Using Attribution IQ in Freeform Tables using-attribution-iq-in-freeform-tables

Attribution IQ allows you to change the attribution model to any of ten rules-based models on a Freeform column on the fly.

Transcript
Hey everybody, it’s Doug. In this video I want to show you how you can customize and configure attribution in freeform tables in analysis workspace. So here we have a freeform table and I’m just going to drag in marketing channels. And for my metric I’m going to pop in orders instead of occurrences. Let me just replace that. And so what I’m talking about here is that we can go into this gear over here and we can jump down to the bottom and it says use non-default attribution models. So we click that and you can see that by default this is going to be last touch. And that’s very common for conversion type variables and in this case marketing channels. But this is very configurable. Now we can choose between different models. So if I do this dropdown right here you can see that there are a number of different models that we can choose from. Last touch, first touch. And so these are probably all known to you and you can look them up in the documentation. But you know, in this case the last value is given all the credit for the order. In this case, the last marketing channel. I’ll get more specific there. The last thing they did to come to the site is given all of the credit for the order. First touch, maybe it’s the very first thing that they did to come to your site that you’d like to give the credit for the order. And there’s linear, which is split it up and participation gives everybody all the credit. And you can see all these different ones here you have same touch, you’ve got U-shaped, and these little images can give you an idea of what it does. So for example, with U-Shaped, you get 40% to the very first touch and 40% to the last touch and then split up 20% in between. You have J Curve and Inverse J, which give more credit to either the last or the first touch. You’ve got a time decay, which you can use here and you can set the half-life on those here. And then you also have a custom model that you can use where you can choose how much goes to the starter, the closer, and all the players in the middle to split up. And then you can select your lookback window. So how far do you actually want to look back to see the starter, the players, the closer, et cetera. So in this case it says 30 days, but you can see here that you can choose a number of different options really anything up to 90 days there. And then you can choose visit, and you can choose visitor. So visit obviously is, you know, everything had to happen in that same visit. Visitor of course is all the visits for that visitor but limited to the reporting window. So depending on the timeframe that you have set to the report. And then you can set a custom time. But even if you do that, the maximum is 90 days. So if you try to go beyond 90 and I’m going to type in like 99 it’ll actually just put in 90 as you can see there. And so in this case, I’ll go ahead and set that to our custom settings here and lookback of 90 days and apply that. Our numbers now change and are applicable for that new custom attribution model. And if I hold command I can actually now add some more columns for orders. I’ll just kind of add those and we can then change per column the attribution model. So I have my custom one there. I’ll leave that one up front. And I’m going to choose down here and go into the settings here and we’ll choose last touch on that one. And then in this last one I’ll go in and choose something like J Curve, whatever. And so now you can compare these different numbers with these different models. There’s no limit to the number of models that you can compare here. So if you want to compare all of them, great, that’s fine. And of course you can compare them on different metrics. You can also add segments. So if I want to say, you know, just for weekend visits I can drop that in here and it will update the information just for weekend visits and show the data according to these attribution models. For that matter, segments can also be added at the column level. So, if I go up here and I’m going to choose products and I’m going to choose selected products and just grab one of these products and I can add that as a filter and it will apply orders for that product according again to this custom attribution model. Okay, I’ll delete that for the moment. Now, breakdowns are also supported. So when I go back to my dimensions and let’s just say I want to break it down by mobile device type. And I’ll come down here and we will break down our natural search by mobile device type. And I have that attribution model still at the level of the column. But now I can also for that breakdown, even choose a different one. So in this case, I can mouse over this column right here and I can edit that. I can change any of these settings. I’ll choose a different one here, maybe a U-shaped and apply that. And now again, in the overall marketing channels I have orders at the custom attribution model here. But when it gets down to the device types within natural search, I am using a U-shaped for my attribution model. So in case you have different teams, for example, working on different marketing channels, and if each team has a different way that they like to do attribution, you can get very specific with these attribution models and so that they can understand attribution within their own channel. Attribution models can be applied to any dimensions really. So eVars and props and page type, internal promotions, search terms, whatever, right? So you can use this great feature to really understand the attribution on your site. Have a great day. -

Each attribution model fully supports segmentation and breakdowns, and allows you to make unlimited comparisons on any dimension (even props!).

Attribution IQ is available exclusively for Analysis Workspace, for customers have the following packages:

  • Adobe Analytics Foundation
  • Adobe Analytics Select
  • Adobe Analytics Prime
  • Adobe Analytics Ultimate
recommendation-more-help
b5d9c99f-be9f-4b96-8809-4e7d6ae353ba