2015-11-23

Posted by Paddy_Moogan

Content marketing is hard.

The problem is that the process looks easy. You brainstorm some ideas, choose one that you like, design and build it, do some outreach and you get traffic, links and social shares. Job done.

It's a bit like link building, where someone may say, "Just build great content and the links will come."

Unfortunately, it's very rarely that straightforward.

Yes, sometimes you can get lucky and something will fly with little effort. But anyone that says that content marketing is easy has probably never done it over and over again. This is one of the reasons that I really liked this post last week by Simon Penson, because he admitted that he'd failed many times before getting it right. Simon pointed out that the plan he shared just increases the possibility of success — it doesn't guarantee it.

In this post, I'm going to share our process for putting together a content marketing campaign. It doesn't guarantee success either, but I'm positive that it puts us in a much better position than if we didn't have a process at all. We're always trying to improve this process, and it's never going to be 100% perfect. With each campaign we do, there's usually something we add or take away which also reflects the ever-changing nature of our industry. It's also hard to manufacture and force that "ah-ha" moment, when you get a great insight into something and which then generates a great idea. Although this slide deck by Mark Johnstone helps make sense of how we get those moments in an excellent way.

One thing to point out before we get into the meat of this post is that it's not just about "big" content. Our role as digital marketers (many of us with an SEO background) goes much wider than content that is purely designed to generate links and social shares.

A content strategy needs to include more than just one type of content, and for most clients at Aira, we do multiple types of content based on their objectives. But that's a post for another day, because today I'm going to talk about our process in the context of content that is designed to generate links and social shares, driving traffic as a result.

There are five broad steps in the process:

Research and idea generation

Idea validation

Production

Promotion

Conversion

Step 1 – Research and idea generation

It's easy to dive straight into brainstorming and idea generation, and sometimes, that can work. However, I'd always recommend a period of research into an industry prior to this so that you can get a feel for what's been done before, what has worked, and what hasn't. This can mean that you go into a brainstorming section far better equipped to generate ideas that may work.

One thing to point out at this stage is that you shouldn't put yourself under pressure to come up with a completely new idea. It's great if you can, but the reality is that it's unlikely that something has never, ever been done before in some form or another. So you shouldn't put this pressure on yourself. The following quote is an apt one:

"An idea is nothing more or less than a new combination of old elements."

This is from the book A Technique for Producing Ideas, published in 1939 by James Webb Young. It's a short — but excellent — read, and I'd highly recommend it.

I think you'd agree that over 75 years later, this quote is even more true now!

Therefore, a big part of the thinking behind our process is looking for inspiration in what others have done and asking ourselves if we can do it a little bit better or a bit differently. I'm certainly not saying you shouldn't try to come up with brand new ideas, but don't let an idea fall by the wayside just because it has been done before.

I'm going to frame the rest of this step by saying something: The most successful* content that you find will come down to at least one of three things:

The story – If something has a strong story or hook behind it, it's more likely to grab attention and be picked up by mainstream news websites and publishers.

The data – Often tied into the story but not always explicitly, if you have unique data or data that has been sliced/interpreted differently, it can be of more interest to someone.

The production – Sometimes a piece of content may just look visually stunning, and that is enough to generate links and shares.

There is one more which I want to point out, but it's been deliberately left out of the list above. The one other thing that can make a piece of content successful is an existing audience to market that audience to. A prime example in our industry here is Moz, who has a very large, existing audience. This means that this very blog post is more likely to get links and shares than it would if I published it on my own blog, which has a very small audience.

This is important to remember because, when looking at your competitors and the success of their content, the numbers may be skewed a bit because of the audience they have. I'll show you how to offset this below.

* Successful, in the context of this post, means generating links and social shares that drive quality traffic. Success can mean many things to different businesses, so I just wanted to remind you of this.

Find your content competitors

The first key step is to research your content competitors, and it's very important to recognize the difference between your product/service competitors and your content competitors. Let's look at an example.

