The long tail of search demand has been around since the dawn of web search and, since that time, search marketers have been attempting to tap into the powerful stream that high quantities of unique content can provide. I recently came across some great data from Hitwise (about 1 year old, but still highly relevant) showing off just how substantive the long tail can be. Bill Tancer's post - Sizing Up the Long Tail - gives some stats:
...the head and body together only account for 3.25% of all search traffic! In fact, the top terms don’t account for much traffic:
• Top 100 terms: 5.7% of the all search traffic
• Top 500 terms: 8.9% of the all search traffic
• Top 1,000 terms: 10.6% of the all search traffic
• Top 10,000 terms: 18.5% of the all search trafficThis means if you had a monopoly over the top 1,000 search terms across all search engines (which is impossible), you’d still be missing out on 89.4% of all search traffic. There’s so much traffic in the tail it is hard to even comprehend. To illustrate, if search were represented by a tiny lizard with a one-inch head, the tail of that lizard would stretch for 221 miles.
Top 10,000 Search Terms by Percentage of All Search Traffic
The truth is my research is still greatly understating the true size of the tail because:
• The Hitwise sample contains 10 million U.S. Internet users and a complete data set would uncover much larger portions of the long tail.
• The data set I used filtered out adult searches.
• I only looked at 3-months worth of data (which were some of the slower months for search engines).
To help put this in perspective, I made a few spiffy charts that can help to illustrate these points:
In this first chart, you can see a representation of Hitwise's data from the four chunks Bill broke down.
In this next representation, I'm showing the classic "long tail" style curve, but color-coded to help show the various areas of keyword demand. Note that you could conceptually say that the 9,000 of the top 10,000 terms should technically fit into the chunky middle. Bill classified them thusly in his post, but I tend to think that at those demand levels, we're still talking about "head" of the curve figures.
For both of these graphics, there's a large, high-res version available by clicking the chart. You can find lots, lots more on our Free Charts page :-)
Ah I love the long tail, but I always think as amazing as the tail is, if you cut off the head then the tail won't work. Allow me to explain.
Most websites are going to have visitors arrive through a variety of means and often through totally unexpected search terms. As this article suggests, people that enter very specific search terms are likely to be very targetted traffic and if your website is worth it's salt, then they are much more likely to convert.
However, in my experience most long tail searches are a function of a "head" phrase" related to your business/product or service. For example if you sell "red widgets" that might be a highly competitve term, but you have still got to optimise for it because your converting long tail is likely to be something like "cheap red widgets that work with blue gizmo" or "discount red widgets free delivery". So your well optimised page for "red widgets" is far more likely to rank for these phrases.
So, my point is you don't optimise for the long tail, it's just a positive effect of proper SEO keyword research.
Yes, and no.
I agree that you still need to target a "declared search term" on the page/URL you are optimizing, and to some extent it is nearly impossible to directly optimize for the long tail due to its nature and scalability factors, but I wouldn't say outright that you don't optimize for the long tail.
Long tail optimization is first and foremost about crawlability ... removing the hurdles and exposing a site's content to the bots.
And then scalably, through programmatic optimization based on pattern research, strengthen a site's potential for ranking for torso and dominant long tail patterns; e.g., Identifying Long Tail Patterns.
I don't really see how that differs from what I said a geat deal to be honest, especially with regards to a quality site. Although you have clarified quite nicely one of the factors (crawlability) I meant by a site "being worth it's salt" (not the most technical definition by me there!)
I still stand by what I said because in practical (as opposed to ideal) terms, getting bogged down in the never ending-long tail just isn't possible. I'm on an hourly rate after all. That doesn't mean I don't love and embrace the long tail, I just find it grows naturally out of a well maintained head and body.
Sorry, guess perhaps you read more disagreement into my statement than was there. Just trying to highlight that an absolute statement about not optimizing for the long tail may be a bit extreme.
And being "hourly" as you stated, then the concept of long tail patterns should be ideal for you... it's about shifting more of that tail to the left.
Yes, you need to focus on the head and torso, and you need to continue to develop out rich, on target content, but then what?
I've researched a number of sites and have seen that, across the board, usually well, well over half of the search phrases that sent traffic to the site only delivered once or twice over the last 30 days. Guessing that the site you optimize for follows that pattern, I'd want to be tapping into any optimization strategy that increases my optimization efficiency and potential to grow the site's torso and tail. In doing so, it also strengthens the head.
