Episode #015 – Cheat Codes – How To Use Data To Get Ahead

You're listening to the REI marketing nerds podcast, the leading resource for real estate investors who want to dominate their market online. Dan Barrett is the founder of Ad Words Nerds, a high tech digital agency focusing exclusively on helping real estate investors like you get more leads in deals online. Outsmart your competition and live a freer, more awesome life. And now, your host, Dan Barrett.

Dan: Alright, hello everybody and welcome to this week's episode of the REI Marketing Nerds podcast. How are you? As always, this is Daniel Barrett here from AdWords Nerds and I've been traveling the last couple weeks, I've actually had a couple different seminars, masterminds, stuff like that that I've been to and really digesting a lot of awesome, fun stuff that I really can't wait to digest and break down and give to you folks, because I really view my role here as I'm going out, I mean there's testing, developing myself inside AdWords Nerds are going to other seminars, really high level marketing people and I'm bringing that information to the real estate investing space. That's my role here, that's what I like to do, that's what I hope you're getting out of this podcast.

One of the seminars I just went to was data driven marketing. If you know anything about me, you've known I've been kind of banging this drum for a long time. I'm very big on data. In fact, when I got started in the real estate investing space, I first started working with investors as clients, I didn't know anything about it, I really didn't. I've been doing pay per click and SEO for like ambulance companies and plastic surgeons and plumbers and all sorts of random stuff. Investing space was totally new to me, totally new. Really there wasn't anybody out there that was teaching that stuff, and in fact, I still remember shock Sean Terry, shouts up to Sean Terry who's a really, really incredible guy, really successful investor. He was one of the only ones that was talking about AdWords or any of that stuff. So I had to kind of a make my own way, there wasn't really a blueprint laid out that people were sharing, we had to figure it out. The only way you could figure that stuff out is by using data. You've got to get out there into the market, see what works, see what doesn't work, take that information in and use it to make a better version of what you're doing. And essentially that's still what we're doing today. I mean even if you come into, you come and look behind the scenes at AdWords Nerds, that's essentially what we're doing, we're constantly launching new account builds, new keyword portfolios, new ad types, whatever it is, and we're seeing what connects with motivated sellers, what actually works.

I wanted to lay out a couple different frameworks you can use to use this within your own business, because it's true, we have a lot of data, we've got a lot of investors, we've got millions and millions of data points every month that we look at, but it doesn't need to be like that, and you can take this data mindset, this data first mindset and apply it to your investing business, whether you're trying to do a deal a month or you're trying to do hundreds of deals a year, you can apply data to what you do in order to make it better, in order to get more leads, get leads more cheaply, get leads more quickly. It's really a life changing process. But it all kind of gets overwhelming. Right? There's a lot of stuff to look at, there's a whole bunch of tools, so I want to break down some easy ways you can do this. Okay? Some real simple ways. I'm going to try to keep this high level, because honestly like marketing channels, tools they change very rapidly. Today we got AdWords at Facebook, five years from now I don't know what we're going to have. I want to keep this high level because the core concepts never actually change. You can apply the core concept to any marketing channel. Whether it's direct mail, pay per click, cold calling, door knocking, literally anything you can apply this framework to. And that's really, really useful because that means I'm going to give you the tool, you can take it forward and use it for years and years, not just for the couple months following this podcast.

First framework I want to give you is RFM. This is a really powerful one if you can understand it, and I'm going to give it to you through the lens of pay per click, but you can apply this to a lot of different businesses, a lot of different contexts. RFM essentially is a way of predicting who is most likely to become your client or buy something from you or sell you their house, whatever that transaction is, who is most likely to go through that transaction with you in the future. And the only way we can really predict this is by looking at who's purchased from us or sold to us in the past. So let me take a step back, break this down a little bit. RFM stands for Recency, Frequency and Money. Now what does that mean? It means if I take a list, let's say you sell camp, you got a candy shop. You run a candy store. You still penny candies, because who knows, I'm sure that's still a business in 2018. You sell candy and you want to try to make a list of people that are most likely to respond to let's say an ad you're going to run promoting your candy company. Who are you going to send that ad to? Well you can send to everybody under the sun, everybody in the world, but the response rate's going to be bad. Most people are not going to be able to buy candy from you, most people are not going to want to come to your store, most people aren't going to have the money. Whatever it is, you just said everybody on Earth, probably not going to do super well, it's going to cost a lot of money.

