"The time has come when advertising has in some hands reached the status of a science... Advertising, once a gamble, has thus become, under able direction, one of the safest business ventures. Certainly no other enterprise with comparable possibilities need involve so little risk."
        - Claude C. Hopkins, Scientific Advertising

Thus was the opening of advertising pioneer Claude C. Hopkins' influential book, Scientific Advertising. That was in 1923. If Hopkins were alive today, with his focus on method, testing and analysis, he would undoubtedly have been a search marketer. In the modern Internet-led age, search is the most trackable and accountable of all advertising media. Search uses sophisticated technology to offer advertisers total visibility as to which keywords are converting into traffic and sales, and which aren't working.

It is this level of accuracy, enabled by partnering with a stable and trusted search agency, that will help search marketers see their brands ride high though the recession.

Paid search is efficient, built on a cost-controlled model, making it the most recession-proof medium in the advertising world. Because search is pay-as-you-go, marketers can spend as much or as little, as slowly or as quickly as they see fit. And they can make these changes on the fly - something that can't be done with traditional media.

Quality score based algorithms used by the search engines add complexity to the variables of search. As quality is improved, minimum bids go down and rank goes up, giving a search campaign the ability to become progressively more efficient.

Even so, volatile markets, which occur during recessions, are a difficult environment for any advertiser. If advertising spend is cut, then sales volume is lost. If spend is increased, then one is exposed to significant risk on the margin. Achieving success in a volatile market and ensuring the achievement of daily business goals can be accomplished with the right balance of the complex interplay of science and technology.

A successful campaign requires specific skills and sophisticated technology to generate solid results. Search executed badly can be wasteful guess work. It is vital not to waste budget on keywords that are too expensive. To get ahead of competitors, mining the long tail effectively is crucial, and to do that, it is essential to correctly use advanced predictive modeling and automated optimization technology.

~ By Jonathan Beeston, Client Services Director, Europe

Cross channel marketing was a common theme among many of the panels I attended. Investing in the technology to track, analyze and then optimize your marketing helped inform not just the marketing teams but the rest of the organization as well.  Understanding demand for products and ROI trends helped to understand inventory trends as well as how to best allocate a company’s marketing budget.

In the Web's Growing Importance panel the speakers discussed catalog resources and how budgets are quickly being shifted to the more efficient online efforts. The trend for retailers has been to weight more importance in the online knowledge of the business. Even in evaluating who should lead the marketing efforts, one speaker noted the trends he saw were leaning more towards a director of online marketing with 8 years of digital experience versus a VP of catalog marketing with 20 years of experience. I spoke on the panel, Tracking Keywords to determine ROI with Steven Broussard, the Marketing Director for Golfballs.com.

I shared with the audience some tips for setting up a search account and how to best track for optimal optimization.  Specific to the retail category, if a company sells products with different sales cycles and varying order sizes, it is best to have an account structure that allows the marketer to separate their data for better tracking. I also shared tips on how to use negatives to funnel traffic to the right campaign or ad group. Tracking everything discreetly is the first step in calculating your ROI. Tracking at the most granular level possible – the keyword level, match types, geo regions, etc - allows the advertiser to utilize the data to effectively optimize their SEM campaigns.

As retailers continue to weather the economic storm, it's imperative that marketers optimize their campaigns and track more effectively to achieve a greater ROI. The sooner this happens, Internet retailers will be poised for success during the recovery.

Bill Lan
Vice President, Global Account Director

Typical rules-based methods are based on older technology and require a large amount of human management which both undermines productivity gains and introduces the potential for errors. Fundamentally, this approach will not calculate optimal bids on multiple positions across over 10,000 keywords.

A campaign consisting of two keywords that can be bid to 10 different positions would require 100 trials to find the optimal solution for those two keywords. Increase the portfolio to 10 keywords, and that number increases to 10 billion combinations. Even this limited campaign cannot be calculated effectively by individuals using Excel spreadsheets and applying rules to keywords. Once the three search engines are added, with the vagaries of keyword match types, content versus search and different syndication, the number of decisions required to figure out optimal positions is far beyond the scale of any rules-based technology.

To read more about this topic please download our whitepaper.

By Nick Morley
VP Sales, Europe

Having seen Comscore's release that Bing's searcher penetration rose again in the 2nd week after launch, we were curious if our data would show Bing's click share gains expanding, holding, or falling back to previous levels.

Efficient Frontier's latest analysis is good news for Microsoft. According to our data analysis, Bing expanded its share of paid clicks for the two weeks post launch. Bing's share of paid clicks is up 13% for the second week post launch as compared to pre-launch. And, it represents an incremental 5% lift over the first week. 

