1 / 4

What is the algorithm behind a proper marketing attribution model?

Marketing attribution aims to help marketers get a better picture of when and how different channels of marketing play contribute to the conversion events. You can then use that information to inform future budget allocations.

todybecke
Download Presentation

What is the algorithm behind a proper marketing attribution model?

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. What is the algorithm behind a proper What is the algorithm behind a proper marketing attribution model? marketing attribution model? Understanding the steps that a customer takes before converting can be as valuable as selling itself to marketers. Attribution models are used in the customer journey to assign the credit to touchpoints. For example, if a consumer purchases an item after clicking on a display ad, crediting that whole sale to that one display ad is easy enough. But what if the consumer has taken a more complicated purchasing route? Initially, she may have clicked on the company's display, and then a week later clicked on a social ad, downloaded the company app, then visited the website from an organic search listing, and converted in-store using a coupon in the mobile app. That is a relatively simple road to conversion these days. Marketing attribution aims to help marketers get a better picture of when and how different channels of marketing play contribute to the conversion events. You can then use that information to inform future budget allocations.

  2. Models for Allocation Following are a few of the most common models of attribution: Attribution by last-click With this marketing attribution model, before converting all credit goes to the last touchpoint of the customer. This one-touch model does not take into account any other commitments which the user may have with the marketing efforts of the company leading up to that last commitment. Attribution with first click The other one-touch model, first-click attribution, gives the first action the customer took on their conversion journey 100 per cent of the credit. Before converting it ignores any subsequent engagements that the customer may have had with other marketing efforts. Linear apportionment This marketing attribution model of multi-touch allocation gives equal credit to every touchpoints along the path of the user. Attribution of time decay This model gives more credit than touchpoints further back in time to the touchpoints that occurred closer to the time of the conversion. The closer an event takes to time, the more credit a touchpoint gets. Attribution in the form of a U The first and last commitments get the most credit and the rest are equally assigned to the touchpoints that occurred between them. In Google Analytics, 40% of the credit is given to the first and last commitments, while the other 20% is distributed equally across interactions. Marketing attribution is algorithmic, or data-driven. When the attribution is handled algorithmically, there is no pre-determined set of rules for the assignment of credits as is the case with each of the above models. It makes use of machine learning to analyze each touchpoint and create a model of attribution based on that data.

  3. Vendors typically don't share what their algorithms take into account when modeling and weighing touchpoints, meaning the marketing attribution results may vary by provider. The data-driven attribution by Google is just one example of algorithmic modeling of the attribution. Attribution made to custom As the name suggests, you can create your own attribution model, using your own set of rules to assign credit to touchpoints on the conversion path, with a customized option. Benefits, limitations of attribution How to choose a model, and make it? Marketers face the ongoing challenge of being able to stitch together all the different touchpoints available to their clients for a grand view of marketing attribution. Improvements have been made, with greater capacity to incorporate mobile usage, in-store visits and phone calls into models, but perfection is elusive. Attribution has never been more important from a marketing measurement perspective, given the increasing fragmentation of platforms and the types of media that marketers have at their disposal. Unfortunately, the attribution nature is one where the goal posts are constantly moving and just like an asymptote, we will never be able to reach the 100 percent attribution point. Find one that makes sense, and stick with it. Whether it's first touch, last touch, or blended, it's really important to get everyone on a team to buy in and stay with it over time.

  4. As marketers invest in more channels and digital media, it's only getting harder to get a unified view of a client's journey. This will become increasingly complicated by increased investments in influencer marketing and Amazon where the creation of unified IDs presents significant challenges. While many of the major players like Visual IQ are looking through partnerships to solve some of these challenges, we will still face major challenges on the horizon for out-of-home (OOH) media as an example. In addition to the customer journey tracking that the marketing attribution platforms of Google and Facebook) provide, we will probably see the development of variance analysis solutions within the platforms that will enable marketers to better understand the existing impact of their strategies. The key takeaway here at an overarching level is the convergence of data across platforms, and the ability to understand channel-wide interactions in both impression and click capability.

More Related