Every business professional wants to get the most out of their money, but how do you follow the trail? Here’s how attribution modeling helps digital marketers achieve the best ROI through quantitative measurements and credits based on touch points and conversions.
Every business wants to get the highest return out of their investments. And while you can easily try to read and understand the data that shows what is happening to each active investment, sometimes the trail is difficult to follow and there isn’t enough data available to determine the best sources of success.
Understanding the consumer journey helps a business understand how, where and why a conversion was made.
Education is important for growth. And learning how to measure growth is the first step to success. Once areas of growth are identified, a business may focus on those areas to achieve the highest return on investment (ROI) for their marketing endeavors. The best way to show a return on investment is by creating attribution models that will help you in the long run by providing a clear image of your customer and identifying which initiatives are really paying off.
· Assigns credit to touch points in conversions
· Follows set of rules
· Uses multiple model types
Attribution modeling uses a rule or a set of rules to determine what point of the sale gets the credit. There are multiple attribution models that assign credit to certain touch points like user experience, specific campaigns, email marketing and other marketing initiatives. Finding the best practices will maximize your efforts to find new customers and retain old ones by understanding the user journey and what efforts contributed to the final sale.
It is important to try different models because many times it is not about the destination but rather the journey of figuring out what made a sale possible.
· Gives credit to just one touchpoint
· Most commonly seen is in the first interaction and last interaction
· Ignores any other interaction customer may have taken
A very popular model is the single-touch attribution that assigns 100% of the credit to one of the touchpoints in the campaign. For example, if the customer clicked the first paid advertisement and still went through multiples channels, such as email or social media, all the credit would be given to that paid advertisement. This can be reversed as the credit can be given to the very last touch point the customer clicked before the sale while ignoring the other touchpoints.
This type of modeling does not just have to include the first or last interactions done by the customer. It can also assign credit to any element in the campaign. Google Analytics provides users with an example that contributes all credit to the last AdWords click before the conversion.
The main issue with this model is that it ignores the consumer journey from start to completion (conversion). When determining what digital marketing strategies would work out best, no one simply puts all their eggs in one basket. They know that consumers behave differently, and that decision was probably made due to the combination of several factors. However, finding the touch points to focus on is a process that takes time and lots of data. This model may be an easy way to explain to a client what worked, but in order to truly capitalize on your ROI, a different model may be better.
· Takes all touchpoints into consideration
· Models may spread credit equally, based on time or position
· Credit can be assigned through custom models
A multi-touch attribution model allows you to understand the full picture of what, where and why a consumer made the purchase. By using this model, you can discover what touch points in your campaigns carried the most weight. For those that consistently generate a higher conversion, a business should leverage a higher investment. For efforts that are shown to be unimportant in the consumer journey, these areas should be re-evaluated for their importance, and reduced where possible to better allocate budget to more productive campaigns. A great starting point into multi-touch attribution models is by using a linear model.
· Great start to multi-touch attribution
· Distributes credit equally
· Difficult to determine what is working best
With this model, a business measures purchases based on an interconnected process, understanding that a consumer goes through multiple channels before that final click. If you have a campaign with four elements, a linear model will give 25% credit to each element in the campaign. This is an easy way to understand the data but is lacking the critical input that other models may provide.
If every element in the campaign automatically receives the same amount of credit, then is becomes very difficult to determine what is working and what is not. This becomes a grave issue when you start figuring out how much you want to invest into each effort. Without knowing what worked in the past, it is nearly impossible to predict the future. That is where a time decay model comes in.
· Assigns most credit to touchpoint that led to conversion
· Touchpoints that led up to final touchpoint are given less credit
· Does not show what touchpoint was organically found by customer
· May does not give enough credit to influential touchpoints
A time decay model focuses just on multiple factors over a course of time that led to the final conversion. This takes every element into account and measures by a window of time – giving the most credit to the element in the campaign that ultimately led to the purchase. It then also shows what elements did not perform well over that period of time, so a business can understand which efforts may be a waste of time.
A major drawback to this model is that it may not include the original element if it was prior to the measured timeframe, and therefore does not show how influential it may have been. For example, if a video is a part of a marketing campaign, it may take time before it gets to the eyes of your consumer. If a consumer saw your video and it greatly influenced them two months before the purchase, it will do poorly against email that the consumer saw two hours before the purchase. Even though the video may have been the major influence of the buying decision, the email may be the only measure of success in a time decay model. To avoid this, a position based model is needed.
· Assigns significant credit to first and last touchpoints
· Other touchpoints receive same credit
· Some touchpoints may not need equal treatment
A position based model offers the idea of a linear model, while also incorporating the element of time. The benefit of this model is that it’s focused on the start of the campaign as much as the final conversion. Other elements within the campaign will receive the same amount of credit. Where previous models only focused on the start and finale, this model of attribution gives equal measure to both start and finish while still recognizing the consumer journey to purchasing.
This model is a great starting point to leverage the most in-depth analytics for a higher ROI.
There is no silver bullet in marketing. No magic wand that will get people to click a few times and end up with a purchase. It is important to recognize this and create a model that will suit your needs. Focus on what has worked, experiment with what may work and do not let anything stop you from trying. Measure your results through the attribution model that fits best with your campaign and objectives, and incorporate that newly gained knowledge into developing your next strategy for the highest ROI.