Let's say you're a travel website. You may be trying to rank for keywords such as "flights to New York" or "holiday apartments in Italy" because you provide those things. You'll have competitors who are trying to rank for the same kind of keywords and of course, you should take a look at what they're up to. However, there is a whole other section of websites who don't compete for these type of keywords, but whom you can learn a lot from when it comes to content. In this example, those websites are travel bloggers and publishers who have travel sections. They produce the exact kind of content that generates links, social shares and traffic — exactly what we're trying to do with our own websites.

Examples in the travel world could be Nomadic Matt and Jayne Gorman, who both produce great content that generates links and social shares. If I run a travel website and I wanted to learn what content can work well in my industry, I'd definitely take a closer look at these kind of people for inspiration. They may even be people who I could partner with on content ideas, but that's a bit outside the scope of this post.

It's pretty simple to find our content competitors. The quickest way is to think of a few non-commercial keywords. Examples related to travel may be "guide to New York City" or "planning a trip to Italy," which are likely to show search results that include publishers/blogs as opposed to direct competitors. You can also use the keyword search function in Buzzsumo to do these kind of searches:



The results will show you content that contains this keyword, ordered by social shares

If you're not familiar with Buzzsumo and would like to learn the basics, take a look at this post that I wrote on Moz a few months ago, which talks about this and shows how we use the API for one of our internal tools at Aira.

Finding stories and topics

Once you've found a handful of content competitors (we try and find at least 4–5), it's time to start taking a closer look at what they're doing. Buzzsumo allows us to do this quickly and easily; all we need to do is run a domain search and use an advanced search operator to search multiple domains at once:



You just need to put OR in between the domains that you want to do research on. The resulting search looks like this:



As you can see, Nomadic Matt is dominating the results, which is likely to be because of a combination of writing great content and having a larger audience than the other websites we searched for. This is a good example of where we may actually want to temporarily remove him from the list, so that we see a more diverse set of results. However, you can also just download a CSV from Buzzsumo and filter his domain out if you wish.

The important step here is to scan the list of results to try and find patterns and trends. In the screenshot below, I can immediately see a pattern:

Some of the best-performing posts are lists. We can see this quickly by noticing the numbers at the start of the title. Going a bit further down, I notice another pattern:

Lots of these posts are "How to"-style posts, which are clearly popular with his audience due to them featuring high on the list of results.

It doesn't take long to start noticing these patterns. Make a note of them and we'll come back to how we're going to use them later.

Another way to find patterns is to analyze the titles in bulk. We can do this by doing an export from Buzzsumo so that we get a list of titles:

You can then copy and paste these titles into a word cloud generator tool, such as Wordle, and get something like this:

You'll need to remove common words, such as the website names and domains, but the result above is basically a summary of the words that get the most shares — which is really handy to know in bulk. Again, make a note of these kind of themes.

I know what you may be thinking at this point: What about links? Buzzsumo can give you backlink data, but you have to click on each individual result to get it. This is fine for a small number of articles, but we're trying to do bulk analysis. So instead, we're going to export the results to a CSV and then upload those results into URL Profiler, which can fetch link metrics for us in bulk.

These are the settings you want:

You can select your choice of Mozscape, Majestic, or Ahrefs data, or all three — it's up to you. The point is that we need to know how many links our content competitors are generating to their individual content pieces. The results will then look something like this when you export the results to Excel:

Once you've got this, you can do some pivot table magic to make the data easier to consume. Here are the settings that you need:

Then you'll end up with a graph that looks something like this (you can, of course, make it prettier!):

As we can see, A Luxury Travel Blog is leading the way in terms of generating links to their content, so they're worthy of a closer look. The beauty of this process is that Buzzsumo does a pretty good job of excluding the homepage from their results, so the results are showing links just to the content they produce. From here, we can do a deeper dive into their links using Mozscape, Majestic, or Ahrefs — whichever you prefer.

Before moving on, I want to mention a few other tools that we use in this step of the process. Epic Beat is very similar to Buzzsumo in that you can enter a domain or keyword to find what content is being shared the most. Combining the results from Epic Beat and Buzzsumo can give you lots of information on what is working for competitors in your industry:

Another cool tool — which is more for qualitative analysis rather than quantitive — is Brandtale, which curates digital content/advertising campaigns on large publishers. Sticking with our travel example, I can browse their travel section to find brands who are running campaigns:

I can drill into any of these and see what these brands are doing and if I can learn anything. Trust me, running content campaigns like this on large publishers, such as National Geographic or the New York Times, is expensive. A lot of work will have gone into them, which means they're worth looking at.

