Content is King, that is basically the one line you hear everywhere these days. At the same time the number of companies that have started their own blog has increased dramatically. This also shows by the number of blog posts that are published every day. It is 2.7 million!
Looking at this number you might think how can it even be possible that you will manage to attract viewers to your particular blog if they have such a variety to chose from?
Well, I found that actually most blog posts out there are just not looking appealing and are not optimized. That leaves a huge potential for you to compete with. So by being better optimized than your competition and putting some extra thought and effort in what you are doing, you will be able to get that traffic you are after.
When you are starting out blogging either for yourself or for your company the first thing your are wondering is where to even get started and how are you able to create a content roadmap over the next couple of months that contains already those interesting articles people want to read?
To make life easier for you and help you get an idea how you can even approach the topic of planning and preparing content, I have started out and analyzed 1260 blog post headlines of some of the best blogs in the marketing niche.
Alright, I am excited. So let’s dive in!
Top marketing blogs in the spotlight
To keep the data relevant and ensure that my findings are of significance I have looked only at blogs that have the same type of authority, backlinking profile and content production frequency.
To get this data for all the blogs I have used the Moz open site explorer. If you don’t know that tool yet, have a look at it, it’s one of my favorite SEO tools!
Here are the blogs I have selected:
The personal blog of Neil Patel, one of the top content marketers on the planet. On his blog Neil discusses a variety of different online marketing topics, giving especially newbies in digital marketing a great overview and orientation where to get started. Started not too long ago Neil produces content with an extremely high level of speed and quality, doing this consistently he managed to push his personal blog amongst the top ranking marketing blogs.
Not just their software but also their content is awesome. If you are an SEO or planning to become one this is definitely a blog you need to save to your bookmarks. Started from a SEO blog Moz now is host to a vibrant community of digital marketers that share and publish their knowledge and findings directly on the Moz blog.
Specialized on creative SEO tactics that drive a ton of links, this blog hosted by Brian Dean, is also a must if you are starting out or being stuck with link building. The blog itself doesn’t have as much content like any of the other blogs I have analyzed but each piece is a detailed guide that gives you a ton of useful information.
Kissmetrics is a software company that has created a very successful analytics software built for marketers focused on getting actionable tasks out of their data. Naturally the core focus of their blog is analytics in web marketing. If you are looking for good and useful information around the topic of analytics I can recommend checking out their blog.
Semrush another software company that has started out with a strong SEO focus but nowadays is developing a full stack online marketing software that you should definitely check out, is running another successful marketing blog. Although you can find a variety of different marketing topics you will find their core focus to be SEO.
Now that I have introduced all the different blogs let’s have a look at how they are comparing on domain authority and root domain backlink count. In such quick comparisons I like to look at those to indicators the most as I feel they give a good understanding and feel how well a website is optimized and performing out there.
Here is how they are holding up:
As you can see all five blogs are competing on a similar level. On top of that all of them are discussing similar topics which makes them great for my analysis.
Good, now that I have my blogs let’s look at what I am trying to accomplish with this headline analysis.
How to write headlines that drive social engagement?
What is the reason why we are actually writing blog posts and are producing content in general? Well, that is kind of obvious right? To generate more sales and revenue!
But how can you actually measure the success of your blog posts? Or let alone measure the success of someone else’s blog posts?
If you are looking at your own blog posts you have many ways how to actually measure success. Starting from time on site, over the touchpoint in the customer journey, to the actual hard number of leads that have been generated by this blog post.
But when we are looking at the blog posts of someone else we don’t have access to those KPIs. We could be looking at backlink count of one blog post or keyword ranking volume, but I find those a bit vague indicators of a blog post’s success.
The most obvious data I am also able to extract easily was the number of shares and comments different articles get.
Now, when you are new to blogging and you are just wondering and starting out with selecting the topics you want write about it is of course quite hard to get any kind of similar attention, traction and engagement on your articles as the top blogs are having.
Also something that is quite easily forgotten, top blogs that are either making a lot of money or are part of a business with established resources do have bigger resources to produce content on a higher or more frequent level. So what is it that we can actually look at that will help you get started with your content planning and give you at least some kind of reassurance that what you are doing can be interesting to your audience?
The probably most obvious part of a blog post: The headline.
I think that by looking at the headlines alone we will be perfectly able to get a better feeling and understanding of what type of content and also what characteristics of a headline will set you up for more social engagement on your blog posts.
Scraping website data with the help of importXML - step by step
This is my favorite part in the process. It’s data collection time. In order to be able to analyze all those 1260 headlines across those 5 different blogs we first need to extract them.
You have a couple of ways to do that. If you are skilled enough you can try building your own script that automates this process for you. Another option would be going for a software solution. There are a couple of tools out there that help you extract content from different websites, such as import.io or mozenda.
If you are neither skilled enough in scripting nor you want to spend that kind of money for a software you can go with a simple spreadsheet formula. (That is also what I did).
The formula is called importXML. If you haven’t heard of it yet and you want to dive deeper than what I am going to show you here, you should read this in depth guide by distilled.
The importXML formula allows you to extract structured data from a website into a Google spreadsheet. Once you know and understand the basics it couldn’t be simpler (and more fun) to run around scraping websites!
