The Basics of Content Analytics for Publishers
Given magazine publishers’ surfeit of competition from free, high-traffic websites and services that disaggregate content (think Facebook, Google News, and Flipboard), it’s time to refine methods for determining what your content is doing for you, and for making sure you’re squeezing as much revenue out of it as possible. That’s where content analytics comes in.
Your editorial, web, and marketing teams have surely been using analytics to a degree for some time. Google Analytics is the most usual suspect. But not enough publishers have taken full advantage of analytics to understand how their products perform and contribute to fundamental business objectives.
Content analytics employ a combination of processes and metrics that enables anyone with a content-heavy website to examine what each page is doing for business—whether that means ad revenues, subscription sales, newsletter signups, or any of the myriad other ways that publishers monetize their content online.
Content analytics is an invaluable strategic tool. It allows publishers a very granular peek at what different types and individual pieces of content do for business and also offers a tool to gather meaningful insight for better business decision-making.
Here we’ll provide a quick summary of some of the metrics and processes that publishers can put to practical use. (Content analytics is a big topic, so take this as a primer more than the definitive tome on the topic.)
There’s No Way Around Google
With 1.2 trillion searches a year, Google is by far the biggest player in search. Fortunately, it also provides a tool allowing anyone with a website to fine-tune their strategy: Google Analytics.
After a little bit of low-level coding, Google Analytics (GA) will provide a deluge of data on every aspect of your audience: operating system, time on individual pages, time on site, revenue, conversion rates, and many others. It’s perhaps the essential content analytics tool, but due to the flexibility and depth of its offerings, diving in can be a daunting task.
Define Your Objectives
In order to take advantage of any content analytics tools, first ponder how you’re making money, and how you want to make money in the future. Here are a few potential use cases:
- Is your content meant to encourage users to sign up for an email newsletter, but you aren’t getting as many signups as you want?
- Are you looking to increase ad revenue by increasing site traffic?
- Are subscriptions your most lucrative revenue stream, but less than 1% of your site’s visitors are willing to open their wallets?
In each of the preceding cases, different metrics will apply. Let’s take a look at the metrics that might be useful in each case, and why they warrant your attention.
What it means: GA collects detailed sitewide information on visitors, notably age, gender, location, their browser’s language settings, and, crucially, affinity categories (which will tell you if a visitor to your site has visited a site about sports or cooking lately).
What it’s useful for: SEO consultant Stephanie Chang advises that publishers start by paying close attention to the demographic tools in GA, because these tools deliver a more accurate and comprehensive picture of readers’ age, location, gender, OS, and other interests than most publishers otherwise have access to.
Chang suggests gearing content to your readership’s other interests. For example, if GA reveals that significant numbers of a publisher’s visitors have also visited cooking sites within the past week, there may be opportunities to publish food-related content that’s still within the publisher’s subject area. Geographical data can help target content, too: If you find out that your readership tilts toward the Midwest, a few features on Chicago wouldn’t be a bad idea.
All three business objectives mentioned above rely on growing core audience, so everyone needs to pay close attention to their visitors’ profiles. Demographic data is more useful for commissioning new work than it is analyzing old content, though, because of sample size issues.
What it means: Unique visitors (or “uniques”) means how many individuals have seen a given webpage (or have visited any page in a given site).
What it’s useful for: The statistic can be skewed a hair by individuals clearing their caches, which deletes the cookies that Google uses to monitor user behavior, but it reflects the best available information on the overall size of a given post’s audience.
Sites that sell display advertising through programs like Adsense rely heavily on unique visitor statistics, so case 2 above will find this metric utterly indispensible. It’s also essential for 1 and 3, though, because it reflects the total number of people a given page or site has available to convert if nothing else changes.
What it means: Conversion rates mark the percentage of visitors who meet a particular goal, usually making a purchase. There are also goal-based conversion rates that will tell you how many people (for example) take an action like submitting their email address.
What it’s useful for: Sitewide conversion rates vary considerably depending on what you’re trying to get people to do—if your business is selling a $12,000 annual subscription to a high-end, niche information service, your conversion rate will be lower than if you’re trying to induce visitors to subscribe to a lower cost, populist publication.
75% of established ecommerce websites have a conversion rate of less than 5.3%; content-based websites typically have lower conversion rates. Cases 1 and 3 will want to pay close attention to this metric, and if articles on a particular topic tend to have lower conversion rates than other topics, it’s a clear signal that the content isn’t working. Conversion rate is less important for strongly ad-reliant sites like case 2, but absolutely vital for cases 1 and 3.
