Understanding Where Your Leads Come From With Attribution Models

This video is for anybody who is spending money on marketing or advertising, trying to drive traffic to your website and generating leads or sales. We all want to know what is it that I spent money on that worked. Is it natural SEO or is it paid advertising? Is it social media? What is it that’s generating my sales and my leads, okay. If you’re looking at any kind of reports and you’re talking about lead sources, you’re using an attribution model. Now, if you don’t know you’re using an attribution model, then I can almost bet with certainty that you’re using the last interaction attribution model.

What that means is you are attributing that lead to the last interaction with your website, okay, so doesn’t matter if they’ve been to your website 15 times before, if they typed in a search term, clicked on your listing in Google, and then filled out lead form. That was their last interaction. That’s what you’re tracking, but let’s think about how we discover and buy things on the web. We may have actually searched for something, discovered a new website, and when we landed on their website, they may have dropped a cookie and then started retargeting us all over the web, everywhere we went, right. Now, I clicked on that ad again. I’ve come back to the website a second time, still not ready to buy.

A week later, I’m on social media, on Facebook for example. I see a post that my friend likes and happens to be the same company. I think it’s great, I click back on that as well, and now finally I decide to go ahead and buy the widget or fill out the lead form. Now to that company if they’re using the last interaction lead model or attribution model, they’re going to assume that that was a social media activity that brought them that customer, right. We went to the website three times and only one media got credit for that, right. You can actually go into Google Analytics and you can actually look at different attribution models. We’re going to talk about what those attribution models are today.

Most people are using Google Analytics, even if they’re using some kind of other source. If you go into Google Analytics, and on the left-hand side, you look for conversion, a sub-item under conversion is going to be attribution model, and there’s several attribution models in there. The first one is the one that we covered, okay, and that’s last interaction. That’s the most simple. You’re assigning 100% of the credit to the last interaction. Now, right below that is another one that’s very similar.

There is last interaction except for direct, so it’s basically saying take out the time that they literally typed my JuicyResults.com domain and then came to the site. What was the one before that that brought them that was not direct? That make sense because if they already know about website, they may be coming back with the intent to purchase. None of your marketing really made that happen. That was something before that that you want to attribute that to, okay, so there’s several types.

There’s another really interesting one called linear, and linear is actually going to go through all of them and it’s going to basically assign an equal credit to each touch point. In our example, there were three touch points, so each of those would receive 33% of the credit. There’s also a first interaction model and some marketers who are building their brand and they want to know what was the first way they discovered my website, okay. I don’t care what brought them back. I want to know what that first interaction was. By the way, you can toggle back and forth between these, so you can look at the same data and say show me the first interaction and then show me the last, the most recent interaction, so you can toggle back and forth.

Now, there’s also one called time decay. Now, time decay is really interesting because it’s a lot like linear except for that it assigns a little bit more credit to the more recent ones, so it’s basically assuming that the most recent interaction got more credit than the second interaction which got a little bit more credit than that original interaction, right, because you actually drove that person to take action. Now, I think one of the most interesting ones that you should consider is position-based attribution model.

The position-based attribution model would basically get 40% of the credit to the first and to the last interaction. In the example we went through, there were three interactions with your website. The first was natural SEO. The last was social media. It doesn’t matter how many times they came in the middle, all of those which share the remaining 20%. In our simple example, the retargeting would have gotten 20%, but in an example where then they became 10-15 times, the first and the last would still get 40%. All of those middle interactions would divvy up that remaining 20%.

Why would you want to do that? Why is that interesting? Well, if you’re spending money, you’re marketing, and you’re getting people back to the site, we all know that it can take lots of touch points before somebody decides to do business with you. You don’t want to rule out things that are bringing people back to the site like retargeting, social media marketing, natural search, even direct if you’re doing some offline marketing that you know is bringing them back to the site, but you still want to give those things credits. You don’t shut that marketing off, but you don’t necessarily want to allocate the same amount of budget to the ones that are giving you a landslide on the initial interaction and the last interaction, right. The thing that brought them there for the first time and anything that actually cause them to come back and actually commit to the sale are the lead.

Attribution models are great thing to know about. It’s a great thing to help you slice and dice your data, and I hope this helps you have a more quality conversation about your marketing.