Marketing attribution is something marketers have long sought out for ROI. Today Sandro goes over what attribution is, some different methods used for it and all about Facebook’s free attribution tool.
Digital Marketing Attribution 101
Liz [00:00:00] Hey there, welcome back to Liz and Sandro’s Marketing Podcast. We are the podcast that brings you tips and insight on all things digital marketing from social media, websites, reputation management and more. Today Sandro is gonna be talking about the basics of attribution.
Sandro [00:00:19] Thanks Liz. Since the dawn of marketing there has been one thing that’s been elusive for businesses and marketers alike – ROI – or Return On Investment. How much are you getting in return, or what kind of boost in sales are you getting in return for an increase in your marketing budget? Hand-in-hand with ROI is attribution. So what is attribution? Most of us know what ROI is but I’m here to give you an overview of that second part. When the World Wide Web began, people started selling products online, it was really exciting because marketers and businesses believed they could now attribute precisely which marketing brought in which customers.
Liz [00:01:01] And actually track the source of that customer. Which you could probably do much easier when the Internet was a lot simpler.
Sandro [00:01:09] Much more easily. Two decades later, things aren’t that simple and we still aren’t there yet. 120 years ago, a guy named John Wanamaker famously said, “Half the money I spend on advertising is wasted. The trouble is I don’t know which half.” This is an oversimplification but the question still remains, how effective is your marketing mix and where should you invest to grow? At its simplest, attribution is the process of matching an action, like a purchase or getting someone to sign up for your email, matching that action to a specific marketing channel like TV or Google or Twitter or Pinterest, so that you can measure the impact of that channel and determine the return on investment.
Sandro [00:01:59] But here’s the issue today like Liz was saying, people generally don’t just buy products using one device or one channel in a single session, at least not most of the time. There’s more consideration required with multiple visits to websites or TV or radio or even billboards. Or heck, I was driving in today and saw the Goodyear blimp over the stadium downtown because there’s an Indians game today. Especially the higher the price of something you may be buying like a washing machine.
Liz [00:02:22] Sure it’s a longer sales cycle so you’re going to be thinking more about it. Or a car.
Sandro [00:02:28] That’s my next example, yeah. With multiple advertising platforms competing to be responsible for your end results, or purchase, it can be difficult to attribute the final action to just one channel. So an example, let’s say a car. My mom just recently purchased a car. It took her two months to decide what to buy. I don’t know her specific pattern but, just making this up, let’s say she saw an ad on TV, then she Googled some information, then she saw a review online, then she heard something on the radio, then she saw an ad on Pinterest and finally she saw an ad on Facebook.
Sandro [00:02:59] That’s five touch points for the journey: TV, Google, radio, Pinterest and finally Facebook. Five steps before she finally purchased. So which one of those five do you give the credit to? Which one should get attribution for the sale? This is the problem today and this is a problem many companies are beginning to tackle including the big three – Google, Facebook and Amazon.
Sandro [00:03:26] So here’s some examples of different methods of attribution. The most common and the one most default to is last click. So on that example of TV, Google, radio, Pinterest and Facebook, Facebook would get 100 percent of the credit for the sale for last click attribution. There’s also something called first click. So the TV would get the 100 percent of the attribution because she saw the ad on TV first. That’s another method to do it. Those two aren’t really that great to be honest.
Sandro [00:03:59] There’s even credit or linear, meaning you divide 20 percent, let’s say the car was thirty thousand dollars, the return on investment, you divide that thirty thousand by five different channels evenly. That’s six thousand apiece. So TV would get six thousand of the credit, Google would get six thousand of the credit, radio would get six thousand and so on. That’s even credit or linear.
Sandro [00:04:19] There’s positional 30 percent which gives 30 percent of the first touch and the last touch and then divides the remaining 40 percent amongst the remainder of the channels. There’s also something called time decay. So time decay one day or time decay seven day. Basically it’s the last touch, which would be Facebook in this instance would get 50 percent, then Google would have gotten 30 percent and the radio would’ve gotten 20 percent. Basically it decays over time. The first one would get the least amount of credit but it would get some credit still. So there’s time decay also.
Sandro [00:04:59] And another one is something called data driven and that’s basically a model developed by Facebook themselves so it’s proprietary to them and divides the attribution amongst different people through their own model that they’re not going to tell you. Google has this also and they update it periodically.
Sandro [00:05:10] So there’s different methods to do this. All that probably makes your head spin and you can test and consider all of these. Generally the more expensive your product, the longer you want to open that window. Facebook allows you to keep the window open up to 90 days to track data and see how many touch points there are. However, a good place to start for most businesses is that one day decay just to divide the attribution, give it more weight to the most recent and less weight as time goes on.
Sandro [00:05:40] So again this is an overview. We can talk about each of those at length but I’ll stop there in terms of that explanation. However, I wanted to point out that in the race for attribution tools between Facebook, Google and Amazon, the current lead by a huge margin is Facebook. This past October, they released a 100 percent free attribution tool that works across multiple channels.
Sandro [00:05:59] Meaning during the setup process, they will help you set it up to receive data from Google, both organic and paid Google Ads, Twitter, Pinterest and a bunch of others. Currently Google and Amazon only give attribution for their properties. Nothing outside of their walls. They do have something in beta, but nothing has been launched yet. Only Facebook does this right now. And from what I can tell, people love it. I personally have trust issues with Facebook.
Liz [00:06:34] I do too. They have just had so many slip ups. I just, any, I’ll still continue to use Facebook, but I take everything that they tell me, all of the data that they give me with a grain of salt. I just don’t trust them. And that’s not to say that I 100 percent trust Google either, I take their data with you know with a, I put it, I try to put it in the bigger context, but I will say I trust Facebook less than I trust Google. I am very anxious for Google to release their version of this tool because I’m going to trust them a little bit more.
