Demonstrating the impact of marketing efforts across all channels is one of marketing’s biggest challenges today.
It’s not easy to do. But with the right approach and technology, you can have confidence in the data, which will help you communicate impact to your leaders and make smarter, data-driven decisions moving forward.
In this post, I’ll use web content as an example to demonstrate how we can measure performance.
Here are the first two steps:
- Report on high-level revenue contribution
- Measure content performance by type
Let’s get started.
Reporting on High-Level Revenue Contribution
For this example, let’s define web content as blog posts and ebooks. If someone views a blog post or ebook as their first interaction with your brand, then fills out a form to download an ebook, the right measurement and attribution technology will allow you to assign revenue credit to those assets.
So the first step is to report the monthly sum of revenue attributed to all your content assets. This is a great KPI because you can track it over time, set goals, and drive the growth of content’s influence on revenue.
While measuring revenue contribution is a good baseline, in order to grow the number, we need to know more.
The Data Needed to Optimize
For improving and growing content revenue performance, we need to know what types of content generate the most revenue, how varied the revenue performance is, and which content type performs best along a standardized measure.
By understanding these details, managers can decide where to invest their budget, where to do testing, and identify areas worth improving. Eventually, you need to know which assets are the best-performing, so you can promote it more, learn about content topics most relevant to buyers, and so on.
Here’s the revenue performance by type, based on the dummy data above.
Now that there’s a baseline data for these assets, the next step is to compare the performance of these content types in a standard way. Cumulative revenue numbers could mask the fact that some content types are more efficient at influencing revenue than others, with certain types generating higher levels of revenue simply because of volume. So to avoid any bias in the data, let’s calculate revenue per asset.
To do so, for each content type, divide total revenue by the number of assets for each type. This provides a better idea of revenue per asset for each type of content. See the example results below. Notice how the top performing content types changed when the data is standardized and revenue per asset rates and identified. This indicates that on a per asset basis, there might be a scenario where more revenue is generated from a webinar, even though cumulatively, there was more revenue generated with definitive guides.
Even with this level of detail, I still caution you to take these results with a grain of salt. Let me explain.
Setting Expectations for Future Revenue from Content
Before taking the next step to optimize and invest in content, it’s a good idea to understand expected values for revenue.
In the above example, even though there’s a generated rate (revenue per asset), it’s still not clear what actual revenue will be in advance of creating the next round of content assets. It’s possible that there are plenty of assets but only a small portion of those are getting revenue credit. What does this mean for expectations?
It means you need to investigate further. Here are few things you should look at next:
Rate of Success: For every asset, what portion actually generates revenue? This is defined as the number of assets getting revenue credit compared to the number of assets that don’t. Knowing these success rates are a helpful place to start to troubleshoot content types with low performance.
Revenue Variance: How spread out are the revenue numbers? If some assets generate a lot of revenue, and many assets have low or zero revenue attribution, then your content return is highly varied and predicting future revenue might be more difficult—which begs the question, what is content performance correlated with? We’ll cover this question below. So a great exercise is to compare the revenue distributions of different content types. The shape of the distribution will tell you how varied the returns are, and what revenue number content assets are centered around (i.e. the average).
Time Dependence: It takes time for assets to accumulate revenue credit. The length of time since the publish date should be taken into account when looking at revenue data. Assets may have accumulated more revenue due to being older. For example, the length of time it takes to rank in organic search could be a factor. Revenue attribution amount may be correlated with time. Understand this time dependence for your content assets will help you predict when you can expect a certain amount of revenue from your assets.
Other correlating factors: The amount of money used for content promotion can vary by asset, which can impact the likelihood for success. The location of the asset on your website and how difficult it is to navigate and find are also correlating factors to keep in mind.
Via Marketo blog