Now you could shrink that down and just send to everybody who's ever been interested in your store. Let's say everybody on your email list, everybody who's ever bought from you, everybody, everybody that's ever been a customer. That list would do a lot better, a lot better. But we could actually get it to be even better than that by applying this RFM framework to our list of clients. Let's say we take a huge list, everybody who's ever showed any interest in our candy store in the past, so we've got a big list of all those people, and we go through and we grade these people on the RFM scale. The first step is recency. How recently has this person purchased from my candy store? Did they come in this week? Did they come in last month? They come in this year? The more recently this person has purchased candy for me, the higher their score is going to be. And this is something that really blew my mind when I learned this, and it's still kind of counterintuitive to me, but the people who are most likely to buy from you in the future are the people who have most recently bought from you. This is just the truth, statistical truth of nature. You look at people that are buying shoes, some of those people are going to buy, the addicted shoe buyers, they are going to be buying shoes all the time, they're going to buy shoes every single day, and so if you just take the whole sample size of the store of the people who have most recently bought shoes, well those people are just statistically most likely to come back in and buy shoes in the future. So we've got recency, we grade our whole list of candy buyers based on how recently they've come into the store.

Now the next one is frequency. How often do they buy from us? It's one thing if they've just came in and bought all this candy, but it's another thing if they come in and they buy candy every single week. So every single week. Or if they buy 10 pieces of candy a year, or 100 pieces of candy a year or a 1000 pieces of candy a year. So what is the frequency with which these people buy candy from you, and I'm going to grade them there. So we got recency, I've got frequency, the last one is money. How much money do they spend, how much money are they worth? Do they come in to buy the penny candies or they buy the king size candy bars for $10 a piece or whatever it is? Who spends the most money with me? So everybody on my list is graded on these three categories. The recency with which they have bought candy for me, the frequency with which they buy candy for me, and the amount of money they spend. And the people they get the highest score across all of those categories are the people that are most likely to respond to my ad. So if I take that ad I'm going to run for my candy store and I just send it to the top 20% of my past clients that score high on the RFM categories, that ad is going to crush it, it's going to crush it. It's going to get an incredible response rate, people are going to buy that candy, it's going to be awesome.

Now, we're talking about candy. How do we apply this to real estate investing? I'm going to get specific here and I want to talk about pay per click, because you can actually apply this model to the ads you are running in AdWords, or in Facebook. Really anything. I'm going to focus on AdWords, but you can absolutely do this in Facebook. Let's look at the keywords we are running. How recently had the keywords that I'm looking at generated lead for me? How frequently do they generate leads for me? And when they do generate leads for me, how likely are they to become a deal? This is how we grade the keywords that we run. We don't just say how many conversions do we get, we look at it on an RFM basis. Certain keywords don't generate a lot of leads, but those leads turn into deals, and so they've had to be scored highly. Some don't necessarily generate a lot of leads per month, but they generate leads very frequently. This is multiple ways of looking at the same data, but allowing us to get at the core of this question we're answering, which is like how can I spend the least amount of money and generate the highest amount of deals. This is what this is about, the RFM model. You can apply this pretty much any kind of market you want to do, any audience you're looking at, you can look at zip codes that you target this way, what's the recency in which a zip code has produced a deal, what is the frequency this zip code produces deals, what is the total deal value this zip code produces. That's the RFM model you can apply to the zip code, you can apply it in the neighborhood, you can apply it to the streets that you are targeting. This is a very powerful model that you can apply across all of the market that you do that will permanently change the efficiency with which you produce deals. It's pretty freaking awesome, man.

Now, I want to get to another data framework which is probably one you're familiar, with which is 80/20. 80/20 has gotten very popular, people talk about a lot. If you're not familiar, the 80/20 idea here is based on this economic principle, it's called the Pareto Principle, and it basically says 20% of the inputs tend to produce about 80% of the outcomes. Like the classic economical example was, is like looking at bean plants. And about 20% of all the bean plants produced 80% of the harvest. Then this guy looked at the economics of Italy at the time and he said, "Wow, about 20% of the families in Italy own about 80% of the wealth." So this is the 80/20 principle. It's been applied pretty much everywhere, it's been popularized in several books, and so people tend to know about it. But it gets hairy when you try to apply it to your actual market. It's like, "What are you saying, I need to cut out 80% of my marketing?" Right? And usually that's not how it works, you don't necessarily just cut everything out, but let me give you a few examples of how you can apply the 80/20 principle to for example your pay per click campaigns. And again, I'm going to use pay per click as an example because it's what I like to talk about, or whatever, that's my baby, it's my baby. You can apply this to really anything, direct mail, any kind of marketing you're doing. Door knocking, call calling, driving for dollar, bated signs; they are all able to be analyzed as well.

