad can fill, and thus take away from competitors, only serves to better draw the searcher’s eye toward your ad and your tempting array of links.

Similar to organic sitelinks, this paid ad extension operates best when a campaign lends itself to multiple landing pages. If your paid search conversion funnel can take multiple approaches, sitelinks provide a unique opportunity to direct searchers down different paths toward potential conversion.

The Negative: Ad Constraints & Lack of Control

A major drawback is that sitelinks can only be set at the campaign level. Most account structures have a broad campaign theme broken down into specific topical ad groups. It can be difficult to generalize a sitelink that encompass all potential ads that could appear in that campaign, hindering the micro-targeting that otherwise an account structure offers. This lack of control is further augmented when you don’t get to choose which ads will appear with sitelinks. This is left to AdWords deeming your ad (relative to competitors and ad rank) as ‘high quality’, an attribution almost as difficult to manipulate as the quality score assigned to keywords.

Since we all love data and analytics, what if you wanted to see which of your individual sitelinks was bringing the most clicks? AdWords does not show data on specific links, which would be a great insight for any data-driven account manager. Monitoring conversions is not available in the ‘Ad Extensions’ tab, which would make the most sense, but can be found by segmenting ads or keywords by ‘Click Type’. More of a user-interface complaint but certainly room for improvement.

Sitelinks: Room for Improvement But Still Valuable

AdWords sitelinks have been a benefit to our campaigns, helping drive up average CTR and placing our ads above competitors on the SERP. Certainly there are areas for improvement, though the simple fact that buying a sitelink placement costs no more than a typical ad shows what a benefit this extension can be to your campaign.

Furthermore, ad spend has increased 159% over the same time last year with an increased number of keywords purchased. As advertisers see more conversions pay out on paid search placements, confidence rises and companies start to carve out more budget in their marketing spend for paid search advertising. This fact is further bolstered by Media Post News reporting that consumers are “twice as likely to click paid search product listings” on a retail website that is using Google Site Search. Running relevant ads on the Google display network (previously called the “content network”) can help product listings convert and small business can capitalize on this targeted environment.

We previously discussed the Forrester forecast of marketing spend by format back in 2009, with preliminary 2010 spending reinforcing the estimates and on pace to continue showing positive gains. eMarketer also suggests that overall US search ad spend could be up 15% this year, a trend that would benefit both small businesses looking for new conversions and the search marketing industry.

Advertisers have previously had the ability to include text ads within Google Image search results but it was just recently announced that image ads would become available as well. The option to include a thumbnail image adjacent to text will be called “Image Search Ads” and is a boon to advertisers looking to capitalize on the reported 1 billion page views per day that Google Images pulls in. This new ad format is a benefit to advertisers by allowing products to be shown in a setting where users are already searching for related images. According to Ben Ling, Google Director of Search Products:

“People often search images for queries about shopping, travel and entertainment, which lend themselves well to advertisements.”

Certain keywords, like “converse shoes”, might lend themselves better than “pet unicorn”, for example, though advertisers will also make this distinction and only bid where it’s relevant and to their benefit. The sponsored image ads, as usual, are offset by a shaded color background to distinguish them from the organic results.

Opening the door for future rollouts, Ben Ling also commented that the company would explore adopting the format on other products, including the main search results. Image Search Ads present a new avenue to help the niche advertiser looking to include a visual allure to his ad and campaign tests will show whether the investment pays off with clicks and conversions.

In addition, consumers exposed to both influenced social media and paid search ads are 77% more likely to search for that brand compared to those exposed to a brand’s paid search ads alone. Not only did search volume increase when combining social media and paid search, but click-through also saw a 50% increase.

“Every day consumers express their intent via search. Now, we better understand how that intent is established via social media and the interplay between the channels,”
Chris Copeland, GroupM Search

“This finding provides strong evidence that investing in social media marketing can both increase initial brand consideration and drive higher conversion rates once the consumer has decided to purchase”
Graham Mudd, Comscore Inc

The study suggests that searchers who use social media are more engaged overall and may be more likely to search with intent to purchase when compared to those who do not use social media.

Research has shown that display ads have a similar effect on search volume and click-through, especially when combined with paid text advertising. Now, it seems, influenced social media can join the ranks of advertising channels that act as an influencer and multiplier of search behavior.

Also increasing, not unexpectedly, is the share of total marketing spend that interactive will garner over the next 5 years. This amount, however, is a substantially lower than the amount suggested by 2007 Forrester research which predicted that total interactive marketing spend would reach $61 Billion by 2012.

As search marketing continues to incorporate other avenues of interactive marketing, it will be interesting to see how they continue to delineate between pure search marketing and email, display, or mobile marketing. If email newsletters are posted online when released and optimized for search, doesn’t that count as search marketing? When Facebook pages show up in brand related search results, is that social media or search marketing?

Some interesting things to point out:

- They are always positioned above the text ads. Unless your ad has the required quality score and bid to appear at the top of organic results (left column), your text ads on the right are substantially lower on the page when compared to the current layout.

- They are always in sets of 3 (when the images are shown)

- They are often, but not always, the same top three results in a similar search on google.com/products. When they are not the same top three from each, there is no discernible pattern regarding how they are chosen/displayed. See the two searches below for [mr clean magic erasers]. The first, done on google.com, reveals product ads from Office Max, Sears, and eCost. Searching the same query on google.com/products does not even yield a result from Sears (that I could find).

- They still need some work. See the screenshot below for a search for [cascade action pacs]. Looking for a grille for your Ford Mustang complete with flames? Too bad…they are temporarily out of stock.

- Google appears to be testing various types of product ad layouts. Some with photos, some without. Below are screen shots of two versions of the same search for [febreze].

