Big Site, Big Search at SES Chicago

I sat in on a unique session today at SES Chicago called “Big Site, Big Search,” which explored the role and shape of Search Marketing in large and even global corporations. Kevin Ryan moderated, with Bill Hunt of Global Strategies International, Amanda Evans of WebMama.com, Inc., Olivier Lemaignen of Intuit, Randy Peterson of Procter & Gamble, and Carol Kruse of Coca-Cola all coming together for a stellar panel with some interesting perspectives.

Meet the panel:

Carol offered some great insights into the unique problems of scale that organizations of Coke’s size face, both internally with respect to change and externally with respect to meeting the demands of a global brand in a globalized medium.

Olivier’s primary concern was for the education of the client.

Randy came out as the strongest proponent of using large organizations’ size to take firm control of the kind of campaigns of scale only they can produce (and, therefore, botch up).

Bill really beat the drum for integration of large-scale SEM’s and SEO’s to the structures of the company—this came across as sort of the inverse conservative corollary of Seth’s 9AM vision towards a new-economy Giza, that if you as the SEO aren’t going to change a Fortune 500 overnight through its website architecture then first and foremost make sure you aren’t slathering on the hot fudge.

Amanda pushed the idea of identifying specific site architecture problems and fixing those hurt areas. She positioned her methodology as very sympathetic to the needs and goals of a client with respect to SEM: a site shouldn’t be redesigned, she argued, for the sake of THE INTERNET, but for the sake of the client’s goals on the internet.

Do you know who I am?
Though it almost seems self-evident, the first question we should ask, the one that I was particularly glad to hear the rep’s of these behemoths address, is why on earth companies with such brand recognition should need to reinvent themselves for the digital space. Coca-Cola is, anywhere and everywhere, Coca-Cola, right?
The answer seems to be, “Yes, but not automatically.”
Some of the stunning results of Coke’s successes in their recent pilot online marketing and optimization projects:
Coke’s optimization efforts (which the panel didn’t have time to go into detail on) correlated with a 139% improvement in registrations, 109% improvement in conversion rate, and CPA was controlled to an only 6% increase. Even for a household organization like Coke, it’s important to push results beyond branded terms, a change which Coke has seen as a result of its pilot programs.
This takes on an interesting twist, however, when we examine unnecessary attempts to use the internet to make a brand even more mass-market when there is an established base of consumers online right now looking for your product. Bill advocates a strategic and judicious approach to SEM, saying big brands shouldn’t be shy about optimizing with emphasis on their brands if they work.

So that’s the why, but how do you do it? As Carol pointed out, “search at a consumer package goods company is a tough sell,” especially because sell-through is hard to measure with lifestyle branding; this means that search optimization requires a lot of energy in advocacy of it in order for meaningful change to occur.
How do mega-sites internally structure online marketing shifts when, as Bob Tripathi pointed out in the morning’s Usability & SEO session, there are numerous concerned parties and everybody’s got an agenda? Or how does a mega-brand like Coke externally maintain its online image when it’s competing against eBay for rankings on branded terms? 4 answers to these questions:

How to convert friends to your search cause.
Carol said that fundamentals like missed keywords that apply to your site, showing your weak points to management, can open the door to the discussion, and case studies of big successes by comparable companies can then be very compelling in making the case for the necessity of implementing these changes. Amanda and Olivier also felt competitive analysis was the most powerful way of making your case to make change.
Randy promotes the “path of lease resistance”: find another person in your organization who can be an advocate for your cause, and fit your approach to the company’s language.
Bill kept it simple and clean: “take five words.” Don’t get any expensive tools, just do the search, get volume estimates, and make your case in the context of missed opportunities. Once the door is open, you can step your campaign up and make a more detailed case and lay a program out for
Kevin pressed the issue with a focus on the IT department. How do you get these gatekeepers to your site’s eventual form to work actively with you on your SEM program? Both Randy and Olivier agreed that giving kudos along the way to the department showing increases in revenue resulting from their efforts can go a long way to winning them over to your program (and Olivier’s firm takes this as far as setting up financial rewards tied to improvements of online performance). Carol shared the effectiveness in her projects of setting common metric benchmarks for everyone to shoot for.

How big is big? Going world-wide on the web.
Bill points out that optimization, design, and user experience are very different from country to country. Deep internal structures need to address the language in these foreign countries; if you’re pushing various sites with similar content you want to be careful you’re not running into a duplicate content issue. Right now, Bill says there are serious problems with differentiating between countries on the directory level (Google is picking up the domain and now the subdomain differentiations), but the “/country/”, formatting is extremely problematic right now. This echoed today’s earlier Usability & SEO session, in which the panel addressed problems of engaging users across multiple languages by doing more on an international site than simply translating the body copy and publishing.

Synchronization: How to get everyone on the same page.
Bill, from a strategic point of view, approached the problem of syncing multiple teams in a large department up with one another by understanding and working with the existing structures: he pointed to sitewide style protocols as a great resource for finding initial common ground between different groups.
On the process & technical side of things, Randy emphasized how crucial it is to determine what stats everyone is going to be using—get your SEM and SEO groups using the same data set and determine how they’ll access this data. This streamlines strategy discussions. Randy also dealt in a roundabout way with the problem of turf wars in response to Coke’s practice—although Carol underscored the importance of challenging SEM’s and SEO’s to make sure they’re giving you what you specifically need—of outsourcing all their metrics analysis, and only having a few Search Marketing people on the staff. Randy’s work at Procter & Gamble gave him a much different and more metrics-based perspective, and he insisted on the importance of taking as much of the analysis and implementation in-house in order to ensure that they have better control over the data and, as a result, over the ultimate development of programs on their site.
Later on, following a question from the audience as to how to determine which battles a Search-Enthusiastic Marketer should pick, where do you draw the line between becoming “that *** search guy” and becoming a mover and shaker with respect to your company’s online presence, Ryan returned to the idea of establishing best practices that everyone can refer to. This makes it less a personal confrontation than a goal everyone needs to address. Amanda advocated an incremental approach that would allow people to accept the budgetary needs and Carol suggested moving up to a high level on the site where the money will be and the management will be able to push for change more authoritatively.

Mistakes to not make: Kevin and Bill see stupid people.
Bill called on marketers to challenge their agencies. You can’t blindly accept the agencies’ thinking—your metrics might not mean what they say they mean. “Stupid money”: don’t just throw money at your problem to make it go away. You might end up with an expensive site that’s not crawlable [read “flash”], so do your research, and learn what it is you’re trying to build and what the rules are. Among the biggest mistakes Bill sees right now is the overextension of brands when they have perfectly good target keywords and audiences right at home (I mentioned this above).
Carol played a little nicer, emphasizing the need of education, and that with the right approach, people can be won over, and she wasn’t the only one to argue the importance of educating co-workers and clients in the importance of the media. Olivier pushed for this as well, and the incremental steps that Amanda advocated as well as the cost-effective and proving measures of targeting a few low-hanging fruit as suggested by Randy and Bill.

The lessons I took away from the panel were:
1. Find your Allies
2. Make your case and make your converts
3. Educate, educate, educate
4. Build and market your organization for what it is: if I’m not putting words into Bill’s mouth, I think the conservative path of proving the worth of the medium to these large organizations in their own field is exactly the way to go. However much change may be demanded, the change can’t happen overnight; but it needs to be started.

All in all, it was insightful advice from a panel of very singular experts in the field, and though the discussion was very much centered on large-scale optimization projects, the lessons of advocacy, measurement & accountability, and hands-on adoption of the medium are hardly exclusive to the big companies.