Are you an investor who wants to dominate your local market? Do you want more leads and deals online? Then download your copy of the Motivated Seller Blueprint absolutely free at www.oldsite.adwordsnerds.com/gift. What are you waiting for? Go to www.oldsite.adwordsnerds.com/gift right now to get your copy of the Motivated Seller Blueprint

Let's say you have an AdWords campaign or Facebook campaign that is spending money producing leads and you want to drop your lead costs. Well how to do that. First thing I would do is I would actually look at it by device. If you go to AdWords, there's a report, they're called "dimensions". You can go in there you can see by device, you can see like how many desktop computers have converted and what did that cost, you can see how many tablets have converted and what did that cost, you can see how many mobile devices have converted and what did that costs. And you can see, like alright, some of your leads are going to come from desktops or are going to go from mobiles or are going to come from tablets, but like what percentages are coming from which devices and what do they cost. Usually what you can find is that either desktop or mobile is a much bigger chunk of your leads and they tend to be significantly cheaper. So a lot of times, let's say you look at your AdWords campaign and you find out that like okay, actually most of my leads are coming from mobile. Those leads are actually cheaper than my desktop leads. Okay, cool. So now we know this. We know that we have a one out of these three possible options that's producing the bulk of our leads, what do we do? Well we don't have to cut out desktop leads. That's not necessarily what 80/20 is telling us, it's not saying like just kill the 80%. You could do that, right, but overall you're going to get less leads, you are probably going to do less deals. Probably won’t work out for you the way that you wanted to.

Instead what I like to do, when I find a relationship like this, when I find that either desktop or mobile or tablets is outperforming everything else, what I like to do is put in a bid modifier. All that means is that I tell Google, "Okay, hey, Google, whenever you see a mobile person searching, I want to bid more than if they're desktop person." Because I know in this scenario my mobile leads are more valuable to me, they're a bigger chunk of my leads, they're cheaper than my average cost, they're more valuable to me, so I want to pay more to get more of them, but I don't have to do that across the board, I can just pick out the pieces that I want. You can do the exact same thing for day of week. Maybe 80% of your leads come on Monday and Tuesday. Well you don't have to cut out Wednesday, Thursday, Friday, Saturday, Sunday, which is actually a significant number of people do do that. You don't have to cut out the whole rest of the week, because you're going to miss out on a lie, you're going to leave money on the table doing that. Instead what you could do is use the bid modifier and say, "I want to bid more on these few days where I'm getting a lot of value." You can do the exact same thing an hour of the day. Maybe most of your leads come from 12 noon to 2 PM. Well you don't have to cut out the entire rest of the day but you can tell Google, "I want to bid more on these two hours of the day where I get most of my leads."

Zoom out again and look at what we've done. You say like, "I want to bid more on the 20% of devices that produce 80% of my leads." And then I said, "I want to bid more on the 20% of the days of the week that produce 80% of my leads, and I want to bid more on the 20% of the hours of the day that produce 80% of my leads." And you put these layers on top of each other, 80/20, 80/20, 80/20, what ends up happening is you get more leads for significantly reduced cost. You're putting more eggs in that basket where you're getting the most value. It's a really, really powerful process. And again, doesn't have to be for pay per click. You can look at the zip codes that produce the most responses to your mail, you can look at the area codes that you call and produce the most responses to your cold calling. You can look at the streets that produce the most responses to your door knocking. Whatever it is, you can apply this basic framework, and end up getting a lot more bang for your buck. It's really, really, really, really powerful.

The final framework I want to throw at you is a little more general, it's a little more abstract. We got deep in the numbers there, we got a little nerdy. I hope you guys enjoy that. Something I want to throw out to you just to understand and this is probably the nearest they are going to talk about, I promise, is statistical significance. And I'm going to throw this out here with a warning that most people tune out when I talk about this, and it is absolutely dangerous to your business. I am not exaggerating, you can data yourself out of business. You can data yourself into bankruptcy. I want to tell you a story, it's a true story. This is someone we were with, and he was doing his own testing on his website. So we were doing his ads, but he wanted to also run his own tests. He said he has someone on staff, they're used to it, they have a bunch of ideas. Cool. He was testing a different page and he would find one that was the winner and then he would kill the loser and he would run the winner, and then he would do it again. He'd put these two pages up against each other, and he'd find one that was the winner, then he would kill the loser and would run the winner. He just kept doing that, doing that, doing that. He would test two, three, four different web pages a week. We were really blown away by how much testing he was doing. He was testing and testing and testing, he was finding the winner, finding the winner, finding the winner, and then all of a sudden, he's out of business. Literally out of business. Just hadn't done any deals, hadn't gotten any leads, scratching his head, couldn't figure it out. And we couldn't figure it out either. Honestly, I wasn't sure what the issue was. And so I sat down with him, we were doing a post-mortem. Literally, just going over this thing and seeing what happened. I realized what he would do is he would run these pages sort of like page version A and web page version B, and he would run traffic to both of them, and he would get like two or thee leads on one at the end of the day, and let's say he didn't get any on the other one. He would say like, "Okay, day is over, test is over. Version A is the winner, it got two leads. Version B didn't get any leads, and so version B is the loser. Let's try the version A.", and then the next day he would start over.

