Everything I Know about Marketing I learned from Google

Gian Fulgoni, comScore

Gian Fulgoni
Executive Chairman & Co-Founder
comScore
2/8/10

Which of my lessons resonates with you the most and why?

“Let the data decide” is a great lesson.There is an abundance of data available in the digital world and, if it’s harnessed effectively and correctly, it can provide terrific insights. At the same time, we need to recognize that there are online issues such as cookie deletion that need to be taken into account lest they have a deleterious impact on the accuracy of a variety of metrics.

“Test everything” is very good advice.There is great learning in understanding what works and what does not before making major commitments of money. Smart marketers are doing this as they evaluate the shift of dollars from traditional media to digital. I would add that it’s important to measure both the short term and long term impact of digital marketing programs. Branding advertising has a latent impact that needs time to be observed and measured. Avoid becoming so direct response oriented that the time needed to build brand value is ignored.

“Don’t rely on SEM alone.” Don’t attribute 100% of the credit for a sale to the final click on a search ad. It’s folly to ignore the marketing [and] branding efforts that are needed higher up the sales funnel.

What makes Google such a unique company? Why has it been so successful?

Google has brilliantly leveraged the power of computers and information to both dramatically reduce their operating costs (such as the cost of selling) while simultaneously providing real value to Internet users and a valuable service for marketers, both big and small, that drives incremental sales.

How can marketers take insights gleamed from Google to inform creative and/or media strategy?

One of the key things to realize when using search marketing is that while you can get high lifts in sales, your reach is limited by the number of people searching using the terms [or] words you’re targeting. So, it’s critical to overlay display on search campaigns. With search you get high lift but limited reach. With display you get lower lifts but higher reach. comScore research shows that by combining search and display into one integrated campaign, it’s possible to often get synergy, where the combined impact is greater than the sum of the parts. It’s also clear that marketers need to maintain a branding effort while they’re running search campaigns (which are more direct response in nature). Successful branding efforts will result in greater click throughs on search ads.

Lesson #11 is “Brands can be answers.” What research has comScore conducted that shows search can drive a lift in brand marketing attributes?

Here are several links to research we’ve conducted that measure the branding impact of search:

http://www.comscore.com/Press_Events/Press_Releases/2007/10/Search_Marketing_for_Consumer_Packaged_Goods_Companies

http://www.comscore.com/Press_Events/Press_Releases/2006/03/Online_Impact_of_Offline_Buying

http://www.comscore.com/Press_Events/Press_Releases/2008/01/Importance_of_Search_Engines

http://www.comscore.com/Press_Events/Press_Releases/2006/03/Online_Impact_of_Offline_Buying

One of my favorite research studies that addresses the issue of the branding impact of search is this from Enquiro.

Lesson #13 is “Your competition is broader than you think.” How can marketers use comScore to assess their competitive landscape?

Using the comScore panel of 2 million global Internet users, we see consumer activity at all web sites. As a result, we can help marketers better understand who their real competitors are and what’s happening on competitors’ web sites. How many visitors and buyers do competitors get? What are they buying? What are their demos? How did they get to the site? Which competitors are stealing market share from me? How are they doing that? What search and display tactics are competitors successfully using to build their business at my expense? Is the market growing and what is the breakdown of market share?

These types of studies that use a panel go way beyond what can be done with web site server analytics, which can only describe what is happening on the operator’s own site. Once a visitor leaves the site, the server-side analytical packages “go dark”.

One particularly powerful analysis that can be conducted using a panel is a “visitor/non-buyer study”. Here, we examine the behavior and characteristics of visitors to a web site who didn’t buy at the site but did buy in the product category at another site. We can even survey these people to better understand attitudes and motivations. This helps web site operators understand who their real competitors are and what can be done to convert more of their visitors [or] non-buyers into buyers.

You’ve often said that too much emphasis is placed on the click. Why do you believe that is the case? What can be done about it?

With click rates on display ad campaigns averaging only 0.1%, either this means that online advertising doesn’t work or it says that the click is basically a direct response measurement metric which ignores the latent impact of advertising. ComScore research confirms that it’s the latter. Educating marketers and publishers to this reality is critical. I think most smart agencies now get it. Using behaviorally-tracked panels, one can compare the behavior of ad-exposed people with control groups of non-exposed and measure the impact of the ad campaign over time –- even with minimal or no clicks on the ads. At comScore, we’ve been aggressive in trying to educate the industry with white papers, blog posts, press releases and conference presentations. I think we’ve had a positive impact.

How can marketers accurately measure the impact search or other online ads have on offline sales?

The key is being able to link the people exposed to online ads with their offline buying. Behaviorally-tracked panels are perfect for this. For example, at comScore, we have our panelists’ permission to use their name and address to link their online activity and ad exposure (which we see through our tracking software) with their retail buying behavior (which is captured when they show their retailer loyalty card at checkout). Then, by comparing the offline buying behavior of ad–exposed people with a control group of non-exposed people we can isolate the effect of online ad campaigns in lifting offline sales. This approach works for both search and display.

Here’s a link to the results of our research into the effectiveness of display campaigns in lifting retail sales of CPG brands, showing that the impact of online ads rivals that of TV.

McGraw Hill Professional Amazon
Amazon Canada Amazon United Kingdom
SEARCH