We have already mentioned that the last cookie counts method falls far too short when it comes to evaluating the effectiveness and efficiency of marketing channels. Especially in affiliate marketing you can feel the repercussions, since only when it comes to a purchase a commission will be paid. Therefore affiliates might leave empty handed even through their channel may have had an initiating or influencing effect on the purchase decision.
This is why there are questions about which attribution model should be used in order to evaluate marketing channels correctly. This does not only apply for the evaluation of affiliates, but for the evaluation of all online marketing channels. No online marketing specialist can answer this simple question right away, because in addition to the first cookie wins principle and the uniformly distributed model, there are a multitude of alternative scenarios.
The following example shows that the results can be very different depending on the attribution model you use, especially if you have several touch points.
- Touch point 1: view at partner A (banner viewing without click)
- Touch point 2: click at partner B (banner viewing with click)
- Touch point 3: SEA with click on the product
- Touch point 4: URL input with sale
If you apply the last cookie counts principle, both affiliate partners would receive nothing at all, whereas under the first cookie counts principle, Partner A would receive 100 percent and therefore the full commission. For a uniformly distributed attribution each touch point would receive 25 percent of the commission. This means, a view would be “worth” just as much as a click. In this scenario both partners receive 25 percent of the commission. Therefore the different attribution scenarios would have significant effects on the remuneration for both partners. This also applies for the other two touch points (channels).
This example illustrates that choosing the right attribution model is not a piece of cake and needs to be well thought through and considered. Also, in this example, the first cookie counts and the uniformly distributed attribution model are not necessarily the best method to evaluate the channels’ effectiveness and efficiency correctly. It is the online marketing agency’s task to develop individually adjusted models for their clients, which not only take the order of the touch points into account, but also the channels themselves, the used advertising mediums, the position within the customer journey, and the differences between a view and a click.
QUISMA has developed an individual evaluation pattern, which has been applied to multi-channel clients for quite some time. This method takes into account a multitude of factors, guaranteeing a realistic evaluation of the channels in terms of efficiency and effectiveness. At this year’s dmexco - September 12th and 13th in Cologne, Germany - a performance specialist will present the results of individual attribution models on the basis of selected industries.