Setting up the proper attribution is highly essential for any marketing analytics. With a proper marketing attribution model, marketers can measure the ROI of their channels and campaigns, and optimize them to get better results.
Yet, with so many different marketing attribution models available, choosing and implementing a model might be confusing, especially for marketers that are not familiar with the subject. So, here we will discuss five of the most popular marketing attribution models, as well as their advantages and disadvantages.
Without further ado, let us begin with the first one.
The first-touch attribution model appoints the whole credit to the first action/touch point in the journey. So, this model mainly focuses on the lead generation aspect. Let’s use an example to better illustrate this model: assume we are an SAAS business selling cloud software, and a prospect signed up for a free-trial demo after reading a content, where they stumbled upon the content from organic Google search. Then, after the free-trial period, they finally converted after getting a nurture email.
So, with the first-touch attribution model, the credit goes to the very first touch-point, which in this case, organic search. The main idea behind this is fairly simple: unless the prospect is aware about your business, there won’t be any sale ever. This model can be ideal for businesses where the conversion is just a few steps away from the first interaction, like a lot of B2B businesses.
Pros: very simple to set up, and can be very effective for marketing strategies that are focusing mainly on awareness and lead generation. Also very simple to track.
Cons: quite obvious, this attribution model neglects the importance of any channels and campaign beyond the very first touch-point. So, it won’t give you a very accurate picture of your overall marketing activities.
In essence, last-touch attribution model is just the opposite of the first-touch model: it gives full credit to the very last touch-point before conversion. It is a very popular attribution model, and also the default attribution model in Google Analytics. So, as you can see, the main focus of this model is the action that triggered conversion, while completely ignoring the steps before that point.
Let’s use the same example as above to better illustrate this model. With a same case, the last-touch attribution model will give full credit to the nurture email sent just before the conversion.
Pros: also very simple to set up, especially considering it’s a Google Analytics default, and so you technically don’t need to set up anything. The model focuses on the channel that directly impact revenue. This model can be extremely useful if conversions are your primary goal.
Cons: every buyer’s journey is unique and there can be countless ways to achieve conversions. The last touch attribution model doesn’t give credit to all the steps that happened before the point of conversion.
The linear attribution model, as the name suggests, divides equal credit on all the available touchpoints. So, all events are treated with the same importance, and a customer journey with 20 touches would give 5% credit to each of the touch-point.This model can be useful when you have a lot of different touch-points involved and are confused on how much attribution you should give to each of them.
Pros: the linear model is an upgrade over the last-touch or first-touch model, since it addressed the touch-points that happened in the middle. The middle stages of the customer journey can be valuable, and with this model you can see patterns that will be hidden otherwise. As mentioned, this model can be useful for companies with overly long customer journey.
Also, this model can act as a benchmark to compare against other models due to its even distribution of credit.By using this model together with last-touch or first-touch model, this model can help fill in the blanks.
Cons: it is, realistically, impossible that every action truly contributed equally towards the conversion. So, in a sense this model will always be inaccurate.
Time Decay Attribution
The name time decay comes from how this attribution models assign credits to the touchpoints: the most credit is given to the interaction that directly happened before the conversion, while actions that happened before receive less and less value the further back they are from the conversion.
Pros: this model is significantly more realistic than the previous ones: it recognizes every single action leading up to the conversion, while still giving the most credit on the activity that caused the conversion. Marketers can use this model to truly optimize the actions that cause conversions while also evaluating and optimizing the actions that played their part in building up the conversion.
Cons: while this model can provide an excellent tool for conversion optimization, this model still lacks the ability to recognize the touchpoints that introduced the customer to your brand (the lead generation).
Position Based Attribution
The position-based attribution model is essentially a combination of the time decay model and the linear model. In this attribution model, 40% of the credit will be assigned to the very last and very first actions. On the other hand, the remaining 20% is evenly distributed to every single touch-point that happened in between. For example, if there are seven touch points in total, the five touchpoints in the middle will receive 4% of the credit for each of them.
Pros: this model allows you to fully optimize the first and last touches, which are indeed generally the most important ones. Yet, position based model still gives a portion of the credit to every single touch-point.
Cons: assigning too much credit on the first and last touch-points can be unrealistic depending on your overall marketing strategy. For example, if the first touch point is “just” an email blast, would it make sense if it receives that much credit over the expensive paid advertising that contributes so much more to the conversion?
Every single marketing attribution model on this list, as well as others not discussed here, will always have their disadvantages. Every single business is different, with their own unique approach to marketing, so ideally a custom-made attribution model designed specifically for the business is the best approach.
Yet, applying a custom-made model can be expensive and time consuming, and you can still get significant benefits from using the 5 basic attribution models discussed above.