Attribution relates to website conversions, such as a purchase or a lead, and what marketing channel is responsible for that conversion. Many marketers only stick to one when evaluating channels. Often times this data is skewed and doesn’t paint a very accurate picture. While there is no 100% right way to look at conversions, there is a way to paint a better picture.
What are the attribution models?
Often the default setting for conversions, the last click model gives 100% of the credit to the last touchpoint within Google Analytics. For example, if a user were to first see a paid ad, then see an ad on Facebook, then go to the site directly, 100% of that conversion is given to direct site traffic. While this doesn’t paint the clearest picture as to the whole multi-channel marketing process, it is useful to know which touchpoint lead to the sale.
Just as the name implies, this is the inverse of the last-click model. This attribution model gives 100% credit to the first touchpoint in the marketing process. For example, a user sees a paid search ad, then later comes to the site from Facebook, then comes back through a retargeting with a 10% offer on it. 100% of the conversion is credited to paid search, although the 10% offer could be what pushed the user to finally make a purchase.
The linear model is an unbiased model as it distributes the conversion evenly across all touchpoints. For example, if a user visits the site 5 times, each time from a different source and on the fifth time they make a purchase. 20% of the conversion will be distributed to each channel. This model should be used by marketers when looking at the overall effectiveness of a marketing channel. The general saying is that it takes 7 touchpoints with a user until they ultimately make a conversion. The linear model takes this into account and gives an average for all the touchpoints.
The time decay model gives the most credit to the touchpoint that is closest to the conversion. As each step is further away, it gets less of the percentage of that conversion. This model is perfect to use as an evaluation tool when you want to see which touchpoint is most likely to tip the scales to a conversion.
Position based conversions give the most credit to the first touchpoint and the last touchpoint. For example, if a user first finds us on Facebook, then goes to the site directly, then clicks on a retargeting ad and makes a purchase. The majority of the credit will be given to Facebook and the retargeting ad. This model is perfect if you are looking to see where a user first saw your brand and where were they finally ready to complete their conversion.
Which model should you use?
The model you should use really depends on your specific business needs and what marketing channels you are using. There is no right or wrong marketing channel to use, only varying degrees of transparency into what sources or channels are leading to conversions.
If you don’t have a very clear understanding of what tactics are leading to conversions, we’d love to help clear up any confusion.