It’s time to reconsider how we think about long tail keywords. And it’s also time to reevaluate the process we use to research them.
Search Google for ‘long tail keywords’, click on a result in the top 10, and you’ll get a definition that goes something like this:
You’ll also get a crazy long process for finding long tails to target, which will normally involve:
- lots of manual Google searches;
- multiple tools;
- lots of ‘rinsing and repeating’;
- huge spreadsheets;
- hours of ‘exhaustive’ research;
- trying to shoehorn these keywords into your header tags and content
But you know what?
These articles are all out of date.
And the research processes are a complete waste of time.
Now don’t get me wrong:
Long tail keywords are still super important for SEO.
But these days:
Finding (and ranking for) them is simple.
In a moment I’ll show you how. But let’s start with that definition.
Why Most ‘Long Tail Keyword’ Definitions Are Wrong
Most definitions for long tail keywords start with some arbitrary length.
Depending on which article you read, that length might be 3 words+, 4+, 5+…
Um… so which is it?
Well actually, the correct answer is none of the above.
Because the thing that makes a keyword long tail or not has nothing to do with length (although admittedly you won’t find many 2 word long tails).
So if it’s not about length, then what is it about?
What Really Makes A Keyword Long Tail?
There are two elements that truly make a keyword long tail:
- Search volume
And actually, the two go hand in hand.
Because the more specific you get with a search phrase, the less volume there is likely to be.
In fact, many long tails will have ZERO existing search volume. That’s because around 16-20% of daily Google searches are for completely new phrases that have NEVER been searched for before.
Which might make you think that they are a waste of time.
But you would be wrong.
Because while individual volumes are low, long tail keywords actually make up around 70% of all search traffic on the web.
So even though individual volume is low, the fact there are so many phrase combinations (probably an infinite number), means that the collective volume is high.
And as a bonus, because they tend to be very specific, long tail keywords can be great for conversions.
Turns out some people weren’t persuaded by David’s explanations, so I decided to jump in and give some extra details.
Search volume or “search demand” (or simply “how many times people search for this keyword per month”) is the ONLY thing that differentiates “head” keywords from “tail” keywords.
The name “long tail keywords” originates from an illustration called “search demand curve” (which you can see on the image above), where the “head” of the curve consists of a few keywords with insanely high search demand and the long “tail” represents a huge amount of keywords with very low search demand.
And obviously the number of words in a keyword has absolutely nothing to do with differentiation of keywords into “head” or “tail” of the search demand curve.
David also mentioned “specificity”, but it is rather a byproduct than a defining trait of a long tail keyword.
And I have a great example to illustrate all of the above.
Here are three keywords:
All three of them have the same number of words in them. All three of them have the same level of specificity. But one of them is long tail keyword while the other two are head keywords. Can you guess which one is long tail?
Now let’s go for that definition.
What Are Long Tail Keywords?
Long tail keywords are phrases with low individual search volume, and often a clearly identifiable search intent. Collectively, these long tail phrases make up the majority of search traffic on the web.
No mention of phrase length and much simpler right?
(If you don’t agree, then let me know in the comments)
So with long tail keywords defined, let me explain why I believe those multi-tool, laborious research processes should be consigned to a Wikipedia footnote.
Google’s Transition To Topics Over Keywords
One of my favourite analogies about the way Google ‘thinks’ is AJ Kohn’s wonderful description of search engines as ‘blind 5 year olds’.
“Words are of great importance to search engines. It’s one of the easiest ways it can categorize a page. But it is not reading the page like you or I. A search engine wouldn’t score well on a reading comprehension test. Instead it’s trying to understand the page by what words are most prominent, based on the number of times a word is mentioned and the size and placement of those words.”
AJ Kohn blindfiveyearold.com
It’s easy to grasp:
Search engines are dumb. You have to spoon feed them information.
But that analogy is from 2008. And a lot has changed since then.
Our blind 5 year old is now a teenager. She’s getting smart.
Here’s a simple example.
