What is Attribution Modelling

Download now

Download "What is Attribution Modelling"

Attribution Modelling. If you are in the same position as many other people and don't have a clue about what it is and how it is used in digital, then listen up.

Attribution Modelling. If you are in the same position as many other people and don't have a clue about what it is and how it is used in digital, then listen up. With the help of Social Steve, Display Dan and PPC Pete, Kevin simply explains how different attribution models are used to attribute sales and conversions in digital.

Video Transcription

Imagine that your business or organisation has 3 sales members. First, you got Social Steve. Last month we went out and made €1,000. Then you've got Display Dan. He made €2.5k. And then you've got PPC Pete. He's absolutely killing it. He went out and made 10K last month. So if I was to ask you who was the best member of the team was. It's pretty obvious, isn't it? But what if I was tell you something else. What if I was to tell you that Social Steve actually went out and brought in 100 new business leads. Display Dan, he brought in 20 but PPC Pete. He actually only brought in 5 new leads. Who is the most valuable member of the team now? It's not quite as easy is it?

Attribution Models are the rules or set of rules that are used to determine how conversions and sales are attributed to various touch points along the conversion path. So let me give you an example. Imagine that a customer of your website first saw an ad for your business on social media. He clicked that ad and we went to the website

but he didn't buy. Next he saw a display ad which reminded him of

that product that they saw and wanted to buy. So then they decided to Google your business where they find a PPC ad. They then convert, they then buy the product and you make a sale. Which of these channels should get credit for that conversion?

In the last example, the last click, the PPC click would be given credit for the sale because it's typically used as the default. But what about that click on the social media ad and what about those display banners? Surely they helped the conversion as well. So we need a way to analyse that and see how they're impacting sales. We do that by looking at your stats through window of

different attribution models.

So let me give you a couple of examples of attribution models.
The first and most common is 'Last Click'. We already mentioned that. That's when the last click before the conversion gets all of the credit. The opposite of that is 'First Click'. And this is when the first click gets the credit, that would be the social media ad. It would get the credit for the final conversion that happened on PPC.

Another common model is the 'Linear'. This is where we give equal credit to every touch point. So the social ad, the display ad, the PPC ad. They would all

get equal weighting. The final model is 'Time Decay'. This is where we attribute the conversions based on the proximity of time to the actual conversion.

With the clicks being closest to the conversion getting the most weighting.

So which attribution model should we use? Currently, the default attribution model is last click. And that's a great way to see which channel is actually leading to conversions. But that's not the full picture and if you are advertising on multi channels like social media and display, it's important to have a bird’s eye view of how those channels are performing, how they're helping drive the final conversion. So it's not a case of changing to a totally different attribution model. It's more about at this stage, being aware of how your data would look through the window of a different attribution model. So it's quite complicated, it's in its infancy and it's something that advertisers are really having to become better and better at because they want to make sure that no matter where they are spending their money they're getting a return for it at the end.

If you have any questions on attribution, please feel free to leave us a comment below and until next time, thanks for watching and take care.