Moderated by Rebecca Lieb – ClickZ
Christine Churchill - KeyRelevance “How many people thin k that keyword research is the most exciting/sexy part of search marketing? (laughs) KW selection criteria: relevant to site. Keyword popularity. Stage in buying process (user intent). Competition. Feedback.
Will go more into detail about stages in the buying process, as well as some tool demos. Buying process…where are they? Problem recognition---information search---select alternatives—evaluation of alternatives---purchase decisions. (source: Phil Coulter’s book) Three types of behavior: 1. navigational search. (ie type in AA.com.) 2. Informational type searches—research…how to do something, or what kind of products available. 3. transactional--- the more interactive type…purchase, subscription, making a donation, etc…
Getting inside the mind of the searcher. Research vs. purchase. For example: “car reviews” vs. “low mileage ford mustang” vs. “fast auto financing.” Stage in buying process…what types of words to use when. Personal demographics such as age/gender, and using proper words. Psychological (FUD- Fear, uncertainty, and Doubt). To compete, you need to know just how competitive the word is. You need to evaluate how active the competitors are. Are they doing PPC? How much are the bids? How optimized are the sites? What’s their linkage situation? Anchor text check. If you choose a keyword that you see a bunch of .edu’s and .gov’s in the results pages, then you may chose a different one.
Shows a very long list of keyword research tools, including, WordTracker, Google, Yahoo, Nichebot, SEO Research labs reports, Keyword intelligence, etc and others. Starts with an example of the Overture tool. Starts with a very general keyword, such as camera, which brings up a list w./count. The neat thing is you can click on each result and find further keywords that are more focused. Keyword Discovery: a pretty easy to use tool. Trellian data showed 343,694 results for the term “Camera.” Also has the ability to do a breakdown by time of year. The SEOBook suggestion tool, an open source tool, which takes all the free tools out there and compiles the results.
Next…WordTracker…she calls this one “the granddaddy.” She doesn’t used the KEI analysis, but this is a decent tool as well. Nichebot gives a top level view of some of the factors regarding the top sites for that keyword. (***looks very cool). She thinks the Google traffic estimator is OK, but not completely accurate. Also speaks about the new kw tool available within the AdWords/AdSense login area. Also talk about Trellian’s competitor intelligence tools. Keyword intelligence by Hitwise uses some neat data about competitors too. AdGooroo competitive intelligence tool also provides solid data. Keyword Analyzer from KeyRelevance.
In summary…when selecting keywords, use multiple considerations including relevancy, competition, and user intent based on buyer’s cycle. Understand the “why behind the search and you can better target how to respond. Test keywords and make adjustments.,
Lori Weiman – Keyword Max Her presentation will be focused on the paid side: diff methods to use to grow paid listings. Mining your referring URLs: organic and paid. Word Building with excel. Keyword research tools best features. Case study.
Mining your referring URLs, or the URL found in log files that show where it came from. When mining for these referrers, you should use a conversion tracking tool that will log them. Get a reporting tool that displays the conversion rate broken down by keyword. Try to get (for paid referral) the actual keyword AND the URL. Look at how to pull information out of organic listings. “Mine your P’s and Q’s.” shows a referral URL Google and MSN: look for a “q” and Yahoo look for a “p” in the string, which directly precedes the keyword used.
Mining URLs: look in log files, then find referrers that convert, then mine referring URL’s for the search query by using P’s and Q’s above. Mining paid search: Run analytics, find converting keywords, then find referring URL’s. This will help you find longer tail keywords.
How to use Excel to create long lists. Describes the process of “concatenating.” This is a system (which I use and love, btw) that combines columns in excel in order to help you grow your keyword list. (funny thing…there was a bunch of technical difficulties, and even my laptop froze up during the last part of her presentation – must be the spirit of Doug Heil at work here) From memory, Lori ended up with a nice short case study about a “Photo Marketing” firm, which is the term used in their industry, and the problems they had when they bid on that term. Ends up that they were very low in converting, mostly due to searches for “marketing photos” and “photos of marketing” (which she was surprised there was such a thing.
John Haney – Beyond ROI Concentrates more on the theory than the tools in his presentation. “Not all keywords are created equal.” When dealing with discovery, study your site carefully (remember that you may think differently about your product than those searching for it), know your competition, and remember that existing keyword popularity tools are simply to be used as guides. Study your log files, visit your referring URLs (mine those pages for possible keywords), and look for patterns that aren’t obvious. Remember that current events can sometime skew results.
Discovery is really a “best guess.” Log tail works. The more keywords you have, the likelihood of your being found increases. If you use a short tail word only, you are missing out on many relevant clicks. Remember that a search is juts a fragment of a person’s complete thought. When building longer tail, remember that anyone can discover keywords…it is up to you how far you will take it, and many others will simply get tired looking for more. Recommends “embracing the path less traveled.” Using analytics: It is time to get down to the science. Ask others what they think your site is about. The technology you rely on should be smart enough to help identify patterns. Everything ideally should feed together- tracking analysis, and research should feed on each other.
The “Dragon Tail” is a spike caused by a current event or an annual happening, or some less obvious ones like an election year or the search behavior after a hurricane, for example. The re will be peaks and valleys in the tail…for example more searches for automotive type things on “race day.” Hot market items come and go…so be ready. For example, when the Steelers won the Superbowl, many related searches happened. He repeats: keyword research never stops. He thinks you should find at least one keyword a day to add or remove, or you “are taking too long of a lunch.”
Shaun Ryan – SLI Systems Wants to talk briefly about internal site search. Finds that what people type into their own internal search boxes are often what they type into when searching in SE’s. Used the example of mining the data from Hollywood.com to find popular words. Not only should you be finding out what people are searching for, but also what they are clicking on at your site. Shows a great example of what is discovered when analyzing the entire realm of possibilities for a particular product. You can then categorize the keywords based on what they are clicking on, as well. You can also examine the seasonality of search term by analyzing your search box queries.
How to get the data? Instead of manually examining logs, use a developer to write a script. Of course search and analytics tools will also get you this info. Small case study about someone who tried this, and quickly got thousands of keyword possibilities. To repeat: mine your own search box queries and you will find some great keywords.
This is part of the Search Engine Roundtable Blog coverage of the New York Search Engine Strategies Conference and Expo 2006. For other SES topics covered, please visit the Roundtable SES NYC 2006 category archives.
SES NYC Tag: sesny2006