Attribution is one of the buzzwords floating about in digital marketing, but attribution models are an essential part of Google AdWords, and it is worth doing your research and taking note.
This article will give you an overview about what attribution is and how it works as well as the benefits that changing your attribution model can bring to your company regarding your knowledge of what the customer is looking for from your marketing.
Defining attribution and attribution models
Attribution is defined as “the action of regarding something as being caused by a person or thing”. Attribution models demonstrate the various scales and ways that conversion and revenue should be weighted. Models can be very basic or complicated, but the final aim is the same: understanding your customer’s journey and, therefore, the best way to get a response from your marketing.
There are seven main attribution models:
1. Last interaction
Last interaction is the most common model and also the easiest one to understand. Google has been pushing people away from this model as it is very simple and fails to unify multiple channels, unlike other models.
In this model, the last click of the customer’s journey receives all of the credit. For example, if the user saw an online advert and noted it down to check out later, then a few days later went to the website and purchased an item, it would only be the website that received credit for the transaction rather than a paid advert.
2. Last Non-Direct Click
The Last Non-Direct Click attribution model works similarly to the Last Click attribution model but does not account for direct traffic. In the example used above, the paid advert would get all of the credit.
3. Last AdWords Click
Last AdWords click is simply the campaign that drove the last AdWords click and is mainly used in PPC.
Google AdWords would still claim the sale even if the user utilized another channel like SEO. AdWords, however, does still give you more rapid costs and revenue compared to other platforms. This is especially important if you are doing daily campaigns and need information readily and rapidly available. If you use alternative attribution models, you could spend a lot of money on campaigns, which you discover a few days later were not working.
4. First interaction
This attribution model is relatively simple but not much used. It gives all the credit to the first interaction a customer with the company. For example, if the user saw an advert for your company on a website and clicked it but later went and accessed the website to purchase something, all the credit would be given to the first click. Unsurprisingly, this model is mostly used by advertisers who want to see how customers first learnt about them.
5. Linear
The Linear attribution model takes into account all the channels that have been a part of the user’s journey towards conversion and gives them equal credit. An excellent example of this would be if a user saw an endorsement for a company on Instagram, a billboard advert on their way home, a YouTube advert and searched the company on Google (and made a purchase having clicked on the website), all of these would be given equal weighting.
6. Time Decay
This attribution model works in the same way as the Linear attribution model in that it does not give 100% of the credit to any one channel. However, there is a major difference: time.
The Time Decay attribution model gives less credit the older the marketing channel the user clicked on. In the example used for the Linear attribution model, the Instagram endorsement for the company would receive the least weight in contrast to the Google search, which was closer to the conversion time. This can be an advantageous model for companies whose conversion path takes longer, for example, with luxury goods or even holidays.
7. Position Based
The final attribution model we will explore here is the Position-Based attribution model. This gives the most credit to the initial point of contact with the user but also the final interaction that caused the conversion. The intermediary steps then get the remaining credit.
This model makes much sense in that the first point of impact was probably strong enough to make an impression that leads to the final point of contact and the conversion. It also gives credit to the other channels (20% split across them), recognizing their importance.
In conclusion
Attribution models are extremely useful ways to measure the relative importance of your various marketing channels. It is worth undertaking research and discussion to decide on the most relevant model for your company, so you can make the most out of the data you have.
Having this information from the attribution models enables you to decide where to spend and where to save in your communication channels. You can also build custom attribution models with Google Analytics, which then enable you to specify the weighting of various conversions. There are also other models you can explore, such as Data-Driven Attribution in Google AdWords.