It really depends on your strategy - if you're talking about on-page keyword placement and basic optimization, that's true to a point, but those tactics only get you so far. Long-tail SEO is more about content marketing and creating valuable, niche content that naturally attracts long-tail search. At the same time, this rich content draws long-tail anchor text and links. The days of just optimizing a site for a couple of keywords are over, IMO.
useful post rand, and helpful diagrams for use with clients.. however experience teaches they'll still want to be No.! on top keyword 'x' I think its a impossible battle to win.
You can win that battle - you just have to shift the client's metrics. It's hard to get them to ignore ranking unless you can put another number in it's place. Start to educate clients on actual search traffic, for example, and just how little of it (to their own site) comes from that one top keyword, and they eventually may change their tune.
I had a client who obsessed about ranking until we opened up their deep pages to spiders and tripled their search traffic (almost all of it long-tail). Suddenly, that one keyword didn't seem so important anymore.
That's some really good advice right there. I get many clients who assume that ranking number 1 on the SERPs for keyword x is the best place to be when they're wrong.
The only problem with this is that some clients just don't like to listen. By God, we should all try!
I quite like to create segments in GA to track the keyphrase a client is obsessed with and its long tail variants.
So for example, if the client is obsessing about the phrase "blue widgets" i'll produce a segment that includes all "blue widget" phrases.
In this way when the client comes back and shouts about not being #1 for "blue widgets" you can reply..
"ah but your traffic for blue widget related phrases has increased 182% in the last month"
;-)
Another great post on the long tail. But I'd like to take the issue one stage further - does anyone have any evidence that long-tail searches are more likely to result in a sale (or whatever outcome your site is targeting)?
I'd say that they're significantly more likely to do so - people searching for "levi jeans" don't sound like they're as engaged as people searching for "size 34 boot cut levi jeans with button fly". The Search Engine Watch article hints at this, but doesn't fully explore it and doesn't refer to organic search.
However, I'm slightly sceptical of people saying they successfully target long-tail terms. Their very nature makes them impossible to predict in a lot of cases and there's too many to optimise for anyway.
Best practise is surely to create big volumes of naturally put together content. (A blog is an ideal way of doing this as I, not coincidentally, blogged about the long tail myself).
Hi A A Catenaccio,
I recently did some research on searches containing different number of words (1 - 18 words long) and found exactly that - long tail keywords do in fact convert better.
See sections 4 and 5 of my post on Long-Tail Keywords
Although it was written with more of a PPC perspective, the principles are the same: people searching for long-tail keywords are generally further along in the buying cycle and are arguably more likely to buy.
I agree completely. Some people are obsessed with ranking for one or two particular keywords, often the ones that don't deliver a good roi. I find the best way is to rely on your analytics to show them the amount of traffic they're getting from the longtail.
@Bloom Media, my name is Harvey Blom ;-) with that said clients want to be ranked in top positions because they want to be next to there competitors some what more of an "ego thing" rather then what would be the best practice/strategy for their business at that time.
I think you should shift your strategy convincing your clients the long tail strategy. As you may know top rankings are often not immediately gained in competitive markets/keywords. Depending on strength, previous seo practices and history of the domain in SEM, it may take some time or may not be possible at all to aquire top rankings on competitive words.
I had many success addressing to our clients. I shift the attention of top rankings to more profit or better yet ROI on our service. That must sound great, right? They can not resist listing to the phrase "more profit" out of your mouth and should give you a window to explain the strategy further.. Explaining to them what is achievable in what time line and what is not.
Getting the long tail will get your clients some QUICK WINS and also put them in a better position to target the Competitive Keywords down the stretch.
If the top rankings are a must they can optain top rankings by using CPC campaigns and better local listings. This should always be done anyway. SEO and SEA are a perfect match ;-)
I'm relatively new to SEO and I fight this battle with clients every day... *sigh*
On a side note.. if anyone is having difficulty figuring out the exact long tail term to target, just use the Google search suggestions that appear below the search box when entering terms to see exactly what long-tail terms have been searched enough for Google to suggest the term.
very true, the wonder wheel is also pretty handy for doing this
Target the head, exploit the tail
Bill's article my favourite SEO article of all time. Understand its significance and your SEO changes.
As you read the article and start to realise the size of the tail you then get blown away by the almost throwaway lines that BIll was just playing around with 10 million searches (it's not many).
The trick for SEOs is to take this concept and apply it to the (relatively) small niches we all target. So Bill talks of the head, body and tail of all searches but in everyday SEO we can think of head, body and tail our target niches. Eg ...
You're targeting 'xmas trees' (the head) but think of all searches contain 'xmas trees' (the body and long tail) as there is more of them - a lot more. And even if you get top of G for 'xmas trees' then ignoring the tail (which is yours for the taking) is leaving money on the table.