As we mentioned in our last post, if Bing's click share gains hold we expect advertisers to allocate additional budget to Microsoft. However, as Danny Sullivan rightly cautions in a recent blog post on Bing, two weeks does not make a trend.  

The coming weeks and months will tell a more complete story on the success of Bing. We will continue to monitor our data closely and keep you updated on our findings.

To make the data as fair as possible, we looked only at high volume, exact match keywords for clients that had continuously advertised since Feb 08. Surprisingly, this eliminates many advertisers, as some companies tactically switch brand terms on and off.  Also, we decided to use a median average (middle point) instead of a mean (adding the total then dividing by their count) so that high impression keywords can't distort the overall results.

The following graphs show the results. The blue line is the median of the sample and the grey area shows the maximum and minimum of the range in the sample. Click-through rate has dropped on these keywords from ~49% to ~34%, not quite as large as the IPA analysis, but significant nonetheless.

The CPC picture, however, doesn't reflect what the IPA reported. Although there has been a big percentage increase of 55% from £0.029 to £0.045, we are nowhere near the absolute levels cited by the IPA. There's two possible technical reasons why this is. First is that the IPA data sample may not be restricted to just exact match, which leaves it open to much more volatility in the keyword marketplace. The second is that the IPA IS using an arithmetic mean for the average, which could be skewed if there are a few relatively high impression outliers in the sample. There's also a spike in May 08 when Google relaxed its trademark policy in the UK.

The IPA's figures are a worry for brands. If brand prices indeed increased from £0.06 to £0.29 there could be a dramatic impact on client's search efforts - a forced redistribution of budgets to brand terms. Such a lift would also suggest both increased competition and a large shift in quality score driven discounts for brand owners. 

Consistent with the IPA figures, Efficient Frontier data indicates that increased competition, in part a result of Google loosening TM policy, has negatively impacted CTR. Contrary to the IPA figures, however, our data indicates brand pricing has not risen to a level that would suggest significant changes to strategy and budget allocation. We're watching the trends and working with our clients to adjust if and when necessary.

Jonathan Beeston
Client Service Director, Europe

Having spent time at Yahoo! focused on the Real Estate category, I spent hours at conferences interacting with individual agents and local brokers struggling to make search work. My conversations with small and medium sized business owners in the real estate category support Borrell's assertion that many feel "hoodwinked and fleeced." Often frustrations from local businesses stemmed from one key statement: "I get the bills but I can't even find my ad." I see the lack of predictable visibility in search as a key factor driving the high churn (50%+) Borrell identifies in SMB search marketing.

Why does predictable visibility matter in search marketing, a marketing channel where we can report on the clicks and transactions that occur directly associating ROI? First, consider the statistic that Borrell cites that just two-fifths of all small businesses don't have websites. Next, consider how few local businesses either enable online purchasing, capture leads via forms or email, or have unique call tracking for acquisition channels like search.  So small businesses are left to look and see if their ad is running as tangible evidence that their investment has value. With its complex algorithms that determine rank, search often does not produce proof of an ad to the marketer searching for it. Local marketers used to seeing their ad in the yellow pages, newspapers, and local television cannot access a similar experience in search.

The challenge of quantifying the full value of search is not unique to local marketers. Large retailers struggle to assess the impact search has on their brick and mortar locations. CPG companies would like to quantify the impact a unique, creative listing may have on consumer's perception of a brand. But valuing search is a particular challenge for the local marketer lacking basic visual proof that their ad is working let alone even showing up in the results.

Perhaps the lesson here is that we don't always need to think of value in complex equations. It is true that data-driven optimization is the right approach for many clients. However, as an industry we should not forget that a compelling ad reaching the right consumer at the right time is by itself valuable. To use a tried and true idiom, we often can't see the forest for the trees.

by Justin Merickel

Importantly, Efficient Frontier data show that Microsoft's paid click share also rose with the launch of Bing. Data show a paid click share lift of 8.1% for the week of the Bing launch as compared to the week prior. If this share lift holds we can expect advertisers to allocate additional budget to Microsoft over the coming months. Paid click growth lagging search query growth could be due to organic listings in Bing delivering stronger relevancy. Bing's new "decision engine" features, such as query-specific drill-down categories available on the top left of the results page, may be allowing consumers to more easily find a relevant organic result.

One week does not make a trend. New Bing users may just be exploring the engine after the hype of launch. But it is undeniable that the needle moved in a positive direction for Microsoft. We'll watch closely as the coming weeks data tell us whether Bing gains additional traction, sustains the lift of launch, or falls back to prior levels.

by Justin Merickel

I first met Rajeev 18 years ago when I was in the Stanford PhD program. Rajeev was always very approachable and had such a wide variety of research interests that, even though he was not officially my thesis advisor, I ended up working with him on multiple research publications. Not only was he very brilliant, insightful and knowledgeable himself in the field of theoretical computer science, he had a wide network of researchers that he worked with and knew exactly how to make the right connections knowing who might benefit from whose knowledge or expertise. It is quite rare to come across people who are both very good at what they do but also very good at making the right connections, I believe this is what made Rajeev both a great academician as well as a successful investor.