Finding data sources

Our next step is to try and find data sources that could lead to us creating a piece of content or a story that can be pitched to publications. I'd highly recommend Statista for this, which is a growing resource of statistics and facts. Sticking with our travel example, here is a snapshot of the kind of data it has available with a simple search:

If Statista doesn't have what you need, a few simple searches on Google will often yield good results. Just remember to do a bit of due diligence on where the data comes from and make sure that it's as sound as possible.

Failing that, can you get your own data? There are many organizations and services out there who will gather data on your behalf. Yes, you have to pay for them, but if you think the data can help you generate links and shares for your website, then it could be worth the investment. Here are a few options:

Google Consumer Surveys

One Poll

You Gov

Toluna

Usurv

Survata

Some of these can be expensive to use, so I recommend using something like Google Consumer Surveys to poll a small sample of people. Then, if the data is looking promising, run the full survey.

Finding visual content

The final piece of this research is finding visual content which has done well and seeing if we can do better. Like finding data, don't overthink this, and start with a few simple searches. Google Images is always a good place to start with keywords such as this:

You can get more specific based on the website you're working with, but what we're looking to do here is scan the results quickly and see if anything stands out to us:

If we find any that look particularly good or interesting, we take a closer look and ask ourselves the question, "Can we do it better?" While some visuals may look okay and performed well, there are often ways to improve on something, such as:

Making the core story or headline more obvious

Making it interactive to make the key messages easier to consume

Making the design cleaner so that key messages are communicated better

There are any number of ways a good designer can make an existing idea much better, and as we discussed earlier, making something beautiful can sometimes be enough to make it successful.

Once we've found something we think we can do better, the next question is how successful was it? One way to do this is to use this Google Chrome extension to automatically do a Google Image reverse search to see how many other websites have used that visual. If the answer is a fair few, then you know that a better version is likely to be of interest to a number of websites.

Putting all of this research together

That was a lot to go through! But trust me, it's worth it. The next step is to take all of this information and put it into a brainstorm session brief for your team. When it comes to brainstorming, many people will say "all ideas are good ideas" — but this simply isn't true.

A brief is very important here, because your team needs to walk into that session with the right information and context. If they don't, then the majority of ideas that are generated may not actually be usable — which isn't a very good use of time.

To make this easier, I've put together a Google Doc template which you're welcome to download and make a copy of. You can find it here.

Step 2 – Idea validation

The more I work on content marketing campaigns, the more I value this step in the process. You can think that you have a great idea, but how do you know for sure? The fact is that you can never predict this 100%, but you can increase the possibility by using a framework to validate an idea.

The key thing here is not the frameworks that I talk about the below, but to make sure you use some kind of framework so that you can consistently and fairly assess the quality of your ideas.

One of the frameworks I'd recommend, which some of you may have heard of, is from Made to Stick by Chip and Dan Heath. I'm not going to go into too much detail here simply because lots has already been written on the topic, including this post from Distilled and this more recent post by Hannah Smith, which references the framework. There is also this summary of the book, which talks about the key takeaways and what the principles of Made to Stick are.

In summary, the book outlines six principles which, through their research, the authors feel make an idea stick in our minds.

Simplicity – An idea needs to be easy for us to comprehend quickly. A good way to test this is to write the headline and see if you can communicate the idea within the restrains of a headline( i.e. you only have a short sentence).

Unexpectedness – While the idea doesn't have to be 100% brand new, there needs to be something new or unexpected about it.

Concreteness – This can often be mixed up with simplicity, but is subtly different. Concreteness is all about the idea not allowing room for ambiguity or misinterpretation of what you're trying to say.

Credibility – The basis of the idea needs to be credible. This can be via credible data, a credible (expert) author or a credible company behind the idea.

Emotion – If an idea provokes an emotional response, we're much more likely to remember it.

Story – We touched upon this earlier and goes back to when we were children. We were told stories and all of us can remember certain ones. We're used to the structure of a story and how it peaks our interest.