Google’s Understanding Of Synonyms And Similar Words
I’m going to illustrate here with some short (or at least medium) tail keywords. Don’t worry, I’ll get to the relevance to long tail in a moment.
Take a look at the following 4 keywords:
- creepy photos
- scary photos
- terrifying photos
- horrifying photos
I’m sure you’ll agree, they are pretty much the same thing. That’s easy for us as intelligent, literate humans to understand.
But, for our blind 5 year old Google it wasn’t so easy.
Unless we spoon fed her the connection by including ALL of these phrases on page, she wasn’t going to be able to tie them together.
Which lead to lots of unnatural sounding, written for SEO content, and one of my favourite SEO jokes:
How many SEO copywriters does it take to change a lightbulb, light bulb, light, bulb, lamp, bulbs, flowers, flour…?
So how about our teenage Google? Can she make the connection?
Let’s find out.
Take a look at the following page content:
Here is the HTML title tag for the page:
<title>The 30 Creepiest Photos Ever Taken</title>
I’m sure you’ll agree that page is optimised for the keyword ‘creepy photos’. The words ‘scary’, ‘horrifying’ and ‘terrifying’ aren’t included anywhere on the page.
But here’s the cool thing:
It actually ranks in the top 3 spots for all of them.
Position data from Ahrefs ‘Organic keywords’ report
Which tells us 2 things about the way our teenage Google ‘thinks’:
- She now recognises the connection between the words
- She also realises that if a page is a good fit for one of these keywords, it should be a good fit for them all
But this is just the tip of the iceberg.
Because that page doesn’t just rank for 4 keywords. It ranks for 564 keywords!
Keyword data from Ahrefs Site Explorer
In the first few results, we can see that Google recognises ‘photos’, ‘images’ and ‘pictures’ as the same thing. The page ranks for them all, without being directly optimised for them.
But the above are still short/medium tail keywords with reasonably high individual search volume.
It’s when we start looking at long tail keywords that things really get interesting.
First though, let me explain how (and why) the above is possible.
The End Of ‘Words On Page’ SEO
It’s clear that Google is now able to:
- Group keywords together into topics;
- Understand words with the same meaning;
- Look beyond the ‘words on the page’ when deciding which content to rank
But actually, despite some recent high profile articles professing otherwise, this is not particularly new.
Because they have been doing that since (at least) 2013.
How Google Hummingbird Changed The SEO Game
In August of 2013, many SEOs noticed an increase in traffic for content rich sites.
I personally noticed the spike on an affiliate site I ran at the time. The site was small (c 30 pages), but each page featured detailed, high quality content targeting a top level keyword (or topic) in the niche.
This increase in traffic was due to the roll out of Google’s Hummingbird algorithm.
75% spike in search traffic after roll out of Google Hummingbird
Google’s stated aim with Hummingbird was to better understand the meaning behind queries, rather than focusing on matching specific words to content on a page.
Danny Sullivan explains this change is his article on Search Engine Land.
“In particular, Google said that Hummingbird is paying more attention to each word in a query, ensuring that the whole query — the whole sentence or conversation or meaning — is taken into account, rather than particular words. The goal is that pages matching the meaning do better, rather than pages matching just a few words.”
Danny Sullivan Search Engine Land
Now I have long been an advocate of writing natural copy for humans (not search engines), and covering topics in depth.
It’s the practice I have followed for as long as I have been building (and optimising) websites.
But for those SEOs who had not, this was a game changing moment. Particularly as it followed Panda, Google’s low quality content killer which launched in 2011.
The End Of Cookie Cutter Content
You’ve probably read articles which state that ‘500 word blog posts no longer cut it’. You may have even heard it said by me.
And Hummingbird is the reason why.
As Google’s understanding of content improved, it began to prefer detailed pieces covering topics in depth, over short pages targeting individual keywords.
And in fact, I picked the keywords in the illustration above for a reason. That’s because our own anchor text guide ranks top 5 for them all.
The long tail keyword ‘how to create anchor text’ is a particularly effective example.