I’ve seen the 80/20 Rule mentioned in passing a bunch of times lately in the SEO world, but it’s not quite being used in the fashion it was originally intended (and later explained in the book The 80/20 Principle
).

UPDATE: Looking for 80/20 Link Building? See this.

The latest mention of this came from John Doherty’s excellent interview of Leo from Buffer which you can find here. Definitely watch it, it’s a phenomenal interview.

But I’m afraid, like a game of “telephone” people hear about this thing called “80/20″ and the true meaning and application of it changes and eventually gets diluted and transformed beyond the original intent. My post and video (at the end) is an attempt to clarify any misconceptions and show an example of how it can be used in SEO and online marketing.

I recorded a NoBoard SEO about 80/20 (which is also embedded at the very bottom of this post – this post actually started as the description to that video in YouTube and just kept growing… anyhoo…)

What 80/20 Is Not

First of all, I am definitely NOT criticizing or “calling out” Leo for using a variation on the 80/20 definition. I’ve heard it misspoken about in lots of places by lots of people, so I think there’s a morphing of the definition going around in general.

Anyhow, in the video he explains they spend 80% of their marketing time on one type of content and 20% of time on the other. Trouble with that is we only have one set of data – time spent on content.

not-80-20

Simply dividing up time spent (or types of content into content A / content B) is only the first set of data. It’s only an input or effort. 80/20 requires comparing this to the result or output.

What 80/20 IS.

A true 80/20 requires two sets of data.

It’s typically used to analyze “effort -> result” or “input -> output”. But the two sets of data can be anything where you can find a relationship between the two (such as where Pareto discovered the phenomena – by studying the % of population in Italy vs. wealth distribution).

I’m just gonna “borrow” an image showing a simplistic example (shh)

80-20-simple

Perhaps Leo was insinuating the measurement of time spent on different types of marketing leading to a result (for example the 20% time leading to an 80% result). But even if so, I want to be sure people catch and understand that implication – as it’s important to getting the value out of 80/20.

There’s another, albeit slightly more complicated way, to view an 80/20 analysis. Again – two sets of data. (I’ve “borrowed” this chart as well… )

80-20-chart

The important thing here is the addition of a 3rd data point – cumulative percentage. You use cumulative percentage to see how quickly you get to 80% (or a high percentage) of the total amount of output.

How To Use 80/20 For SEO

Hopefully that all makes sense up there. To use this principle in SEO, you just need two sets of data – an “input” and an “output”. Some examples;

  • Content -> Organic Search Traffic
  • Backlinks-> Referral Traffic
  • Landing Pages-> Bounce Rate
  • Followers -> Retweets
  • Unique Visitors -> Conversions

Let’s do the first one on my personal site (this one) and hopefully it won’t take long ’cause it’s late :-)

I want to answer the question: What 20% of my content is bringing me 80% of my traffic? Or – where is the imbalance in content effort vs. reward?

Step 1 – Create Landing Page Segment For Blog Posts

This may differ for you – but I want to look at traffic landing on blog posts.

blog-post-landing

Step 2 – View For All Time And Export To Excel / CSV

We want to view the data for as far back as possible (I do in this case anyway, I suppose you could do this annually, quarterly, monthly etc).

And we’re going to export the report to excel.

all-time-export

(note: you can also segment out google traffic, referral traffic etc for a deeper level of analysis – here I’m just doing all traffic).

Step 3 – Clean Up CSV Export

Simple… all you need to do is delete everything except “visits” and delete any landing pages that don’t belong (like image URLs or comment URLs etc).

Step 4 – Calculating Percentage

Here’s what your excel sheet should look like so far – before we get cumulative percentage we’re going to just get the percentage of traffic each page is generating.

cumulative-empty

We can see what pages are bringing traffic, but we need to calculate the cumulative percentage to find the 80% cutoff.

Get the total number of visits at the bottom with SUM

sum-total-visits

Now we’re going to get the percentage – you need to use the total number of visits (calculated with the SUM above) and visits per page to get your percentage. The formula happens to look like this for the first one;

percentage-formula

The formula is =(B2/$B$54)*100 – assuming your cells are the same.

Drag down;

drag-down-percentage

Step 5 – Get The Cumulative Percentage

This is the exciting part where we reveal the 80/20 effect!

