The Subject Of Content Intelligence Spans Many Areas; When It Is Described To Marketers, How Do We Ensure That It’s Not Just Understandable, But Also Relevant?
You’re dead on. Virtually every definition you see of content intelligence poses as many questions as it answers because it spans so many topics.
Fundamentally, the only way for a marketer to understand whether a specific content marketing strategy works is to start with the data. Take your historical data and compare it with competitive data to identify leading metrics. Then you produce content. Then you use data to analyze the effectiveness of that content and work from there.
Take the untimely death of Prince. There were hundreds, if not thousands of articles from news and entertainment outlets, blogs, Twitter, Facebook, and lots more. If you searched online about Prince in the days following his death, Google would present maybe four or fivesearch results at the top of the page, then load up other Prince information from Wikipedia,Instagram, and elsewhere.
On the other hand Google News served up a lot more in-the-now information, but there you would see that the marquee main headline changed hands many times throughout the day, as Google decided one piece of content was more interesting than another. Examining that data later, you might decide that in a case like a famous person’s death, you shouldn’t go with a straight obit (where competition is fiercer) but dust off recollections of the individual.
In addition to simply analyzing historical data to find out what worked in the past, it’s also critical to understand the human side to predict what may work. The better the tool set, the higher the probability you’ll have a winning outcome.
What Particular Marketing Problem(s) Most Call Out For a Content-Intelligence Approach?
Content intelligence solves many problems. One solution we are exploring is analyzing content performance from multiple viewpoints. Historically, marketers have taken a very binary perspective (e.g., “How much traffic is my content receiving?”). Content intelligence must use a variety of methods and metrics to assess performance related to social media, search, paid advertising, desktop vs. mobile, and many other factors. At its core, content intelligence attempts to solve several problems: What to write? Why write it? How does it impact users?
What Are Some First Steps For Marketers?
No. 1 – data. No. 2 – data. No. 3 – more data. Marketers should have a lot of information about how content connects with their users. When I worked at eBay, we used data to understand what gaps we might have in our content output. We found that people want lots of information on products. To fill the gap, we created guides that attract informational searchers — things like how to buy a blender or what’s the best pair of running shoes. Starting with your own data very often can determine where you’re missing a key part of the puzzle.
Some other areas to consider: Examine performance metrics to identify dead content that needs to be removed from your website. Or compare what content works best for your competitors, and whether yours is superior or needs more work. The best content intelligence combines data and business strategy.
How About Examples of Intermediate/Advanced Steps?
One of The Famous data projects is called content recall. Content recall uses the same fundamental practices of brand recall. Essentially, you’re examining users’ feelings and expectations related to your content. For example, you may study all the tweets about a particular piece of content. Within those tweets there is significant redundant terminology (e.g., words with similar meaning or words all related to a particular user intent). By aggregating specific terms, you can better understand what users are expecting. Say you track 10 branded, 40 product-related, 50 commercial-related, and 10 competitive terms. By grouping the terms, you may see that while commercial and product expectations are important to users, a particular page should use more competitive terms to attract buyers from your competition. (Without aggregating terms, you may not have detected that pattern.) This can dictate content edits and changes to improve the overall performance.
What problems/risks Marketers Face Embarking on Some of the Projects Mentioned?
One of the biggest mistakes We’ve seen content marketers make is thinking all traffic is created equal and that generating traffic for your content is the primary objective. Simply looking at changes in traffic leaves you with little understanding of user expectations or the emotional response your content generates. Understanding user intent is increasingly important to overall success — if for no other reason than it’s a major focus for Google’s algorithmic efforts. If We reach a search page that has a shopping cart, one would assume my intent is transactional — that is, We are buying something instead of window shopping. Google has identified similar patterns and associates your content with the expectations of the user. Reaching this level of understanding is far more difficult than the old days of simply understanding the need for content quality.
What Role Does Technology play? Is it a Supporting role or a Primary role?
Technology helps content owners and managers scale their efforts and become more efficient. Yet adopting new technology is challenging for content creators because it interrupts the content production flow. Technology can transform the content intelligence and content production environment, but it must not disrupt content creation in the name of delivering insights to the user.
In summarizing his views, Jordan offers these five content intelligence tips to help your content marketing:
- Look beyond internal data.
- Understand what the user’s emotional expectations might be.
- Use multiple data points to interpret the effectiveness of content (i.e., social data, search data, internal data).
- Consider removing content before adding more content.
- Leverage the content-recall model to understand what users are expecting in your content.