I encountered the phrase “surrogate marker” several years ago, and it has stuck with me ever since. We are surrounded by them, although they are seldom identified or examined. And the world of content marketing is saturated with them.
A surrogate marker is when you use an easy-to-test variable as a substitute for a more-difficult-to-test variable, and assume that the correlation is identical for both. For example, say that you understand that lower cholesterol = being healthier = living longer. So you conduct a series of tests on habits or medications that reduce cholesterol. When you have a scientifically reliable result for your medication, you may claim (and it makes for better press if you do) that this medication increases your lifespan. But what if someone does another study and questions the core assumption that there is a relationship between cholesterol and longer life (something Malcolm Gladwell actually explored this very topic recently in his excellent Revisionist History podcast). If there is no correlation… in other words, if your assumption of the causal correlation between your actual variable and your substitute variable is incorrect… then your science is good, but your conclusions are bad.
Blog post length is one of those things. For a while, it was conventional wisdom that nobody liked to read long posts, so you needed to keep them short and sweet. Then everyone did a bunch of research and found that 1600 word posts rank the best, so now everyone is scrambling to write longer posts. As a freelance writer, I am seeing clients want longer and longer posts, and as a reader of content, I am seeing it increasingly on blogs, particularly in the inbound community.
Blog Post Length is a Surrogate Marker
As I mentioned, inbound marketing is full of these, and this topic in particular is riddled with them. Blog post length is assumed to correlate to:
- high-quality content
- content that is valuable for the reader
- high-ranking content
- sharable content
But it goes even deeper than that. High-quality content is assumed to correlate to:
- higher traffic
- increased brand awareness
- improved visibility
And THOSE things are assumed to correlate to:
- higher sales
- increased revenue
Whew.
I’m not saying that there isn’t a relationship between those three layers of assumptions: there absolutely is. But what everyone takes away from the science is that longer blog posts = higher sales, and that specific correlation hasn’t actually been tested. These are the most robust correlations between blog posts and business sales that I have been able to find:
- SEO leads have a 14.6% close rate, while outbound leads have a 1.7% close rate. But that stat doesn’t link to a source.
- Another statement without a data source is that a high quality blog generates 67% more B2B leads and 88% more B2C leads per month.
Again, I believe that to be true (of course I do – I’m in content marketing!) but it’s also riddled with assumptions. And what’s bothering me is that everyone is asking for extremely long posts without that much to say.
There is No Perfect Word Count for a Blog Post
Asking how long a blog post should be is like asking how long a story should be. A story should be as long as necessary and no longer.
That being said, a blog post needs to be at least 300 words, to make it worth a click. Shorter than that, put it in a tweet.
Blog posts are one aspect of your content strategy, and your entire strategy is to provide value to your audience. The length of the post should be driven by the value you can provide. An ideal blog content strategy would be multi-layered, and would include:
Shorter posts (~400-600 words):
- breaking news in your industry
- update about a product or feature
- tips for how to do a thing your audience wants to do
- events and announcements
Medium-length posts (~700-1000 words):
- how-tos and tutorials
- comparisons and feature explanations
- origins and process descriptions
Long posts (~1300+ words):
- guide to
- history of
- manuals
And there may be occasions, depending on your audience, to produce a very long post. For example, if you are explaining something highly technical and complex, it’s often useful to go back and explain the history and terminology before exploring the issue at hand.
All these types of posts (and all the appropriate word lengths), should be part of your content strategy and part of the overall value of your blog to your customer.
Tips for Long Blog Posts
All that being said, conventional wisdom is also true: people don’t actually want to read a lot. 43% of people skim blog posts. Here are some ways to keep a long post engaging:
- Let people know up front that it’s going to be long. If the headline is “The Ultimate Guide to X”, I know it’s going to be a long post. I love how Cognitive SEO put a TLDR at the top of this post. Let people know how much you expect them to invest.
- Break up your wall of text. Use images (infographics are awesome) and bullet points and headlines to facilitate skimming.
- Be smart about your sourcing. Neil Patel writes epic blog posts, and they are really valuable, but I do feel like he often goes on and on proving his point. If you make a point and back it up with 2-3 sources, citations, or examples, you can move on. If someone doesn’t believe you by then, they never will, and you’re wasting the time of people who are thinking: “I get it. Now what?”
- Provide a “now what”. Of course a blog post should always end with a CTA, but even if you don’t (I almost never do), you should try to bring it back to what is relevant and actionable for the audience. Don’t just make your point and leave it there; if someone has spent a lot of time reading your content, then give them something they can take away from it.
Let Go of Surrogate Markers
Content marketers love metrics because they are fascinating and offer a way to demonstrate the concrete value of the rather intangible process of relationship-building. But don’t confuse word length with quality, or quality posts with increased revenue. Be attentive to your audience and let their needs guide your strategy. Focus on authenticity and value, and less on one specific statistic or another.