Target the head, exploit the tail.
Great post on SearchEngineLand about analyzing your short and longtail search terms:
https://searchengineland.com/analyzing-short-long-keywords-using-google-analytics-29567
Also, for what its worth, using their filters, on my site I've seen the following:
43% higher pages per visit for long tail search visitors
55% higher time on site for long tail search visitors
Just thought I'd share that the data actually bears out the notion that long tail visitors are more engaged...
I just had a quick look on my analytics and sorted by time on site and this stands up true for me, it makes sense now you have said it but I would not have though of it nor worded it quite so well. I will be holding onto it for my next presentation
Here's an example that shows "why most long tail illustrations are wrong".
https://www.google.com/#sclient=psy-ab&hl=en&site=&source=hp&q=why+most+long+tail+illustrations+are+wrong&pbx=1&oq=why+most+long+tail+illustrations+are+wrong&aq=f&aqi=&aql=&gs_sm=e&gs_upl=945l945l0l1337l1l1l0l0l0l0l0l0ll0l0&bav=on.2,or.r_gc.r_pw.r_cp.,cf.osb&fp=5f6e1c9d40277296&biw=1280&bih=592
(Query valid as of November 25th 2011)
Top results are completely irrelevant.
I'm not trying to underestimate the importance of the long tail but there has to be a new illustration and chart to represent it. Not a funky inforgraphic but a valid scientific chart. Some comments here stress the fact that "clients won't understand and they'll ask to rank for keyword x".
I think we owe to give them a better illustration that doesn't need explanation. And let's be honest. How much time to you spend on average to explain it to clients? Shouldn't it be more straightforward?
Excellent post rand. I know i'm a bit behind since it was published and the data is probably not 100% accurate but it would be great to see the difference between then and now considering all the changes.
Good post, Rand! To my mind it's the common strategy of every good seo, marketer nowadays.
An excellent post that underscores the value of the long tail. Another aspect I don't believe was mentioned is the enchanced conversion potential of the long tail. Single world (top 100) search terms are by their nature general and non-specific. The long tail phrase takes into consideration the intention of the searcher - by using a LT the searcher has more certainity about what they're looking for. If you deliver the phrase in the title, and have a a supportive description, the likehood of that searcher becoming a buyer or lead is that much greater.
Legal related KWs are some of the most expensive PPCs terms on the planet - we (at LegalMatch.com) have found that combining lawyer with geo-specificity (and category) brings very targeted traffic that often bring conversion percentages above our most sucessful PPC campaigns (e.g. antioch adoption lawyers). Granted the overall volume isn't there (however it's rising as more and more searchers use the LT - some 66% are using 3 or more KWs - Website Mag 2/27/2009) but having 1000s of LT pages allows you to set more traps in the forest.
Great post, just a thought, in some markets, conversions are mainly on the fat head, for instance, in the online travel industry, you´ll find conversions by hotel names, but 90% of transactions are coming from unbranded terms, such as "London hotels"; is not all about traffic, I agree with the ideas shared here, exploit the long tail, but see and invest time in the part of the tail that is bringing you the conversions..
un abrazo
Mariano.
Your posts confirms what I was thinking and confirming in my way of acting when "negotiating" SEO services with Small Business Companies (my core SEO market).
As all we know, a Client aims always to rank 1st for words like "Phones", "House Rentals"... and we all know that without a big budget it's not possible to have success in the match against "giants".
I always try to make understand the secret of success behind the Long Tail SEO philosophy, and I admit that the Small Business CEO (who are usually the Owners of the firm) after a first bad reaction (they're so proud and sure of themselves), they usually are very receptive and understand the correctness of the Long Tail.
As said here in the Blog many times is really a question of educating your Clients.
Very detailed and fact basis post.
I love the way you people post.
Long tail are so important in every department. whether it is SEO or writing.
Nice post indeed. For our travel wiki Earth.org, we're counting on what you're saying is true :)
By the way: very useful charts. Thanks!
very usable info, thank you.
The question becomes, how does one effectively optimize for long tail, inexpensively... am thinking:
1.) Directory links for long-tail terms
2.) Additional focus toward on-page seo, ie. more extensive product descriptions for an ecommerce site
anyone else have any tips/suggestions?
hey rand,
thanks a lot for this awesome analysis.....
the charts are really useful.........
Very interesting, it makes me wonder whether clients would appreciate the effort taken to optimise and aim at the long tail. I can think of a few who would just say - "I'm not number one!!"