Later after my PhD, when I started eBoodle in 1999, these very same qualities drew me back to him. Rajeev began helping us with no expectations of reciprocity, well before we put him on our advisory board. Rajeev assisted us in a variety of areas from helping me with raising financing, to recruiting people as well as connecting us with business development opportunities (including one with the then fledgling startup Google).

Over the last 10 years Rajeev's profile in the Silicon Valley community has grown a lot with the success of Google whose founders he mentored, as well as other startups that he advised or invested in through his VC firm DotEdu Ventures. Inspite of all his success, he was always very down-to-earth, always available to help and make the right introductions.

Rajeev achieved a lot in his unfortunately short life and I think Silicon valley and the world of technology has lost a great researcher and mentor. I will miss him as a Professor, a mentor and a person I could reliably lean on for help. Even though he is gone, his memory will live on through the impact that he has had on this world through the research papers that he published, the companies that he mentored and the students and entrepreneurs like me whom he taught and mentored. My condolences, and those of everybody else at Efficient Frontier, go out to his wife Asha and their daughters and I wish them the strength to cope with this tragedy and move forward with their lives.

~ by Dr. Anil Kamath

Until recently, targeting data has been controlled by large publishers and ad networks to drive higher CPMs and margins. As Clifford rightly notes, "this combination — real-time [targeting] data and ad exchanges — has monetized what was once considered throwaway space online." The key point is that access to both real-time targeting data and ad exchanges have shifted price control to the marketer and away from the publishers and networks. The result is that while demand builds in the exchange and data auctions, early-to-move marketers stand to benefit from the low prices and good volume available in-part because of the lack of competition. The early mover opportunity in the exchanges is reminiscent of the early days of search marketing with Overture and Google.

One point omitted from the NY Times story was the link between search marketing and the developments in exchanges. Search marketers and their agency and technology partners have been using data to drive bid and copy decisions for many years. In fact, the data-driven, Wall Street-like approach to marketing now being applied to display is core to any successful search engine marketing effort. Efficient Frontier's founder and CTO, Dr. Anil Kamath, brought the quantitative modeling techniques he built for programmatic trading in the hedge fund industry to search marketing in 2002. To those companies like ours immersed in search engine marketing, ad and data exchanges are an exciting development that fit well with our competencies and technology. This "new era" of display advertising is very search-like at its core.

We're actively working on our display and exchange management capabilities. Early testing has shown clearly the opportunity that the exchanges present in delivering efficient results at significant scale to our performance-oriented clients. The opportunity is still evolving as the exchanges develop and improve their APIs, a necessary step for us to automate optimization. We’re thrilled to be a primary participant as the new era of display unfolds, presenting us with a new challenge with a familiar feel.

by Justin Merickel
Vice President of Marketing and New Product Development

The answer almost always is (b). Barring dramatic changes in the marketplace, it is always best to try and spend to an average daily target for the entire month. The reason has more to do with the basic mathematics of the law of diminishing returns than with any sophisticated marketing strategy.

Consider the following example. You have $3000 to spend over three days. Being a smart advertiser, you have predictive modeling based automotive bidding technology at hand. As a result, you have built an efficient frontier as shown below.



A simple strategy of spending $1000 daily would return $3,975 in revenue everyday for an ROI of 398%. However, if we were to spend $500 on the first day, then $1000 and $1500 on the third day (to make up the $500 spending shortfall on the first day) you would only get a 389% ROI. The numbers are shown in Figure 2.





Why? Due to the law of diminishing returns, the incremental ROI always decreases the more you spend. In figure (1), for a $500 increase in spend from $500 to $1000, one gets only $1,352 in revenue, but for a further $500 increase in spend from $1,000 to $1,500, one gets only $1,095 in revenue. Hence, if we spend less than the $1,000 daily target we cannot completely make up for the lost revenue by increasing spend to $1500 for a day. The incremental revenue always diminishes with incremental spend. Therefore, the simple strategy of maintaining a stable daily spend wins.

Things to keep in mind: The analysis above assumes that the efficient frontier is stable. While this is generally true, the efficient frontier usually shifts on the weekends, hence budgets should be adjusted for the weekends. Budgets should also be adjusted for short term seasonality such as promotions and sales as the efficient frontier shifts dramatically in these times.

~ by Dr. Siddharth Shah