Editor note: I’ve been following Justin Kan’s YouTube channel for a while and I found this article’s title on YC’s Hacker News. I just couldn’t find the full article because Atrium.co was shut down in 2015. So I went so far as to copy the full article on Wayback Machine. It’s full of gold, enjoy it.
Every CEO goes through times when they say to themselves “things just aren’t working.” While this might very well be true, what is not true is the feeling that generally accompanies it: “I’m all alone and this is a unique problem.”
After seeing startups from the inside and outside, from seed rounds to billion-dollar companies, I can definitely say that all problems CEOs experience are repeated. Across board meetings, CEO dinners, coaching sessions, leadership retreats–I hear the same startup problems mirrored back to me time and time again.
Justin Kan is an internet entrepreneur and investor known for founding various companies, including Twitch–a video game streaming platform (acquired by Amazon for $970mm). He served as Partner at Y Combinator, where he impacted over 900 companies and funded more than 130. Currently, Justin serves as CEO of Atrium, where he's building technology to revolutionize the $450bn legal industry.
serpIQ is back with more data that we’re excited to share with the SEO community. If you haven’t read our first post on domain age and rankings, make sure you check that out. We got some great feedback on that post and we’ve taken that into consideration for this post, which is number 2 of 5 in our “Why Data Driven SEO Trumps Guru Opinions” series.
As we’re sure you know, SEO is an extremely fast-paced industry. Just this year Google rolled out more Panda updates and the recent Penguin update as well. Both have shifted SEO strategies considerably. It’s very tough for us to figure out exactly what works and what doesn’t due to the veiled nature of the search engines, so relying on your own testing and experiences is going to be one of the best methods for becoming an expert SEO. At serpIQ, we’re big believers in trusting yourself and your own testing as opposed to listening to the opinion of a famous SEO. That’s not to say that what they say isn’t valid – it just isn’t experienced by you in your own unique situation.
We’re here to provide some hard data as a starting point for you. We will be giving you our conclusions, but want to stress that you can and should critically examine this post and draw your own conclusions as well. Let’s get started.
How Much Content do Ranking Pages Have?
When it comes to content length, there are a lot of opinions as to what’s considered “good.” While there are no hard rules, it’s generally assumed that more content is better…but is this true? Let’s kick off the post with a graph of the average content length of domains on the first page of the SERPs:
As you can see, there is a drop in content length as we move from first to tenth position. On average, 10th position pages have 400 less words on the page than first position pages. This does point to the trend that higher ranked sites have more content, but keep in mind that this graph is not segmented in any way – this is just a graph of all of the SERPs we’ve analyzed.
Content Length by Domain Age
It’s tough to draw any substantive conclusions from the chart above, so we took it a step further. We broke our content length data down by domain age to take a closer look at what factors and trends we could tease out. Here are the segments we chose:
Less than a year old
Between 1-10 years old
Over 10 years old
Here we see that fresh domains generally have less content on the ranking page, averaging around 1800 words. Domains 1-10 years old are at around 2100 words of content, while domains over 10 years old close in on 2800 words.
Based on both of the graphs so far, we can say that there is a possibility that higher ranked pages have more content because these pages are also older. Data in our first post showed that 55% of pages ranking in the top 3 positions were over 10 years old, which is a high percentage when you consider the fact that there aren’t a lot of domains >19 years old and that the volume of domains registered has drastically increased in recent years. Additionally, the “Wikipedia Effect” could be playing a big part here. It ranks highly in many different SERPs and typically has a huge amount of content on the page.
For domains less than a year old, it could be the case that new websites are just starting to build up a solid base of content and may not have their pages fully developed.
