When it comes to measuring content marketing, we’re all treading water. Marketing analytics for content marketing is changing, and, we hope, improving essentially every day. Unfortunately, content marketers are often struggling with their measuring tools, but, as we all know, we have to measure and prove and measure some more in order to keep and continue a content marketing endeavor.
In order to measure how content marketing is improving ROI, companies can use several methods that vary in complexity and accuracy. Attribution modeling is a method that companies use in order to understand which of their marketing channels is the most effective, or which turns out the most conversions, and should therefore receive the most investment. There are two major categories within attribution modeling—single-touch attribution and multi-touch attribution.
1. Single-Touch Attribution
Single-touch attribution is a method by which companies credit a conversion to the channel that either was the first interaction (first touch) a customer had with the company or was the last interaction (last touch) before the customer became a conversion. This method is easy to implement and speaks to the first steps of a customer’s journey, but it is not very accurate in terms of attributing proportionate value to marketing channels.
As a first step to measuring your content marketing ROI, single attribution is an obvious choice because it is easy to understand and explain to others. Unique customers’ actions on your website and with your marketing channels are tracked and recorded. Each unique customer who eventually turns into a lead has a first touch with the company’s marketing channels, and each customer has a last touch before it is evident that they are going to become your customer. Therefore, companies simply attribute that customer’s value to either the first channel they interacted with or the last. Simple, but with some obvious holes.
For example, if some channel is never the first interaction or the last interaction, then it may be cut from the budget because it is not converting customers. However, if customers are using that channel to get to know your company’s product or service, or to interact with your marketers, then that channel is valuable in building a relationship with the customer. That value, however, is not indicated when using single attribution. As stated above, although this method has holes, it could be a first start for measuring your content marketing ROI.
2. Multi-Touch Attribution
Multi-touch attribution has taken many forms as companies recognize the shortcomings of single-touch attribution and try to give credit to middle-touch channels. MarketingSherpa shares the data from a survey of several hundred companies, finding that companies using multi-touch attribution models increase ROI by 22% year over year. Some of the following models have been adopted with their respective pros and cons.
→ Linear Attribution Model:
This model records the entire progress of the customer from first-touch to last-touch and everything in between. It then attributes the value of that customer equally across every channel with which the customer came in contact. This method is perhaps a step ahead of the single-touch attribution because it recognizes that there are many channels that contribute to a prospect becoming a customer, but it also utilizes a generic and, albeit easy, oversimplified valuation of marketing channels.
A customer may have interacted far more closely and frequently with the email marketing you employed, however, if they interacted once with social media, then that channel is receiving the same credit as the email marketing. If a company is investigating and adjusting their marketing budget, this method will tell them very little about which marketing channels are the most effective. However, it is a good indication that your marketing channels are being utilized, and if you are tracking a customer’s journey, it can lead to insights about the common path that prospects follow through your marketing channels before they become customers.
→ Time Decay Attribution Model:
The time decay model’s name explains itself—the interactions that occur closer to the time of the actual conversion receive more credit than the interactions at the beginning of the buyer journey. This model, like the others, functions on an assumption about the value of certain points of the buyer journey. The assumption here is that as buyers get closer to a decision, the channels that they interact with are more important or valuable.
This assumption may be closer to the truth than the linear attribution model, but it is still an assumption. The advantage to this model is that its assumption is based off of the idea of the buyer journey in which the potential customer wants more and more from your marketing channels as they get closer to a purchasing decision. However, the assumption still stands as an ideal model in which every customer follows the same path, but customers don’t follow the same path. Furthermore, it may be that the first marketing channels a customer comes in contact with are very valuable to them because they were the first to introduce the product or service and lead the customers to a decision. Again, the model rides on an assumption, but since it attempts to follow the buyer’s journey, it has more potential for accurate value attribution.
→ Position-Based Attribution Model:
This last model’s assumption is somewhere in between single and multi-touch models. With position-based attribution, 40% credit for a conversion is given to both the first and last touch channels, and 20% credit is divided amongst all of the channels that fall in between the first and last interactions. This model wants to recognize the entirety of the buyer journey, but it still assumes that the most valuable channels to the customer are the first and the last. The assignment of value given with the above percentages is, in many ways, arbitrary.
All these baseline models function on assumptions about the customers. There are options to create custom attribution models, and as marketing technologies become more sophisticated, companies should be able to see the attribution model vary with every customer that is closed. The attribution of value will follow more closely the content consumption of each prospect in order to give a company more actionable information about their marketing channels.
The current attribution models may give information about where to increase the budget in terms of channel, but they are not actionable on a content level. When the value attribution data can inform content pieces more closely, content marketers will not simply pump more content into an apparently successful channel, they will use attribution data to create the correct content for the correct customer and put it on the correct channel.
What are the pros and cons you see in attribution modeling? What tools would you recommend for a more precise measurement of content marketing?