Sandro [00:07:12] I completely agree. There’s so many different ways that they’ve violated our trust.
Liz [00:07:16] Yeah.
Sandro [00:07:16] But people seem to really trust and enjoy and believe in this current free product. And people are saying this free product may be as big some day as Google Analytics, another free powerful product. But from what I can tell, they are legitimately giving other companies credit when it’s due. They’re not taking all the credit for a sale. The downside to this Facebook tool is that right now they don’t attribute or can’t attribute or attach to Amazon. So if you make a sale or purchase on Amazon, that’s not being connected to Google yet. I’m sorry, that’s not being connected to the Facebook’s tool just yet. Also they don’t have it for TV or radio obviously.
Sandro [00:07:59] However six months ago, we did report on this podcast about a patent where Facebook, they want their app to be listening all the time on your phone because allegedly they say they want to hear the TV or radio ads you hear so they can attribute more thoroughly. At least that’s what they claim. So this attribution tool from Facebook is completely free. All you need to do is install the Facebook pixel on your website and activate it, as well as set up your other marketing channels like Google and Twitter. And also set up which attribution method you want to give. Like last touch, first touch, seven day decay, one day decay or whatever. And boom, after a few days you’re going to start receiving your data.
Sandro [00:08:29] And from what I’ve seen, people are blown away. One guy who is a big Facebook ads guru, Jon Loomer, he sells courses for online marketing, but he also tries to capture people’s emails and send them free weekly emails. He saw after implementing this, that Facebook ads bring him the most email sign ups. However, Google ads and email sell him the most courses. So after this, he started adjusting his money to spend more on Facebook for email sign ups and started spending more on Google ads and emails to sell the courses.
Liz [00:09:14] Interesting.
Sandro [00:09:14] Yeah. So there’s a lot you can learn from this. And one other great thing, I’m going to give Facebook all the credit because they’re pretty creative here. They also do off-line conversions, meaning let’s say you’re a restaurant, you put out some Facebook ads. Well you don’t know if somebody’s purchased because they purchased there in-store.
Liz [00:09:33] Or you’re an attorney and you sold an estate plan, you don’t typically sell those through a website.
Sandro [00:09:38] Right. It’s in person, correct. Well, you capture, let’s say you email their receipt or you capture their email address there at the restaurant or the attorney does or their phone number. Well, every few weeks or so you upload those sales, you upload that to Facebook and Facebook will match it to who they showed an ad to, and they’ll say, well this we showed this and this turned into a sale. That’s off-site conversions that you can do by uploading emails. And there’s ways you can match it to Square matches up with Facebook if you sell through Square or other products. So they’re being pretty creative, they really want to match sales with ads.
Sandro [00:10:18] So again, there’s a lot to this but the bottom line for me is if you’re selling things or paying people to see your ads to buy things from you or sign up for your e-mail, I think, and it’s a little tricky to set up so you may need some help, but I think this Facebook attribution tool that’s completely free, you don’t even have to advertise on Facebook, you can just implement the pixel and it will start tracking how people start buying things. I think it’s a great way to start to explore how or why people are choosing to buy things from you.
Sandro [00:10:59] Because again, it’s only going to be getting more complicated. Back in the old days, marketing, our professor told us the most famous billboard ad of all time was on the side of a highway and it just said, “Free coffee. Next exit.” And people just flew into this restaurant for free coffee. Attribution, the billboard ad. These days, who knows? It might have been a Facebook ad.
Liz [00:11:12] Yeah. Because I’m thinking, I’m thinking 20 years ago when Google first rolled out pay-per-click, it was just pay-per-click. That was it. And you could track clicks and then sales very easily. And now there’s display ads on every website. I mean it’s just…
Sandro [00:11:29] I mean, is it Instagram, because this tool will track if it was an Instagram organic click, if it was an Instagram paid click, if it was a messenger click, where it came from. So yeah, there’s so many different methods.
Liz [00:11:41] Yeah. And by the way, I love that you simplified that car buying example with your mom to just five touches. Because I think, like the reality is it was probably many, many touches. Back in January I did an episode on what is multi-channel marketing. And this attribution piece kind of plays into that. And I gave two examples in that episode. I’ll just give one right now. L’Oreal, the cosmetics company, they are now saying that it takes 25 touches before a consumer makes a purchase. So, it has increased substantially. Those 25 touches could be across a variety of channels.
Sandro [00:12:22] And it could include your friends, you could say, “Hey are you happy with your car?” How are you going to track that?
Liz [00:12:27] Definitely. So yeah, there will always be some elements to this that are not trackable but we’re trying to do the best we can. And I have not played around with Facebook’s tool yet, but I’m very intrigued to. I’m going to probably in the next week get it going for at least one client so we can see how it works. Of course I’m intrigued to play with it, I’m intrigued to see how they’re going to monetize this in several years because the number of Internet users is like you said, it’s leveled off and people are leaving social media channels, so this could just be another source of revenue for Facebook. Not that that’s a bad thing. You know if it’s a tool that that works well and gives us good data I’m all about that. But I’m just intrigued to see how this pans out.
Sandro [00:13:16] Yeah. Google has their Google Analytics 360 which is a paid version of Analytics. But the free version so powerful. So maybe Facebook will come out with a paid version of their attribution tool.
Liz [00:13:26] Yeah. Great info. I think this has given us a lot to think about. We have some tools to go out and test and report back on. And if you guys have any questions we’d love to hear from you. Send us your thoughts, questions via email or one of our social channels. We are on Facebook, Twitter and Instagram. Feel free to show us some love on iTunes. We are also on Stitcher and Google podcast. That wraps up today’s episode. We will see you next time.