News Source: http://weblogs.hitwise.com
Published: Nov 07, 2008 - 05:34 am
Story Found By: toddmintz 1192 Days ago
Category: SEM
14 Comments
Sorry this site requires JavaScript to be enabled in your browser. See the following guide on How to enable JavaScript in Internet Explorer, Netscape, Firefox and Safari. Alternatively you may be blocking JavaScript with an advert-related or developer plugin. Please check your browser plugins.
News Source: http://weblogs.hitwise.com
Published: Nov 07, 2008 - 05:34 am
Story Found By: toddmintz 1192 Days ago
Category: SEM
14 Comments
Upcoming Conferences
Search Engine Land produces SMX, the Search Marketing Expo conference series. SMX events deliver the most comprehensive educational and networking experiences - whether you're just starting in search marketing or you're a seasoned expert.
Join us at an upcoming SMX event:
- SMX West - Feb. 28 - Mar. 1, 2012
- SMX Munich - Mar. 27-28, 2012
- SMX Toronto - Apr. 25-26, 2012
- SMX Sydney - May 1-2, 2012
- SMX London - May 15-16, 2012
- SMX Advanced Seattle - June 5-6, 2012
- SMX Paris - June 7-8 2012
- SMX Melbourne - Aug. 2012
- SMX East - Oct. 2-4, 2012
- SMX Stockholm - Sept. 15-16, 2012
- SMX Social Media Marketing - Dec. 2012
Latest News From Marketing Land
- Marketing Day: February 10, 2012
- Report: iPhone, iPad Now Drive More Web Traffic Than Mac OS
- Drawing Out Your Affiliate Program Marketing Plan
- Is Yahoo About To Make Another Big Mistake — With Its Display Ad Business?
- Google Analytics For The Enterprise: Perception & Reality
- - See more news at Marketing Land -
Upcoming Webcasts
Learn more about search marketing with our free online webcasts and webinars from our sister site, Search Marketing Now. Upcoming online events include:
- The Inside Scoop on Automating and Optimizing Your PPC Campaigns - February 21, 2012
Latest News From Search Engine Land
- Google’s Recent Domain Registrations: Labs, Screenwise & AdWords
- Google AdWords Gets Shared Lists To Eliminate Duplicate Data Entry
- SearchCap: The Day In Search, February 10, 2012
- Now, Mitt Romney Has A Santorum-Like Bing & Google Problem
- Search In Pics: Google Sugar Cube, Yahoo Valentines & Chrome M&M’s
- - See more news at Search Engine Land -
Comments
Brilliant. This pattern of a really really really long tail dwarfing the head also exists at ’niche’ level. It’s relevant for almost every web site and page. It’s why researching and targeting single keywords almost always misses the big picture. It’s why (at page level) SEO should research and target groups of keywords (keyword niches).
Well, actually Dustin concludes that there is no long tail... all search is tail. That might not be how SEO vendors want it to sound, but he wrote:In summary, the long tail aspect of the search is true, but the data tells us that there may really be no head or body. When it comes to search, virtually all traffic is long tail and the word “long” doesn’t do the length of the tail justice.I agree with Dustin that this "long tail of search" thing needs more study. My current column in Modern Domainer (hitting mailboxes next week?) notes that the long tail is not always where the money is. If you care about ROI you need to target what boosts the bottom line, not what has the highest search frequency. Track and target intent.Most of us understand that search intent is related to conversion performance. Let’s not take a step backwards and return to worshipping search query volume, even if it provides pretty graphs.
Worship volume? No. ROI? Yes. To do so we must measure reponse for different traffic sources, inc searches. This is about searches with keywords. Analysing single keywords is (in all but rare cases) at best a missed opportunity (miss the tail) and at worst a waste of time (sales for single keywords are too small). So analyse groups of keywords (keyword niches): volume and response rate. Volume x response = sales. (Response rate without volume is as daft as volume without response.) Most stats tools let you do this: filter for results containing ’seed words’ and result is analysis (inc response) of a related group of keywords. Compare results for each keyword niche and from non-search sources. Add some data from a keyword research tool like Wordtracker to find out the potetnial size of each keyword niche. Chase the money (ROI). Monitor marginal response rate and sales for each keyword niche and other groups of traffic - adjust accordingly (so you are chasing marginal ROI).
I’m not really sure the data from this post is really very useful. Data within one niche would be far more valuble and I suspect the traditional head, body, tail sort of graph would still appear then.