The key here isn't the framework itself, although that is very important. The key is the ability for you and your team to give each other valuable, constructive feedback on an idea. It's often easy to just say "I don't like that idea" or "That idea won't work," which, even if you're right, isn't the most useful feedback to receive. With a good framework, someone can reference it in their feedback. So if you're using the framework above, you can say "I don't think the idea is simple" or "It's not concrete enough" — this is far more useful feedback to hear and it may mean that an idea simply needs tweaking rather than dumping completely.

As mentioned earlier, this isn't the only framework you can use. Another one goes back to what we talked about earlier:

Do we have a story or an interesting hook?

Do we have unique, interesting data?

Can we make the idea look beautiful?

Answering yes to at least one of these questions can increase the chances of your idea being a success.

Step 3 – Production

I'm not a designer or a developer, so I'm not going to tell you how to design or develop a piece of content. But there are some things that we've learned (sometimes the hard way) when it comes to producing a piece of content.

Function over form

The first thing I want to share here which is important is to remember function over form. Never, ever say "I want an infographic" or "I want a video" or "I want an interactive piece of content." You should focus on getting the right idea first, then ask what the best way to present that idea is. If it turns out that an infographic is the best way to present your idea, then great. But don't start with the form; start with the idea and see where it takes you.

This may help reduce the number of terrible infographics on the web which, unfortunately, our industry is at least partly responsible for!

Mobile-first design

There are stats upon stats showing the growth of mobile, so I'm not going to tell you those again. If you want to do some digging, I'd highly recommend the work and analysis from Ben Evans, who specializes in this area.

In relation to content, what we need to remember is that content discovery is becoming more and more mobile-centric. We typically think of content discovery as someone browsing on their laptop/desktop machines and clicking through from a blog, Twitter, or Facebook. In reality, though, it actually looks more like this:

When someone clicks on a link like this on their mobile device, they expect the content they land on to work perfectly on their device. If it doesn't, the user is not likely to enjoy or engage with the content, let along share it or link to it from somewhere.

This deck from Vicke Cheung does a great job of showing the importance of designing for mobile, along with practical tips for doing this:

Ten Lessons in Designing Content for Mobile from Vicke Cheung

Another key thing here is to let designers design. Try not to restrict them by providing a brief that tells them 100% how something needs to be done. Give them the goals of the piece and some guidelines, then let them design. Of course give them feedback along the way, but try not to be too prescriptive.

Go-live checklist

One of the lessons we've learned the hard way is that in your excitement to get something live, you can forget some of the basics. A few common things that need to be thought about, but are easily forgotten, can be:

Social/Open Graph tagging

Analytics code

Responsive testing

To help with this, here is another Google Doc which you can download and use which contains a few things to remember:

While the things on this list seem basic, it can be very easy to forget!

Step 4 – Promotion

Here is one of the key takeaways: Spend just as much time on promotion as you do on the production. It's so easy to get caught up on design, development, and the idea itself, you can end up spending most of your time on producing it and not nearly enough time on promoting it.

There are three different types of promotion we work on at Aira. These differ by client, but ideally, we spend time on all of the following:

A combination of all three can help ensure that your content reaches as many people as possible. I used to rely solely on organic content promotion via traditional link building outreach/digital PR, but this may not be enough and ignores some useful techniques.

Paid promotion

Paying to promote your content can be very useful in generating traffic to a piece of content, which in turn, can also help generate social shares and sometimes links. Larry Kim goes into detail on this in his post over on Search Engine Land. The basic principle is that you can use paid promotion to get your content in front of writers, bloggers, journalists, and influencers.

There are a few options for how you can do this. Firstly, to reach a wide audience, you can use platforms such as Taboola or Outbrain. These can work well for reaching a very big audience, but targeting options for specific demographics on these platforms is still rather limited.

Wil Reynolds ran an experiment using these (and other) platforms, which is definitely worth looking at:

The $10,000 Paid Content + Paid Linking Test that is 100% Google Safe from Wil Reynolds

Our experience with these particular platforms is very mixed, with it working well for some clients but sending very untargeted traffic for others. So we'd advise starting with a small budget and assessing the quality of traffic before spending too much.

