Various studies over the years have examined the relationship between content relevancy and behavior. Almost everyone would agree that content must be relevant. But what is relevance? According to Wikipedia: "Relevance describes how pertinent, connected, or applicable something is to a given matter." A thing is relevant if it serves as a means to a given purpose.
In the context of this discussion, the purpose of content is to positively influence customer or employee behavior, such as increasing purchase frequency, purchase velocity (time to purchase), likelihood to recommend, productivity, etc.
When we ask marketers and others how they measure content relevancy, we often hear, "We base it on response rate." If the response rate meets the target, then we assume the content is relevant; if response doesn't meet the target, we assume it's not relevant.
Clearly there is a relationship between relevance and response. Intuitively we believe that the more relevant the content, the higher the response will be. But measuring response rate is not the best measure of relevancy. Many factors can affect response rate, such as time of year, personalization, and incentives. Also, in today's multichannel environment, we want to account for responses or interactions beyond what we might typically measure, such as click-throughs or downloads.
So, what is the best way to measure relevancy?
The best-practice approaches for measuring relevancy are many, and many of them are complex and require modeling. For example, information diagrams are an excellent tool. But marketers, who are usually spread thin, need a simpler approach.
The following three steps provide a way to tie interaction (behavior) with content. It's critical that you have a good inventory of all your content and a way to define and count interactions, because once you do, you'll be able to create a measure of relevancy.
The process and equation include the following:
- Count every single piece of content you created this week (new Web content, emails, articles, tweets, etc.). We'll call this C.
- Count the collective number of interactions (opens, click-throughs, downloads, likes, mentions, etc.) for all of your content this week from the intended target (you'll need to have clear definitions for interactions and a way to only include intended targets in your count). We'll call this I.
- Divide total interactions by total content created to determine Relevancy: R = I/C