The first cell in cumulative is just going to equal the cell to the left of it;

first-one-equals

NOW we’re going to get the SUM of the first cumulative, with the second page’s “percentage”

sum-final-drag

Get the SUM of D2+C3 and drag down for all cumulative percentages.

Step 6 – Zoom Out For Results

How many posts or pieces of content did it take you to get to 80% of your traffic? Took me only about ten;

final-analysis

From here you can make your charts ‘n stuff – this post actually does a decent job at explaining.

My example above is almost a textbook perfect “80/20″ – but that’s also another misunderstanding… 80/20 does NOT mean exactly 80% / 20% – it can mean any imbalance which strays from 50/50.

It could be 75/35 or 90/10 or 99/1 – it’s all an 80/20 type relationship.

Step 7 – What’s Really Going On?

Everything thus far was to arrive at the data. We revealed the typical imbalance of input->output. But what’s REALLY going on, and what does this data tell us?

Well if I quickly glance across the top content, I spot some things in common;

  • They are all written about something else popular already, or at least interesting – like Google, or Ben Folds, or Ifttt
  • They were generally timely. I wrote the Ifttt piece right as it was coming out. The Followerwonk piece just after Moz acquired it etc… the “removal of the left sidebar from Google” I know has seen a lot of search traffic, as I posted about it with screenshots before most anyone else, as I think I was subject to an earlier test of it.
  • Some pieces leverage a relationship or involved someone who went and shared it – like the Iftt piece I wrote after seeing a tweet of Rand’s

But more importantly almost;

What’s NOT Working About Posts With Less Traffic?

I see some commonalities;

  • The topic is only important to a few people at most.
  • The content wasn’t timed well.

I could go on, but you get the idea I hope. Find some similarities in the content that isn’t performing well.

One caveat to this, is that some content has been around longer. But conversely, other content had more of a social push since I have a lot more followers now than I used to. Deeper analysis and segmentation would have to be done to factor these influences in or out.

The Takeaway

Use 80/20 to do two things really;

  1. Figure out what’s working and do MORE of it
  2. Figure out what’s NOT working and do LESS.

Here’s the NoBoard!

I HIGHLY encourage everyone to check out the book too!

About Dan Shure

I'm Dan (Google Plus Profile). I've been helping businesses improve their websites since 2007. Improving your bottom line is my number one goal. My obsessive nature and love for SEO as a true craft doesn't hurt either.

7 Comments

  • January 23, 2013 Reply

    Kane Jamison

    So, I like the post, and I agree that it’s typically not used to refer to the concept of 80% of results caused by 20% of activities. But, I want to go back to Leo’s comment.

    Your takeaway (which is completely accurate) is that we should be using the Pareto principle to (A) do more of what’s working and (B) do less of what’s not.

    His comment said that they try to focus on doing 80% of what has worked for them in the past and has proven results. To me, this is really the second phase of utilizing the 80/20 principle – implementation of your findings. Figure out which activity causes 80% of your results, and focus more time on that activity. At the same time, they’re recognizing that what has worked for them in the past may not always work, and allowing for that 20% experimental time to find new activities eventually build into their 80% tactics.

    Love the application to your blog content. Brings a couple ideas to mind on how to apply outside of blogging efforts and your other examples that I may share at a later point, but have to simmer on them for a bit.

    • January 26, 2013 Reply

      Dan Shure

      Hey man… yep, absolutely you have to apply what you learn. I just fear that people are going to think 80/20 means “split up your time into 80% of one activity and 20% of another” when it really means “find out what 20% of inputs result in 80% of outputs”. And I fear people miss one of the main points of 80/20 at it’s core – is that you spend lots of time in deep thought and analysis – to only act on the few things that really matter.

      Thanks man… I’d love to hear how you apply it to SEO!

  • January 27, 2013 Reply

    Sahil

    Great post Dan. SEOs should be careful in not applying the 80/20 principle in all aspects of seo. Especially post panda and penguin updates, the webmasters have to be extra careful in not using the 80/20 in old link building tactics (may be spam or grey hat) that once used to work and will never work now as it is considered as a trip to search engine penalty.

  • January 28, 2013 Reply

    Ted Ives

    You can further apply the 80% to the 80%, what you get is 16% of the overall effort gets you 64% of the results. Makes you really think about ROI.

    This sort of thinking applies well to paid search; it doesn’t mean you don’t go after that additional 20% of conversions though, often that 4x extra effort for it is well worth it.

Leave a Comment

Error: Please check your entries!