Content Length, Domain Age, and Competition Index
The next step we took was to segment the SERPs in terms of their Competition Index, which is a metric that serpIQ generates to determine the difficulty of ranking for a particular keyword. By doing this, we get a picture of how the competitiveness of a SERP affects the amount of content on ranking pages. Our index ranges from 0-100, but the vast majority of SERPs are in the 25-75 range. We decided to look at content length and domain age in these CI ranges:
CI of 25-50
CI of 50-75
CI of 75+
Competition Index Between 25-50
To give you an idea of the type of keywords that would show up in this Competition Index range, here’s an example keyword analysis preview that our users get (this is just the preview, the full competitive analysis report is much more thorough). In this example, “phoenix electrician” has a CI of 31, indicating that it’s not a very competitive niche and would take less effort to rank. Let’s take a look at the content length breakdown.
We see slightly less content in all domain ranges for this CI as compared to the overall graph. It’s not enough to say anything conclusive, but generally it takes less to rank a lower competition SERP so there is the possibility that you could get by on thinner content if you’re trying to rank for something at the lower end of this CI range. When competition is this low you can generally best your competition by focusing on some of the more important ranking factors, like a high-quality backlink portfolio and rock-solid onpage optimization.
Competition Index Between 50-75
“Poodles” is a great example of a SERP that is moderately competitive. The term has a variety of sub-niches that people are competing for, but it isn’t a massive term, especially since it is a sub-niche of the pets or dogs niche itself. Here’s what the content length data looks like for SERPs with a CI of 50-75:
Content length data here is essentially the same as the previous graph, which makes sense because a large percentage of SERPs fall in the 40-60 CI range so there is a bit of spillover in this range compared to the 25-50 CI range.
Competition Index Over 75
Here’s an example of an extremely competitive SERP. “Apache” is a very broad term and has multiple definitions, most of which are pretty popular have a broad range of subniches that can be explored as well. It’s no surprise that it has a CI of 77. How does the content length shape up in these types of SERPs?
Here’s where these CI graphs get interesting. While CI ranges of 25-50 and 50-75 produced nearly identical content lengths over all respective domain age ranges, ranking pages in SERPs with a CI of 75+ have less content overall. Here’s the breakdown:
Less than a year old: 450 less words than SERPs with 25-75 CI
Between 1-10 years old: 300 less words than SERPs with a 25-75 CI
Over 10 years old: 500 less words than SERPS with a 25-75 CI
Our hypothesis is that pages that rank for extremely competitive keywords are getting there with massive amounts of backlinks, social signals, and other factors. In essence, in the aggregate, pages ranking for SERPs with a CI of 75 or higher are well branded and have a lot of other signals going for them, so content length is just not playing a very significant factor in their ranking whatsoever.
It’s evident that a fully fleshed-out site with a lot of quality content is going to be favored by Google. In fact, the lasttwo monthly updates from Google have shown that they’re actively working on improving their detection of poor quality content.
For most SERPs it looks like at least 1500 words is a good target. This isn’t a steadfast rule – you’ll need to adjust this target to fit the niche that you’re in. If you’re working on your own projects or even on client websites, keep in mind that not all content is equal. If writing isn’t your strong point, finding someone who can create compelling sales copy, blog posts or informative content is going to pay off in a big way down the road.
That being said, it’s likely the case that focusing on your branding, quality linkbuilding and social engagement will pay off more in the long run, especially if you’re targeting some competitive terms. As long as you make sure that the content on your ranking pages isn’t too thin, you should be covered in this area.
We hope this has been an illuminating post for you. It’s never easy to figure out what Google’s up to. Algorithm changes are taking place at least twice a day and Panda and Penguin have hit some SEOs particularly hard. To truly gain expertise in the SEO field you need to be constantly testing and evaluating different strategies and we sincerely hope that we have provided some useful data for you.
*Editor’s note: Posted by Kevin Espiritu on serpIQ at April 26, 2012. For some reason, serpIQ was shut down. But I think it is so valuable that I had to “recycled” it from Wayback Machine and keep a copy here for you founders.
I'm the marketing and operations guy at serpIQ. When I'm not working with serpIQ, I run my own inbound marketing consulting business and I'm also involved in the sustainable food, hydroponic, and restaurant businesses.