Get your Free ImportXML Guide
Learn more about ImportXML and start scraping the web like a pro! I teach you the basics in my free importXML guide.
Now to get all the different headlines and shares this is what I did:
Step 1: Open up an empty Google spreadsheet.
Step 2: Open the blog you want to scrape in your browser. In my example I use Neil Patel’s blog.
Step 3: Inspect the page by opening the developer tools. If you are using Google Chrome (which I highly recommend), open the inspection modus by clicking right on your mouse and then click on ‘Inspect’.
Or use the keyboard shortcuts: On Windows and Linux: Ctrl + Shift + J. On Mac: Cmd + Option + J.
Once you did that the a sidebar with developer tools will open on your screen and you should see something like this:
Now you are able to see the html code of the web page.
Step 4: Navigate to the elements you want to extract and prepare your spreadsheet formula.
Now you need to click on the little cursor icon on top of the developer tools, doing that allows you to quickly navigate to the html snippet of the part of the web page you want to extract.
Then hover with your mouse over the headline and click. Once you did that you will see that the html code in which the headline is, is marked in blue as in the image above.
Now you can see that the headline is in the style format of ‘h2’ and in the ‘class’ called ‘title’. This is the information we need for our importXML formula.
Now head over to your spreadsheet and prepare the formula. To give you a quick idea how the importXML formula is structured, here is a quick example:
And here is what you get. As you can see, after typing in this simple formula you will already find quite a big list of headlines extracted into your spreadsheet.
Now you can keep on repeating this process with whatever you like. In my case I also extracted the comments and shares. After doing this for all the blogs I wanted to have a look at here is what I ended up with:
But that 1260 times!
Great! Now that I have all the data I want the next step is to dive deep and analyze all those headlines.
Dissecting blog post headlines to find something epic
First things first. Before we get deeper into what type of headlines we are dealing with here let’s take a quick look at a couple of numbers.
Looking at those numbers we can already see quite some interesting things sticking out. The average headline consists out of 10 words and has 62 characters. This is quite interesting, as of the one of the more recent Google Updates (2016), the SERP snippets got much more space.
This indicates that only long tailed headlines are becoming more and more interesting but it can also help with preparation for the increase of voice search, as our searches will be longer and more human language like.
Now that we are already focusing on the word count of a headline, let’s have a look how many headlines have what count of words spread out over the entire sample size.
To do this I have built three groups of word counts. The first one goes from 3 to 8 words, the second one from 9 to 14 words and the last segment from 15 to 20 words. Here is what I found:
The majority of headlines lies between 9 to 14 words, which is quite interesting. This word count range should mostly be just in the visible part of the entire SERP snippet, which makes it fully readable at the first glance.
But now the question is, does the word count have any effect on social shares and comments? Alright, let’s have a look at that.
Looking at that graph we can see a relatively big spark of average comments and shares with headlines that have more than 15 words in them. While we see quite drop of average shares with headlines that have between 9 and 14 words. The comment section itself seems to be buzzing with short and long worded headlines.
Now we have assessed a couple of basic numbers but let’s try to go deeper and see what type of headlines those are. By looking at the type of headline we can already tell what type of blog post can be expected behind that headline.
I have considered the following types and characteristics of headlines:
- Headlines that start with a number -> Most likely list articles
- Headlines that start with ‘how to’Headlines that start with ‘why’
- Headlines that contain an indication that the article is an analysis
- Headlines that contain the word ‘guide’Headlines that contain a question
- Headlines that contain the word ‘you’
- Headlines that contain the words ‘me’ or ‘I’here...
First let’s look again at the distribution of all those different types of headlines across the entire data sample.
Interestingly, you can see that list articles (indicated by headlines that start with a number) and how to headlines are the most prominent. While at the same time, headlines contain the word ‘you’ quite frequently and make up a majority of all headlines.
On the other side, analysis, why headlines and headlines referring to oneself are strongly in the minority.
Now let’s have a look at how the different headline types perform in our social engagement check.
The first thing that sparks out is the very strong outbreak of average comments and average shares in the analysis headline type.
We can see that this type of headline performs well above average. Another strong amount of engagement can be found in ‘how to’ headlines as well as guides!
If we are looking at the difference between talking about ‘you’ or ‘me’ in the headline we can see a very strong increase in engagement on self centered headlines. However the average number of shares seems to drop in self centered headlines while the number of comments goes up. A part missing here is of course an analysis of the type of comment.
How to write the perfect headline?
The outcome of this exercise is to get a step closer to answer the question how you can actually write an awesome headline for your blog posts that drives engagement.
So let’s put together what we have just learnt. Looking at our data we can state that an engaging headline contains more words than 14, tells a story about what you did (contains me or I) and is an analysis!
I suggest you go ahead and test this for yourself. Write a story, write longer headlines and share all the interesting insights you have found on the web or in the world!
Of course writing and preparing an entire content calendar and content plan is a huge task but by giving you a clearer picture of what type of headlines perform well, you should already be able to create an interesting mix of different article types.
So when you are setting out to create your plan for the next articles you are going to write, don’t forget to write an amazing headline that not only generates clicks but also drives engagement!
I had lot’s of fun with this analysis and I am looking forward to all your thoughts! And if you are tempted to play around with the importXML function let me know of all your cool findings!