Don’t get it twisted........In 2009 and beyond...its moving toward the "fat belly"
rfergie: absolutely we each want data for our markets and niches. But what if the same pattern - tail (with fat bellies and other protrusions) dwarfing the head - exists within our niches? That is really really really useful. and it is so. Don’t forget you don’t see most of the tail in keyword research tools (inc. Google’s and Hitwise) as they are all samples.
Nunney: What if the same pattern doesn’t exist? Saying "it is so" doesn’t convince me.I’d agree with you that most of the long tail doesn’t appear in keyword tools, I’ve heard somewhere that 50% of searches are unique and never seen before. The question is whther or not there is any way to use this information. Even if you knew what it would be, its a waste of time to target a search query that will only be used once. I’d say targeting stuff that far down the long tail would be a waste of time.The point I’m trying to make is that comparing traffic volume across all search queries ever is not useful because no one is trying to compete on all search queries.
As with most things in life, it all depends on context.We work with a large insurance/financial services client that derives massive amounts of visits and conversions from one monster keyword, despite ranking well for tons of "long tail" keywords as well. The one monster keyword accounts for nearly half of all their visits and conversions.On the other hand, I oversee a sports site that derives over 1000 organic referrals per day, and nearly all of that traffic (99%) is derived from longtail keywords.
My primary goal with that article was to expose that how much traffic is in the long tail and that past references to the "long tail of search" didn’t take a look at large enough data sets, which quickly show there is virtually no head or body. Of course, if you filter down to a single topic, the chart won’t be so dramatic - you may see a head and body, but you will likely see a very long tail as well (the hard part, as rfergie explains, is getting your hands on that data). I’ve created similar charts I can’t share from my work in the food industry, where I was very successful at gaining the head and tail terms. There was more of a head and body, but the tail was very long and covered the majority of the volume (80-90% depending on how you slice it). In response to John’s statement:"Let’s not take a step backwards and return to worshipping search query volume, even if it provides pretty graphs."I think, if anything, the article shows that worshipping highly popular terms (the "head") due to their query volume & pretty graphs overlooks the great opportunity in the tail. Long tail charts never make pretty graphs - if you look at the first one, it is probably the ugliest graph I’ve ever created. Imagine if I graphed out the whole entire tail for my data set (14 million terms). My instinct was to immediate use a logrithmic y-axis, but that wouldn’t paint the real picture.I agree conversion and ROI is important, but there is great volume, conversion and ROI opportunities in the tail if you are able to successfully obtain them. Obtaining them is certainly the hardest part, but I’m here to tell you it can be done.
past references to the "long tail of search" didn’t take a look at large enough data setsIf you work with a large enough dataset, that fact alone will render any head/tail conclusion irrelevant.The realization that, if we add up all the world’s intents, goals and curiosity (+ typos), its aggregate volume will dwarf any single search query is not something I would find earth-shattering.In fact, this conclusion can be reached without any data, simply using common sense.
rfergie: "Saying "it is so" doesn’t convince me". Quite right. I give some evidence here. Others have also done so above. And I explore some of the consequences for SEO here.You don’t just target single keywords or long tail keywords. You research, analyse and target both by thinking in terms of groups of keywords (keyword niches).When optimizing a page, focus on a niche’s core or seed word but make sure there is a lot of copy to reach deep into the tail (Aaron Wall compares this trawling rather than fishing with a single hook).hugoguzman: On the finance site, the tail might contain many times more response than that single monster keyword (and I’ll bet it does) but if you don’t get the traffic, you don’t know.
Like @hugoguzman says, some verticals have "money terms". As Dustin reports, searchers are all over the landscape with queries. As nunny aptly outlines, optimization requires one to research around the content to determine where the money is. As SocialSEOMaven notes, there is a new "belly" showing up, which I think probably comes from influences like Google’s suggest integrated into the search page.All great duscussion... great to see on Sphinn. How much is due to Google being the only game in town, and *not* reflective of users/user intent?
@nunney - that’s just it, though. We do get the traffic for a ton of longtail terms.As @johnandrews reiterated, it’s about researching to understand the specific landscape you’re working with.There’s no "one size fits all" solution.
hugoguzman: I’m fully with you on no "one size fits all". Key is to have a process - of research, strategy and action - that gets refined according to results, preferably marginal results. One guide for the process is: if there’s money being made in a niche, refine the analysis, in theory until you reach a single keyword.Shame we don’t both have your specific example’s stats in front of us. (This chat, needs terms pinned down and details looked at to get much further.) I’ll bet we would agree.johnandrews: it is good to see a real discussion. Sadly rare. By the way, what’s the "belly"? I thought that was a joke.