Now the problem with that may not be immediately obvious. And in fact, to most people it isn't immediately obvious. I think investors are more mathematically inclined than most folks, and so a lot of you guys may be screaming in your cars right now or wherever you're listening to this, because you know what I'm going to say, but like what was the issue? Well the issue it, there was no statistical significance. If you want to test something and you want to know if version A or version B is better, you have to have a significant amount of data before you make that decision. Like let's say you imagine it like I'm running a hot cocoa stand outside my house and I'm testing 2 types of hot cocoa to see which people liked better. I've got peanut butter hot cocoa and I've got peppermint hot cocoa. Which one are people going to like? Well one person comes down the street and they said, "Oh peppermint hot cocoa, that's my favorite." And they buy it and I say like, "Cool, test over. Peppermint hot cocoa wins." What's the problem? The problem is, I only tested one guy.

The next 1000 people that come, is it still going to be peppermint that wins? I don't know. The fact of reality is, nobody knows. If you test a very small amount of data, you are very likely to get weird outsized results. It's very common if you only test a couple of conversions or a couple of leads or a couple hundred points of data, then it's going to look like one page is way better than the other page or one postcard is way better than the other postcard. But the fact of the matter is, if you test it on a much bigger population, those numbers might look very different. And so this guy in this situation is just testing and testing and testing, but he's never getting enough traffic in order to really know which page is better. So he's just kind of making random decisions. And so even though he felt like he was testing it he felt like he was improving and he felt like he was doing everything right, he essentially was just letting random chance dictate what he showed of his business to his potential clients. I'm going to tell you, if you let random chance run your business; the chances are you're not going to do super well.

So what is statistical significance, how do we avoid this problem? Well look, there's a lot of complicated ways of calculating it. If you go online and you google "statistical significance calculator", you can find, you can pop in your numbers, you can get real, it will tell you what percentage chance of certainty you have and all that stuff. You can find all that stuff online. I will tell you the general rule of thumb that I use, which is you need enough. And how much is enough? I don't know, but it's probably more than you think it is. Okay? So generally, I won't... Like for example, if I'm testing a landing page, and this is a really rough rule of thumb, but I'm testing a landing page, I want at least a 100 people on each version of that page. Version A, version B. I want 100 people on both, so 200 people total before I start to make a decision about what page is better. And even then, if the pages are, if they're very close, I might need more than that.

Now, do you always need to go all the way to statistical significance? No. Sometimes you're just kind of be able to tell, but I would urge you, urge you, urge you not to follow your gut most of the time. You let the math tell you what's up. This is again, getting back to this basic idea of being data driven and data minded. We all have biases, we all have opinions, we all have ways that we think the world works. And I got to tell you, I've seen hundreds of real estate investor landing pages at this point, I have lorded over thousands upon thousands of motivated seller leads, and I got to tell you that I am wrong the majority of the time on which page is going to do better and why. I just am. I think this page is really pretty, it's going to look great, and it just doesn't work well. And meanwhile, the page I think is ugly and terrible ends up crushing it. You just never know. So what makes us, AdWords Nerds better than most of our competition is not that we just intuitively know everything and we're never wrong and we just make the best decisions. That's not the case. What makes us better and get better results for our clients is the fact that we actually just say, "You know what? I don't know. I don't know which version is going to be the best, I don't know which one is going to work best. Why don't we test it?" That's really it. I mean I wish I could say it's because I'm a genius and everybody relies on my marketing methodology and all this stuff, but I'm actually not great at marketing. I really am not, I'm not I'm important copywriter, I'm not one of these people that just intuitively knows what the audience wants. What I am is good at creating systems that test for the outcome I want.

And so even if I'm not right in the beginning, I know I could find the right answer. The reason I can find the right answer is because I follow the data. I don't rely on my intuition. I let the data tell me what to do. And if you could apply that, whether it's the RFM methodology, whether it's the 80/20 principle, whether it's just the idea of understanding statistical significance, if you can apply this kind of data driven mindset to your investing business, you are going to do more deals for less than your competition every single day of the week. Because I got to tell you, you remember in our previous episode I was talking about the bar is low, I got to tell you, most people do not do this. Most investors are not doing this. They don't want to take the time, they don't want to put in the effort. So if you can do that, you are going to be far above and beyond your competition every single day of the week.

Guys, I hope you enjoyed this episode, I hope this was useful, you can find me online at the at the REI Marketing Nerds Facebook group, that's at oldsite.adwordsnerds.com/group, and as always, you can get all the past episodes of this show, plus show notes, cool links, all sorts of fun stuff, learn more about me at oldsite.adwordsnerds.com/podcast. Guys, this is Daniel Barrett signing off, have an awesome rest of your week, and I'll talk to you soon. Cheers