It will be interesting to see, if/when these tests become a standard part of SERPs, what happens to the click-through-rate of text ads placed below the product ads. Also, I have to believe that the click-through rate of the product ads themselves will be quite high (especially initially) when users see an image where they previously did not.

Will products need to be listed in Google’s product directory to be included? Will they give preference to merchants using Google Checkout? Similar to organic page ranking, will a certain amount of authority be given to the brand’s official ecommerce sites? Or will the usual bulk PPC buyers (e.g. BizRate, NexTag, Amazon) be the major players?

We’ll post the answers here as we learn them.

If the above statement wasn’t true over the last fifteen years, it certainly is now. Yahoo is now committed to the content and ad game, forgoing search to Microsoft. This probably shouldn’t be as big of news as it was.  After all, Yahoo used Inktomi search results to help supplement their directory listings way back in the day. And it was only a few years ago that Yahoo search was powered by Google.

So this Yahoo/Microsoft deal led me to wondering about what effects, if any, this would have on the international search universe. I have some fairly extensive international SEO experience having worked with a huge client with 70 locations across the globe and dozens of web sites catering to each individual country. Heck, we even had two Chinese language sites, one in Traditional and the other Simplified.

So while I’m on the topic of China, let’s take a quick peek at the search engine market share there, shall we?

Clearly, Yahoo and Microsoft have lost the battle in China. And despite multiple media reports to the contrary, Google is alive and kicking in the world’s most populous country.

Because of China’s major growth curve per Internet usage, Baidu is now up to 9% of all global search engine usage. Google’s global market share is actually going down due entirely to Baidu and doesn’t reflect the rest of the globe.

Which begs the question, just how is Google doing outside of North America?

What’s that? You’d like to see a chart showing this data?

Ask and you shall receive.

According to ComScore, Google has 60% of the US market, but 67% worldwide. Impressive. Microsoft has made few inroads abroad, and has 3% or less share in the U.K., Germany and Korea. Yahoo has one pocket of overseas influence: Japan, where its search share is 43%, close to Google’s.

In many top and growing Internet countries, including Germany, Canada, Brazil, Turkey and Italy, Google has more than 80% of the search market.

So no surprises there. Google is still the king of the jungle but there are still plenty of country/language specific engines that are very popular in some countries. But that’s another post.

So there’s a 30,000 foot view of the global  search landscape as it stands now. Next time, I’ll write about some of the things to bear in mind when working on international SEO.

We have organized these moments into the following categories:

• Organic Web Search
• Paid Text Ads
• Organic Image Search
• Organic Video Search
• Organic Maps Search
• Social Networking Sites
• Mobile Search
• Website Landing Pages
• Analytics and Measurement

Download the PDF to learn 50 Ways Search Marketing Can Change Consumer Behavior

The top 5 most common Quality Score myths and misconceptions:

The higher your ad’s position, the higher your Quality Score FALSE
While a change in position may increase your click-through rate, an increase in position alone will have no affect on your quality score. In fact, your Quality Score takes into account and normalizes performance differences resulting from ad position.

Good conversion rates can increase your Quality Score FALSE
Your conversion rates do not, in any way, contribute to your Quality Score.

Poor keyword performance from testing will hurt your Quality Score FALSE
One of the primary advantages of paid search campaigns is the ability to test keywords, ad copy, bids and landing pages in order to optimize performance and maximize ROI. The potentially negative short-term impact on your Quality Score from these tests will be quickly out-weighed by the improved performance that comes as a result of testing.

Exact match terms will have a higher Quality score than broad match FALSE
When using broad or phrase match, the keyword’s quality score is only impacted when someone searches for the exact term being used (regardless of match type). For example, bidding on [custom t-shirts] using broad match, your ad will appear when someone searches [design custom t-shirts]. However, this particular query will not affect your Quality Score because it is not the precise term you bid on. Only when users search [custom t-shirts] will that query’s performance contribute to your Quality Score.

A higher bid or budget will boost your Quality Score FALSE
Increasing the maximum bid amount of your keywords will not increase their Quality Score. This will most likely simply result in an improved ad position which, as mentioned above, also has no affect on your Quality Score.

Increasing your daily or monthly budgets will also have no affect on your Quality Score. It is often incorrectly assumed that if your ad is not showing due to lack of budget, your Quality Score will suffer. This is not true as Quality Score only takes performance while your ad is running into consideration.

What is Quality Score?

If you are interested in learning more about how maximum bid and Quality Score affect your ad position and the actual pay-per-click costs of your campaign, Google’s Chief Economist does an excellent job of explaining the relationship in the video below.

While Google is constantly testing variations of their SERP layouts, this marks the first confirmed change to ad format since the inclusion of the Google Checkout logo within participating merchants’ ads.

CPC vs. CPA

In perhaps the largest departure from their current model, Google will not be charging merchants when users click on their ads, but instead charge them only if a user completes a purchase on their site. Some could argue that such a model encourages merchants to attempt to place their product ads in as many search results as possible without regard to ad quality and relevance.

Under the current CPC (cost-per-click) model, advertisers are motivated to keep their ads relevant to ensure higher quality scores and, in turn, lower costs-per-click. This leads to a better overall experience because users are only shown “high quality” ads. However, under the new CPA (cost-per-action) model, since merchants are not paying for impressions or clicks, there is an economic incentive to cast a very wide net. This creates the potential for an increase in less relevant ads appearing on SERPS, leading to a more cumbersome user experience.

While this is a valid concern, there are several potential solutions. The easiest of which is to simply apply a ‘quality score’ (similar to text ads) to each merchant’s product ad based on click-through rate, conversion rate, etc. This quality score would then dictate the CPA cost paid to Google. The economic incentive for merchants to keep ads as relevant as possible would then be in place.