Our CEO, Charlie, spent the first 5 years of his digital career on the production side of things. Globe-trotting celebrity video shoots, RED camera, the whole shebang. He even co-created one of Red Bull’s most successful video concepts of all time: Raw 100.
During this time, budgets for shoots could range anywhere from $5K to $500K. That’s just one shoot. Taking into account production of dozens of photo and video assets plus digital distribution, a B2C organization’s budget for content can easily top $10 Million.
With so much budget on the line, the days of creative intuition being the sole driver of content direction are beginning to come to a close. That’s not to say that creativity doesn’t have its place. But...
We started RightMetric because we realized that a marketers creative intuition is most powerful when backed by data.
Sometimes that data comes from content analytics tools or a research firm like RightMetric, but it can also come from unexpected sources… Here’s Charlie:
"Back in my video production days, I was working as a video editor at Arc'teryx. During this time the Customer Service Manager came to our team and said that he had an idea for a new video about teaching people how to properly wash their GORE-TEX jackets.
Me, being the young whippersnapper that I was, thought that his idea was boring. I was used to editing videos with rock climbers summiting peaks and skiers shredding powder.
He mentioned that when he looked at the data from his team, this was one of the topics that came up the most and he figured that if there was a video that existed that his team could send to customers, they’d be able to save time and serve more customers on a given day.
We ended up making the video and I remember after we published it on YouTube that it didn’t perform very well and I definitely thought to myself “I told you so...”
Almost 10 years later and that video has the 2nd most all-time video views of any Arc’teryx video sitting at just under half a million as we’re writing this article."
This video has been a smash hit for Arc’teryx across the whole digital landscape.
- It’s generated the 2nd most all-time video views for the brand on YouTube, ever
- It’s still #1 on the Google SERP for “how to wash gore-tex”
- It serves a clear customer need
- It positions Arc’teryx as a thought leader
- Production costs were only 25% of the typical action-packed videos featuring professional athletes going on expeditions as there are extras associated with travel, accommodation, equipment, etc
Brute force (a huge budget) isn’t the only way to create powerful content. A little data— whether it be from customer service, social analytics, or market research — can go a long way.
A lot of RightMetric clients approach us with the task of informing or optimizing their content strategy using data, with the ultimate goal of taking the existing budget and using it to reach the desired audience more efficiently and impactfully. We thought we’d share how we apply marketing intelligence to that challenge.
Where Should We Focus?
This is typically the first question we get from the content strategists we work with. We know that for content to be successful we need to have the distribution channels in mind from the very early planning stages. “Build it and they will come” doesn’t work anymore.
The specific questions often look like this:
- We have a limited distribution budget, what are the 2 or 3 places we should target?
- Where does our audience actually spend time online?
- Where are our competitors allocating budget to distribute content?
- What works well on our Instagram is totally flopping on YouTube and Pinterest. How should we deal with that?
Every company has a lot of assumptions about the answers to these questions, some are right and many are often wrong. As always, we recommend balancing intuition with data.
“Where should we focus?” really is the #1 question in marketing, and also one of the toughest to answer. Luckily, we wrote an article about how to solve for that. Give that a read to cover step 1.
The Framework For Categorizing Content
One of the biggest challenges that marketers face when creating a content strategy is just how to define and categorize everything. The sheer variety of content types is such that it almost defies organization.
Why categorize content? To help content creators to make things that succeed you need to give them guidance that they’ll actually remember and take action on.
The best way we’ve found to do that is to organize content into categories, and ensure that those categories are ultra simple and clear.
For example, how would you define and compare these two pieces of content: a 2 hour Twitch stream of a Super Mario 64 speedrun versus an Instagram carousel of a fitness model providing photo and video tutorials on kettlebell exercises? At first glance, these two pieces of content seem so different as to be almost incomparable. Apples and oranges.
Here’s how we break it down:
- Platform: Where on the internet the content will live ex. YouTube
- Subject: What the content is about at a high level ex. Yoga poses
- Theme: How the Subject is presented ex. Listicle, tutorial, cheat sheet, etc.
- Format: The type of media ex. Video, photo, etc.
- [Optional] Sub-Format: Sometimes needed for clarity on platforms where multiple formats can go together i.e. A video inside a carousel on Instagram.
We’ve found this to be a robust framework that can be applied to just about any piece of content on the internet.
Let’s apply it to the examples above.
The Super Mario 64 Speedrun
- Platform: Twitch
- Subject(s): Super Mario 64
- Theme: Speedrun
- Format: Livestream
- [Optional] Sub-Format: 2 Hour Length
The Fitness Model Kettlebell Tutorial
- Platform: Instagram
- Subject(s): Kettlebell Exercises
- Theme: Step-by-step tutorial
- Format: Carousel
- [Optional] Sub-Format: 30 second videos & illustrated instructions
Works pretty well, right?