Other options are more regular social channels such as:

Facebook

Twitter

LinkedIn

Pinterest

The one I want to focus on is Facebook, where the targeting options are almost scary. But they're useful to us nonetheless. You can do things such as specifically targeting journalists using options such as:

You can put whatever list you'd like in here, but I'm sure you get the idea!

You can also go one step further in targeting people by uploading their email address into the custom audience feature of Facebook:

It's straightforward here to upload your outreach list; if Facebook can find a match for the email addresses, you can advertise directly to those people. If you'd like to go into more detail on this, take a look at this post I wrote last year.

Earned promotion

This is likely to be more familiar to most of us because this section covers traditional link building outreach and digital PR. Essentially, we need to find a list of influencers and contact them in order to promote our content to them. This sounds simple, but can often be the trickiest part of the process... because it's here that you may find out that you don't actually have a great idea! This is why the idea validation step is so important — because it reduces the chances of promotion going wrong.

I've written multiple times about finding outreach prospects before, so I won't repeat everything here. But I will point out my favourite techniques for doing this.

Finding existing lists of prospects

I honestly start every single piece of link building research with these kind of searches:

Simply switch out [INDUSTRY] for your own industry and you'll find more than enough prospects to keep you busy!

Finding mainstream publications and journalists

Here, we're trying to find high-level outreach targets who write for national newspapers and mainstream publications. The value of these can be huge, because many websites like this have the ability to send a LOT of traffic to your website.

Here are a few tools (mostly paid, unfortunately) that you can use for this kind of research:

Gorkana

Cision

Muckrack

A News Tip

You can of course do manual research as well, but these tools can help speed things up a bit.

Owned promotion

This one will depend heavily on what your client already has, but essentially we're talking about using their own channels, such as:

Social channels like Facebook, Twitter, Instagram, LinkedIn, or Pinterest

Their existing blog

Their email marketing list

This may sound easy, but I've worked with some companies where the social team sits separately to the SEO/content team — which can make it harder to get them to work together! If you can bridge this gap, though, it's a pretty easy win to get eyeballs on your content.

Step 5 – Conversion and tracking

So here we are, at the final step of our process, and I want to be really honest about this bit. It can be quite hard to convert a visitor to a piece of content that is designed for links and social shares. These kind of content pieces are often not designed to "sell" to the visitor, so getting them to click across to the main website or a product page (let alone getting them to buy something) is difficult. There are exceptions; this piece from Bellroy is one that comes to mind which is informational but very related to their product:

Generally though, this is difficult to pull off. So what can we do instead?

Micro-conversions

If we can't convert someone into a buyer, what else can we do? One thing we've done for some clients is to try and capture a visitor's email address so that we can then target them on Facebook or via email marketing. Or it could be any other number of things, such as:

Commenting on a piece of content (so you also get their email address)

Sharing a piece of content

Spending a certain amount of time on your content

Build retargeting lists

If someone visits a piece of content, you can build a retargeting list and then advertise to them in the future. There are two ways you can do this:

Advertise your products and services to try and encourage clicks and future purchases

Advertise your future content pieces — this can work very well if you're working on a content series, i.e. a series of blog posts that all tie together

Build retargeting lists based on interactions with a page

This is a post for another day, but it is possible to go more targeted when building your retargeting lists, by building them based on how someone interacts with your content. For example, you can fire Facebook retargeting pixels when someone clicks on a certain link or when someone selects a certain option (if your content is interactive). This means that you can build lists that are very specific, and you can cater your advertising based on the interactions that users have carried out.

To wrap up

So that's about it for today's post! These are the five broad steps that we take for a content marketing campaign, and while we're always iterating on them and improving them, they have increased our chances of success — which is what this is all about. You'll never guarantee success, but whether you use the process above or your own, you certainly should utilize a process to enhance your chances.

I'd love to hear your feedback in the comments!

Sign up for The Moz Top 10, a semimonthly mailer updating you on the top ten hottest pieces of SEO news, tips, and rad links uncovered by the Moz team. Think of it as your exclusive digest of stuff you don't have time to hunt down but want to read!

Show more