What it means: When someone shows up to a webpage, visits one page, and leaves, GA registers them as a bounce. Bounce rates are generally better when they’re lower — it means that you’ve successfully kept people interested. SEO blogger Avinash Kaushik once remarked (somewhat misleadingly, as we’ll see), bounce rate is a shorthand for “I came, I puked, I left.”
What it’s useful for: If one of your pieces of content has higher traffic than average but also a higher bounce rate or lower conversion rate than average, alarm bells should go off. Something on this page is attracting viewership, but once people arrive, they aren’t engaging. Pages with high bounce rates might have user interface problems, or you may not have provided the tools necessary for visitors to find something else they like as much as they did the content that initially attracted them.
Occasionally you’ll end up with a high bounce-rate page if someone who gets more traffic than you do cites a page of yours as an information source. This isn’t always the end of the world—indeed, if you’re selling display ads, it will make your page more valuable regardless. (Although details of its algorithms are always well-hidden, Google reportedly takes little notice of a page’s bounce rate when calculating the value of a display ad on it.)
Bounce rate is, by definition, incomplete: It doesn’t suggest a course of action. It can mean you’ve done a good job of attracting people! It can be a good opportunity signal, though, and as such will be most useful for cases 1 and 3.
Average Order Value
What it means: Average order value, or AOV for short, measures the average dollar spend in each session where any money gets spent at all.
What it’s useful for: AOV is vitally important in e-commerce, and also useful for publishers with merchandising pages, or ones trying to increase the number of subscriptions or events sold per purchase. AOV takes discounts into account, so if you’re offering discounted second or third subscriptions and you want to know how that’s going, or if you’re wondering how a discount code or promotion has affected revenue, AOV is the go-to metric. It’s best for case 3.
The Importance of Goal Setting
The metrics above are the basics, but extracting useful information out of them takes a little more planning and effort. Fortunately, GA also includes powerful tools that show exactly what content is most effective in driving visitors to particular locations in your site. These analytical tools are GA’s scalpels — they allow publishers detailed data on routes to purchase, traffic, and user preferences.
GA allows users to set triggers — called goals — that trip whenever event X happens. That can be an ad click, but it can also trigger when someone spends more than five minutes on a page, or when a visitor buys a subscription, reaches the confirmation page in your merch’s checkout process, or signs up for an email newsletter.
Most commonly, you’ll want to input a URL that customers reach just after they’ve done something that generates revenue, directly or indirectly — perhaps a purchase confirmation page, or a page that says “thank you for trusting us with your email address.” Whichever it is, goals detail all users’ routes to a post-transaction page, telling you what routes users take to get there most often.
Goals are easy to set up and critical to content analytics, because without goals, GA can’t tell a conversion-based business like examples 1 and 3 whether they’re succeeding by their own standards, and what’s helping them to get there.
Content Grouping: An Essential Publishing Tool
Ever wonder whether posts about fish drive users to submit their email addresses more often than posts about cats? Do people who read articles written in Comic Sans buy a subscription less often than readers who see Helvetica? Do you want to find out whether author A has more fans than author B (without adding their pages’ uniques up by hand)?
Content grouping in GA allows you to taxonomize your content, and look at how each group performs compared to others. Content grouping is perhaps the best way for publishers to take advantage of GA’s offerings. It can look at any of the above metrics, not by individual article, but by whatever category you choose.
These categories can include date, author, or as many topic areas as you like; you can also tease out different sales pages (if you have more than one on your site at the same time) to see which offer’s wording converts most often, or to see what the average order value of people taking a particular route to purchase is.
Content groupings are organized three deep, so that each grouping will have a subgroup and each subgroup a set of pages. A sample setup might be organized like this:
- Article 1
- Article 2
- Article 1
- Article 3
- Article 4
- Article 5
- Article 6
In this example, note that Anne and Bob worked together on article 1 — it’s a useful feature of content grouping that each webpage can belong to more than one subgroup.
The drawback to content grouping is the advance planning it takes. If you end up wanting to add a new author or a new subject area to your site’s remit in six months—say, your fishing magazine decides to start running articles on deer hunting—your results for your established categories will be skewed if you don’t add a new category to reflect the addition to your taxonomy, so thinking ahead is important.
Content grouping also doesn’t tell you how well all your legacy articles about fish or cats drew an audience or converted, either—it’s not retroactive. You only start accumulating data once you set the category, so using content grouping effectively means waiting for adequate sample size, and it works best when you have some idea of what kinds of content you might publish down the road.
How It All Fits Together
Using these tools, you can answer most questions you might have about your content. Want to know if people who read Carlita’s articles are more inclined to buy a subscription than other authors’ readership? Set up the subscription confirmation page as a goal, set each of your authors as a content group, and after a few thousand visits for each author, you’ll have your answer.
Related story: The Starting Point for Harnessing Magazine Audience Data