At RightMetric, we call each of these categories a “bucket”. Once you define some high-performing content buckets it becomes much easier to decide what to create.
The next step is performing an analysis to determine which content buckets are working well to reach a specific audience on a specific platform.
Choosing What to Analyze
Once you know which platforms to focus on to reach your audience you’re already halfway there.
The approach you take to determining high-performing content buckets actually shouldn’t vary that much regardless of what data you can get your hands on, but what will vary is the accuracy of your results.
The wider you can cast the net of your analysis, the more accurate your results will be.
From narrow to broad:
- Your own data (very limited, but how most companies do it)
- Manual observation of competitors (labour intensive, but it works okay)
- Content performance tools like Phlanx, BuzzSumo, or SEMrush (relatively low cost, but there’s still a lot of manual labour involved to get accurate-ish results)
- A Content Intelligence service like RightMetric (We’ll cast the net very wide and do it all for you, worth mentioning!)
Regardless of how broad you go, it’s best to analyze content from accounts that you think your audience follows, or that focus on the topic you’d like to focus on. For somewhat robust results, we’d recommend looking at 10 accounts at the very minimum.
Once you decide what dataset to analyze, here’s the steps to take.
Finding The Right Buckets
For RightMetric clients, we perform this analysis on a monthly basis to ensure that we’re capturing shifts in content preferences while they’re still relevant.
- For each account you’re analyzing, calculate the engagement rate (ER) of each piece of content the account has posted in the past month.
- Sort the accounts by size of following, since an ER that’s considered good will vary depending on the size of the account.
- Set an ER benchmark by taking the average ER of all posts in the account cohort, or all posts in the category if you have that data available.
- Here’s the manual part. Start categorizing every post that has an ER higher than the benchmark into categories (buckets). Why is this manual? Simply because there are subtleties that software alone cannot pick up on. This includes things like defining whether a post is aspirational, or if it’s a story of failure, or if it’s educational, and so on.
- Once you’ve bucketed all the posts, take an ER average for each bucket. The highest ERs are your winners.
Okay! The analysis is done. That’s the hard part… or is it?
Content creation is… creative, and there’s a longstanding rift in marketing between “the creatives” and “the data people”.
If you’re a data person reading this, then you know that the final step is finding a way to communicate your findings to the content creators in a way that makes it more likely that all this valuable information will actually get put into action.
In our experience, it’s all about presenting findings that are specific, clear, and believable.
How to Present Content Intelligence So it Actually Gets Used
As we said, for content intelligence to get actioned it needs to be:
The example below is one of the ways that we present content intelligence to RightMetric Insights subscribers.
Why it’s Clear: It highlights the content buckets that are performing well, without clutter. Using colour strategically is always a good idea for this kind of thing.
Why it’s Specific: Examples are provided. The reader doesn’t need to imagine what a “shopping challenge” or “vanity upgrade” is. They can click through to the post to understand what it’s about.
Why it’s believable: The graph you’re looking at here is part of an executive summary, but later in the report there’s several pages of data and analysis to back up these findings. It’s important to always “show your work” so that skeptical power users can be satisfied. If they don’t believe your findings, they won’t action them.
The Process For Surfacing Content Intelligence
The very worst thing that can happen to any research or analytics project is for it to get created and then thrown in the metaphorical desk drawer. Doing this kind of thing requires a significant investment of time and money. Here’s how we make sure it gets used:
Update it Regularly: A lot of marketers are used to flying by the seat of their pants. It’s critical to perform analyses and share results with your team on a regular cadence so that after a while, people get used to relying on the valuable information that’s being provided. It gets built into the culture.
Surface it Regularly: Culture is one piece, but it takes time to get there. The first step is building content intelligence into your workflows. Red Bull’s global social team has a weekly standing meeting called Party Time. They use it to do a post mortem performance analysis on the past week’s content and work that into the plan for next week. Carve out regular time for your team to review content intelligence and build it into their ever-evolving strategy. If you already have regular strategy meetings then you're halfway there. Just make sure that reviewing and actioning content intelligence is an agenda item.
Share it with everyone who needs it: Content touches almost every digital channel, but most marketers don’t think about it that way. Once you’ve put your content intelligence findings into a format that is simple and fast to consume, share it with everyone on your marketing team. It won’t take up a ton of their time and they might use it to drive value that falls outside the typical scope of content marketing. For example, your performance marketing team might adapt their Facebook ad creative using a high-performing content bucket that you identified for that specific platform, or your email marketing team could create lead-nurturing content that has already been proven to resonate strongly with your exact audience.
The Impact of Data-Driven Content
What’s your organization’s annual budget for content production and distribution?
Even a simple content intelligence program can improve results by 20% on average, while increasing the chances of publishing an evergreen smash hit like the Arc’teryx GORE-TEX video that drives views, reach, engagement, and traffic for your brand for years to come.