For all the content out there online, there’s a surprising lack of decent information about content analytics. Perhaps digital marketing departments believe that when it comes to understanding how their content is performing, ignorance truly is bliss?
Well, I’ve spent enough time looking at performance metrics in Google Analytics to empathize with those departments.
That being said, questions about return on investment (ROI) on content marketing spend will inevitably arise, whether you have a developed analytics program in place or not.
Gartner, a major advisory firm that researches enterprise IT strategy, conducts an annual survey to track how much of the marketing budget goes to content marketing. This year, Gartner’s survey shows content creation making up the second narrowest sliver of the budget among the budget categories surveyed:
With such a narrow slice of the pie devoted to content marketing, proving ROI on your efforts is crucial to ensuring that your department gets the funding it needs to succeed.
Dashboards are currently the most effective way for marketers to document ROI for the people who make the decisions that count.
We’ll identify the major challenges in creating content analytics dashboards and offer solutions based on our department’s experience with building dashboards using a range of analytical tools.
Here’s what we’ll cover:
- Linking Online and In-Store Purchases to Content
- Linking Content Production Metrics to Traffic/Engagement and Revenue
Dashboards Address Shortage of Tools for Analyzing Content Performance
Content analytics is an emerging discipline. Right now, end-to-end platforms for measuring content performance don’t exist.
Marketers are thus in the tough position of having to measure performance without dedicated tools.
In “How to Measure and Prove the Business Value of Your Content Marketing Program” (full content available to Gartner clients), Jake Sorofman and Martin Kihn note that “Surging interest in content marketing raises the inevitable question of ROI. The answer? It requires a long-term commitment to yield returns, but the path to ROI should include a comprehensive metrics program.”
There are two basic reasons why measuring content performance is absolutely essential:
- You must either convince executives of your content’s value or perish.
- You need to understand what’s working and what isn’t in order to continue improving content performance.
Building a dashboard is a crucial step in developing a metrics program. Chingho Wu, the architect of our content analytics dashboard here at Gartner, explains that “dashboards are a great way to repeat an analysis so that you don’t have to start from scratch with raw data every single time.”
Let’s move on to looking at how dashboards solve some of the core challenges in content analytics.
Integrating Data Sources in Dashboards Key to Attributing Revenue to Content
The key to proving ROI for content marketing efforts lies in integrating the performance data on how visitors engage with your content with the metrics related to conversions and sales. Dashboards solve a number of attribution challenges, including:
Linking Online and In-Store Purchases to Content
Sorofman and Kihn make a crucial point about linking revenue impact to content performance:
“As a rule of thumb, the closer you are to the point of sale—both physically and temporally—the easier it will be to show revenue impact. For example, a considered purchase like an automobile, which is sold through a dealer network, is more difficult to trace to a content marketing campaign than, say, an online fashion retailer.”
Jake Sorofman and Martin Kihn, Gartner Analysts
Content marketers working for businesses that rely heavily on e-commerce channels will find the task of documenting revenue impact to be a cakewalk compared to the challenges other marketing departments face.
This is because web analytics platforms used to collect data about visitor engagement with content (e.g., Google Analytics) can also be used to track e-commerce conversions. You thus have one source for both kinds of data.
Linking in-store purchases to online content, however, is a task that’s bigger than your marketing department.
In “Understand Attribution and Marketing Mix Modeling” (full content available to Gartner clients), Kihn notes that “if some sales are in-store and some are online, you cannot do attribution without having a method of identifying in-store shoppers (e.g., a loyalty program) and tying them to digital advertising (e.g., by onboarding).”
Linking Content Production Metrics to Traffic/Engagement and Revenue
Determining ROI on content also involves tracking content production metrics. Otherwise, you won’t be able to make strategic decisions about how the time invested in a piece of content correlates with the traffic, engagement and revenue it generates.
Our department simply tracks production metrics manually in spreadsheets, though they’re also captured automatically by some workflow tools and content management systems.
This data will also need to be integrated with web analytics data (from Google, social media etc.), and web analytics data from different sources will itself need to be integrated.
Storing and Visualizing Historical Data Without a Data Warehouse
In a typical BI scenario, data would be extracted from all of these sources and fed into a data warehouse for storage and reporting.
For the purpose of content analytics, however, data warehousing isn’t necessary. Lizzy Foo Kune and Christi Eubanks note in the “Market Guide for Marketing Dashboards” (full content available to Gartner clients) that content analytics doesn’t “require you to manage a data warehouse (most marketing data—visits, clickstreams, impressions, comments—already resides in big data cloud environments)”.
Instead of opting for a data warehouse, most marketing departments will use the following strategies:
- Exporting data from sources to CSV files, and aggregating and integrating data in spreadsheets.
- Connecting directly to data sources with self-service BI tools, configuring data source integrations within self-service tools, and using dashboarding features of self-service tools for reporting.
While Google Analytics offers native features for building dashboards, the functionality is too limited to be useful for content analytics (if you’re interested in why, see the first part of this series).
Content IDs and Lookup Tables Solve Integration Headaches in Dashboards
While data warehouses aren’t typically used in content analytics, our marketing department harnesses a crucial concept in relational database management to simplify content analytics, namely key-value pairs.
In a key-value pair, a unique identifier (the key) is paired with some kind of value in a database. For example, a customer ID can be associated with the amount of a customer’s purchase.
The reason key-value pairs are used in databases is that the same key can be used to link values that appear in various tables.
For instance, a repeat customer will have numerous purchase amounts that may be tracked in separate databases, but the “customer ID” key can be used to link all these values together for analysis.
John Leslie, research manager at Software Advice, introduced the practice of assigning a reference ID to every piece of content we produce. This can simply be done in a spreadsheet:
Once the ID has been assigned, it can be paired with various values (engagement metrics, production metrics, revenue metrics etc.) to view the same piece of content from different angles.
The other advantage of utilizing key-value pairs is simple data integration. Even if you’re just integrating and aggregating data with spreadsheets, you can use the ID in conjunction with the VLOOKUP function and lookup tables to automate the tedious work of rolling up values about a piece of content from various spreadsheets into a unified dashboard.
Options for Pre-Built Dashboards
If you’re looking to simplify data cleaning and integration for your dashboard—which can be a beastly chore—you have two options:
1. A generic BI system that includes self-service data preparation functionality (which we describe in more detail in our report on the subject), as well as a dashboard authoring interface. If you plan to explore data sources other than the ones we list in the chart, and if other business units in your organization will be using the platform, you should go with a generic platform. Our site lists reviews of software vendors that sell self-service BI tools.
2. A platform specifically for designing dashboards. Dundas BI, Domo and iDashboards are all strong solutions for marketing departments, but we also have an extensive collection of reviews of dashboard platforms.