Nice one. Its a strangely pervserse phenomenon; the long tail. Things are, and should be, blindingly obvious but it is only when you do a simple deduction like you did that things really ARE that obvious.
I've done similar things on many occasions and am always impressed (and bizarrely) surprisedon the outcomes. er........Long live the long tail!
Thanks Rand! Beautiful charts as always! ;)
That makes total sense, but how do you find these long tail phrases then? Most of the keyword tools out there will not show these as the volume per phrase is too low.
really good, thanks for share it^
Good stastistics on the keywords on the global search perspective on the search engines thanks for it.
Long tail/long terms keywords bring good amount traffic and selling traffic of product and services but brand purpose it serves very little for it.
4 years later and I still refer back to this post. Thanks again!
The post is 4 yrs old but that idea in seo companies is coming now here... No more keywords not giving revenues but in top10:)
Nice post, Rand
Viewing the search demand curve, seems like 11.6k keywords belong to fat head. However, do you know the magnitude of keywords belonging to the chunky middle that we are talking about? just to know
That graph is great value, thanks. This is why I am a fan of Wordpress...the long tail keywords seem to be quite easily found, even if they don't appear on the same page.
Since I am not very familiar with SEO I always considered it kinda overated. I have began going some reading and I think I realize that is quite important.
Thank you for enlightening me.
Matt
https://wadja.com
I would like to know just how to capture some of that 89% in the long tail - I mean really capture it. Do you have any posting suggestions that could point me in the right direction?
Very nice analogy!
I love the "Fat Head" part of it, it makes total sense, can't believe its taken this long to be put to practice.
Oh, and awesome graphics, I have sneaky feeling they will be making their way into a presso or two... obviously I will source you ;)
I am a big fan of long tail search for my clients. I believe people are becoming more savvy enough to use more words in search terms in order to receive more relevent results.
When I assist clients in keyword suggestions prior to optimization, they send the most generic of phrases...usually 1-2 words phrases. I use those as a guide on determining more relevent phrases. It may include regional areas of business, more detailed words that a person may want, add if relevent, cheap, discount etc on to the generic terms. Point being, the generic or highly targeted phrase is still within the long tail search, thus allowing the ability to be found for the short phrase, but a better chance of achieving top visibility for the long tail search, which as we know converts so much better. To me, short phrase, generic phrase searchers are lookie-lou's and long tail searchers are buyers.
Educating the client on long tail has been very successful and once explained they really do understand the concept and more importantly the ROI of the organic searcher to their site. And yes, in time they actually are found under generic terms as well :-)
Ok...its holiday time and I think I need to buy a turkey. Thanks for the article and may you all have a great day!
Rand, I think you've missed a great opportunity to actually demonstrate the actual search volume of the long tail. In the last graph the actual area under the graph should be 70% of the whole graph. This would leave it really fat and probably look a bit out of whack, but that is the reality it should be showing. Similarly if you add up relatively all of the percentages in the Percent of All Search Traffic graph then it ends up being over 100%, if you don't do that the 70% looks much larger in relation.
Aren't the constituent parts of the long tail for the most part the same people at different parts of their purchasing or research cycle?
If that is true what real insight does traffic for head middle or tail really tell us? More people searching for "small heavy red widget" than "red widget" ?
It might be that if you apply the conventional marketing model AIDA to the longtail stats it just reinforces what we already know i.e. as the consumer psyche moves through awareness into interest and desire then search activity/volume increases exponentially.
So as Bludge said without gaining awareness at the beginning of the AIDA cycle (the head) your effectuveness at the tail may not be great
Does that get more into how you track conversions? Are the conversions coming form a search done for head/middle/tail?
Just wanting to make sure I am following you correctly.
Can you clarify what you are defining as long tail? I've heard several explanations that are only slightly different and am curious what you are calling long tail.
I have defined "long tail" keywords in my Google Analytics as any terms that are 4+ words. For example, "blue dress" wouldnt qualify, but "ralph lauren blue dress with white straps" would..
And the article I had posted above shows exactly the regular expression you would have to use to get those filtered properly.
Thumbs up for that free charts page!
Interesting, has anyone ever worked with the company YourAmigo?
https://www.youramigo.co.uk/news-and-events/youramigo-in-red-herring-top-100.html
They are huge promoters of long tail optimization and its what their product focuses on. I am in discussions with them at the moment.
Agreed, great graphs as usual from seoMOZ. This definitely reminds us to put more emphasis on quality content generation within your specific niche/vertical.
