Finding keywords, trying different ad copy, testing landing pages, bid managing – blah blah blah. You already know what managing SEM is about. But if you crave the new SEM tactic, the unknown search story, the changing market dynamics of SEM that few understand and even fewer talk about, come to this session, where Omniture and an SEM "dream team" panel will push the conference envelope and make you — yes, even you— all-stars: better, more knowledgeable, and aware of what's really going on in search.
· Chris Zaharias, VP Search Sales, Omniture
· David Rodnitzky, VP, Strategy, PPCAdBuying.com
· Terry Whalen, SEM & Internet Marketing Expert, Founder, TDW Consulting
· Chris Knoch, Principal Consultant, Best Practices Group, Omniture
· Vinny Lingham, CEO, Synthasite
Chris Zaharias: I work for Omniture, our goal here is to tell you things that you might not otherwise have heard, stuff that does not get covered at the show. I did not know much about Omniture until I joined about 6 months ago; we are the leader in web analytics and headquartered out of Utah. We have about 500 SEM customers.
I will introduce the panelists and then speak a bit myself. Here are the topics:
- the assumption that the long tail is eternal
- the notion that there are 1001 things to do in SEM
- the assumption that SEM is always the right channel – where it is and isn't the right channel
- fine line between search engines and what they try to sell
- is search is really opaque or are there methods to make things more transparent
The first point is the assumption that the long-tail keeps growing, that finding more keywords is the critical activity (shows some data from HitWise showing percentage of total search volume for the top 2000 search queries). The reality is that search is becoming more navigational. The growth in brand and trademark queries shows that search is become less "search" and more a method of direct navigation. That's why revenues for the search engines are going more towards trademarked and navigation terms.
Any questions on this data so far?
David: Google is not encouraging people to buy the multi-token keywords, and encouraging to use broad match for the tail keywords. Search engines are reducing the efficacy of buying long-tail keywords.
Terry Whalen: People have gotten the point that you don't need to do super advanced queries to get the results.
Chirs Knoch: From my perspective I have seen quality scores declining because there are too many keywords in too many ad groups. I have done tons of campaigns with the long-tail and it's been a couple of years since I have seen a great return on that effort.
Chris Zaharias: Next up is Vinny Lingham.
Vinny (spoke really fast): When you are managing a $200,000 campaign and you spend 40 hours optimizing you can decrease costs by (1%)$2,000 = $50/hour. You can achieve greater scale as there are more gaps in bigger campaign than a smaller campaign – due to market fundamentals (all your competitors bid on the same head words, but not tail).
Smaller advertisers should focus on fewer engines, starting with largest market shares for their geographic areas. Expand once you reach saturation. Theory is only valid for long tail campaigns (5,000+ keywords) – most effective with +1m keywords.
Should run multiple search engines for head words (top 100 usually).
Running big campaigns is great but when you have bad landing pages it will default to bad quality score. We have done tests where everything was right and you get good traffic and good volume. Keep your campaign focused on relevance, good ad copy and good landing pages. What users really want is relevancy. Keep that in mind and you will be fine.
Dave: I see a lot of people who are not tracking at the keyword level – if you are not tracking you have no idea where you should be spending your time. It's very difficult to know which keywords to adjust.
So, I think that SEM is obviously very hot and big corporations get a lot of pressure about spending more on SEM etc. but what I have found is that SEM in its traditional marketing does not always work. For many B2B products today, for people that have such a new product that there are no searches for, content network ads work really well. For example, a DVR, maybe 10 years ago people would not know to search for one, but to see it on a banner ad would be a different story, it raises awareness. So with new products, start looking at some other options. Also, I am really a search marketer not a banner ad guy, so my assumption was that the bigger the ad the better the performance, but it's not the case, the one that converts best for my clients is a square box!
Chris Zaharias: one of the campaigns I admired was by Honda, they were trying to get people to buy their car, but they had a lot of new models that no one was aware of. So what they did was they brainstormed who are their target demographics – and thought about what are those people searching for and interested in – types of animals, hobbies, etc, and did a campaign that revolved brand building around other topics and really increased brand awareness. It had to do with engagement of demographics, there was no demand, but they built up demand based on non-automotive search terms.
Chris Knoch: A common assumption I work on is that content is not as effective as paid search. What will you get out of it? The results are not the same. We've looked at the quality of content traffic. So you've got these engines that make it easy for you to opt in to these other programs, and they mix your costs, but it messes up your reporting. To Chris' point, there has been analysis done all over the place. Looking at Yahoo, over half is not good quality. On Google's network, a good 90% of traffic is considered to be less than great quality traffic. Search is like "permission" marketing, it doesn't interrupt you. That's very different than content, so you see the vast difference in the quality of traffic. But this stuff has so much momentum that they suck you in.
I recommend creating separate accounts altogether for content network campaigns. Don't let the engines up-sell you on that.
One of the most un-utilized ways to improve your broad match keywords is using negative match.
Terry Whalen: So there is the idea that a lot of things in SEM are opaque, we don't know algorithms and what makes up quality score. The point of this topic is not that you can go ad seek out transparency in terms of reports, but what I mean is more seeing the forest for the trees. What I am talking about is in terms of positioning and SEM, if you are a smaller company or a newer company, there is a lot of competitive intelligence you can gain from looking at your competition. You can use screen scrapers (Spyfoo), we can look at landing pages of competitors etc. This is a simple thing that a lot of companies don't seem to be doing this, but when I bring it up they are receptive to it but don't necessarily do it. If there are certain competitors that you know are large and buying the top keywords, most like they are doing a good job so there are probably some things you can learn about what that are doing. I think a lot of folks don't think about that.
Chris: Through any types of Analytics you should look at all the steps. The micro-conversions, what steps, say, people are taking to complete a purchase.
Session coverage contributed by Sheara Wilensky of Promediacorp.