Good stuff, I am always looking for ways to quantify keyword market sizes in business terms and this can help. It would be useful to tie in conversion research around 2, 3, 4, and 5 word phrases to guestimate revenue potential
Rand,
I love the graphical display of the long-tail that you have presented. Our Senior Leadership team is very visually-focused so this will be a big help.
Thanks!
sharon
Thanks for the free charts. I'm looking forward to using them!
Nice post Rand! The charts you use really make a difference.
I always cringe when clients or potential clients say they want to rank #1 for "holidays" or "property" or some other ultra broad search term without realising the competitiveness of such terms and the fact that people that search using such terms are very unlikely to convert.
It's great (not to mention a huge relief) when you explain to the them about the long tail of search and how it is more convertible and easier to rank for etc, and they finally get it.
Build lots of highly focused, unique, well researched and quality content and make sure the engines can crawl it and you will be on your way to long tail heaven.
We at JSM get this all the time. Recently we had a client who wanted to rank for the term IT (!)
Education on long tail and search trends is necessary before kicking of any campaign.
Great post Rand - love the charts.
Surely this 'Search Demand Curve' should not touch the Y axis at any point?
Just as the number of keywords is (theoretically) infinite, so the number of searches is also surely (theoretically) infinite....?
Also, if the curve crosses x=0, this implies that the # of keywords can be zero, which makes no sense.
Just my 2 cents...
Interesting piece, Rand...and I'd add only that in my own canuckseo.com blog yesterday, in looking at how us Canadians differ from you all south of the border, our own long-tail queries fail 30% of the time!
Intertesting stuff here to ponder too....gotta print this one out and sit over same, eh!
:-)Jim
as a mere landscape architect (ie did not know what I was doing regarding SEO) - I started backwards- I named all my pages with about 6 words placed 'long tail's throughout text, headings, photos etc. In the last 4 weeks I shortened page names, headers etc and now in top top 3 for a variety of 3 key words combos, and 1st pages for 2 key words combinations - i am pretty happy but don't really understand how it happened - it was literally overnight
Long tail keywords are the way to go. I recently found a really unique and cool business just by searching (long phrase) for exactly what I wanted.
I agree that a sound SEO strategy must build a solid base in attacting traffic from the long tail. However from say a commercial pressure, you really need to be able to "turn the tap on" so to speak when the sales team are crying in their starbucks coffee cup.
I've worked for one the UK's biggest DELL resellers and that pressure was based on daily (changing sales targets). So there I feel is an working environment where the dynamics may not suit long tail, if the traffic count is lower and volume of sales inquiry is king.
Brilliant piece. Well researched and presented in a thorough and useful way. Will be sharing this.
Your almost close to applying Pareto ruling here with long tail - not quite but close - interesting post Rand.
D
Interesting article, thanks. Targeting long-tail search has been a really effective strategy for several of my clients. As MarketingGrin points out, it's about understanding budgets and devising the most efficient key-phrase strategy.
Great illustration, will use The Search demand Curve one in my Uni presentation tomorrow . Tx
BTW: Why is logging in such a pain? Site doesnt remember where I am when i login and i have to renavigate here (in FF on win7)
Thank you very much for your valuable description
Dont know many sites that are trying to rank for many of the top 100 keywords, top 1000 or top 10,000.
It would be nice to look at some example keywords and look at the longtail patterns that are generated around that first keyword. For us this means adding logical prefixes and suffixes to the keyword to allow us to bid on less popular longtail keywords. Thus exploring what additional keywords to searchers add to simple keywords to refine what they are looking for.
Company A might go after 100 keywords that reflect their business model. Company B might go after those same 100 keywords and the 10,000 longer tail keywords surrounding those first 100 keywords and find much greater success.
I guess I miss some of the orginal posts point. Thank goodness 40% of all searches arent for the same 100 or 1000 keywords. What we need is to figure out which 10,000 or 5,000 keywords make the most sense to the business we are working with.
Long tail search is an effective SEO technique. The main point here is understanding your SEO budget and building an effective keywords strategy in line with your SEO budget. The cost of getting to the top of Google for popular keywords or phrases is very much dependant on your industry and what SEO campaigns your competitors have in place but tends to be expensive.
Long tail search terms are obviously not going to generate the same number of visitors in comparison to being at the top of Google for the more popular terms, however it is far better to be at the top for a less popular phrase that is still well used, generating modest visitor numbers, than languishing at the bottom for popular keyword searches.
If your SEO budget allows it, running a long tail search page alongside your other SEO campaigns for the more popular searches is a great technique of gaining additional visitors that otherwise may not visit your site.