Test Your Pay Per Click Ads
[Tweet “Brad Geddes and Charles Kirkland talk about three ways to test ads in pay per click http://adalysis.com”]
Charles Kirkland: This is Charles, founder of the Media Buyer Association. I’d like to welcome you to the podcast, whether you’re listening to it day, evening, or somewhere in between, you’re going to learn a lot about the world of paid traffic.
Now, today I‘ve got the man that really needs no introduction. If you’ve been into paid traffic, you will know who Brad is. So, Brad Geddes is the godfather of traffic.
So, Brad, how are you doing?
Brad Geddes: Good. Thanks for having me.
Charles: Dude, I am excited. I am super excited today. Now, this is what happened. Brad’s got a software, and I’m going to screw the name up, because I’m very good at doing that. And we’ll just blame it on the southern accent here. He’s got a software called AdAlysis. I screwed it up probably. Who knows?
But first of all, it’s probably one of the most important pieces of software I think I’ve ever gotten an opportunity to even look at. We’re just getting started with it. But, when I initially saw it, I was thinking, crap this thing is going to be $1,000 a month. I know it’s going to be super expensive. And it wasn’t.
But we’re going to get to that. So, first of all, Brad. Why did you create the software? Tell us about what it does.
Brad Geddes: Sure. So, the software, primarily, is all about ad testing. And we essentially have sort of three ways we test ads. We look at best ads within an ad group, which is single ad group testing. We’ll do pattern matches across ad groups, so you can get insights. That’s all about ad testing too.
Ads are the only part of the account users see. Right? Users don’t care about your targeting message. That’s for you. Users think about your ads. And so, everyone in the industry says, oh ad testing’s important. Ad testing’s important.
And when we looked in reality, how people test ads, it goes like this. They set up, you know, maybe 100 ad groups. It could be 10,000 – number’s irrelevant here. It’s all about scale. And then they make two ads per ad group. And then month one, they go in and they’re like, all right, I’m going to make sure I have statistical significance and look at confidence factors.
And the reality is you maybe have three winners. And you’re a little deflated that you have 100 ad tests and three things to do. And so the next month, you do that again. You have five winners. Great. And so we have two months of data.
And the third month you start eyeballing it and saying well, that looks better than that one. I don’t really need to run test results for this. Let’s just pause this loser. And humans are terrible at figuring out patterns. We love to make patterns. So, we thought, you know what? Computers can do this. Right?
Our mantra is, if a computer can do it as well as a person, a person shouldn’t waste their time. People should be given actual data. So, essentually we did it so people could make real results with their ad tests, but then only give people actions when there’s actually something to do, and not waste our time with meaningless math.
Charles: I’ll tell you that’s exciting, because as you’re describing this, I’m thinking, hey! That’s me. Oh, that is me. Oh, third month, that is definitely me. And it’s across platforms, whether it’s Facebook, whether it’s Adwords, whether it’s Twitter. It’s like, you get to a point where it’s like, ugh! Well, yeah. I’ll take an educated guess. I usually probably turn off my most profitable ad campaigns. But you know, we live and we learn.
So really, at the core of what we’re talking about is users could care less about targeting. Only thing they see is the ad – how your ad compares to somebody else’s ad in the world of Google, you know, really you’re fighting your neighbor. Whoever your neighbor is, whoever one, two and three is – that’s who you’re fighting with.
So, what you’re really saying is, basically the thing that matters the most, you get the biggest lift, because you’re not going to get a massive lift from bidding, from adjusting bids, all that plays into a part. But your biggest gains really come from your ads – your ad copy.
Brad Geddes: Exactly. Right? Some part users see. And I also- so bidding’s important, right? It’s controlling the money. But bidding is a transient gain.
So, when you spend time setting a new bid you’re actions are relevant only until you set a new bid. And that’s where testing ads, testing landing pages, adding new keywords, adding negatives, have long-term impacts, which is why from a high theoretical level I’m a fan of automating bidding even if it’s not perfect in margins because the time you save can be so much better used in making permanent gains of adding new negative keywords, or creating, you know, a new landing page test.
Charles: Yes. Where the real gain comes, I mean, at the end of the day, if you’re doing the exact same thing as your competitors, which is basically drowning and trying to you know, come up with something better period, you really don’t come up with anything better than your competitors. You have to think differently. You have to automate differently. And you have to just plain be a lot smarter.
Now when I came across this, I was quite excited. I was super-excited. And I looked at it and I almost didn’t want to click on the pricing because I mean Brad, it’s like, you look at this, and like man, this thing’s going to be $1,000 a month, or at least $247, or $497. And I was pleasantly surprised.
I was actually somewhat in shock. I had to kind of look like, whoa! Well maybe that’s per day. I was like, surprised. I mean, completely and utterly shocked – in a good way!
Brad Geddes: So, I mean I’ve been in this industry a long time, and when you start paying eight people a percentage of spend, you suddenly have nothing left. Right? And also, everyone wnts a little piece here. So, we actually really, really thought our pricing was okay. We’ve got API costs. We have database storage costs. We have processing costs. But our heart- our costs are only related to how many ads you have in our system.
So, we said, you know what? If you’ve got 10 million ads, that is a different price than 20 ads. Right? And so we essentially took a different view of that. So, if you’ve got them, we have people paying us quite a bit of money a month because they’ve got 20 million, 30 million ads in our accounts. If you’ve got 100 ads, it’s a really low price point. I want to make it accessible to everyone, not price out people.
Charles: Oh absolutely! And I was shocked when I saw this because every other platform I think I’ve ever looked at was like, we want a base and we want a percentage of your spend. And you start looking at it like, well, we’re using something here to manage ads across platform. We’re using something here to do relevant testing or bidding.
And next thing you know, all of a sudden you end up with basically, a boat-load of people, like you said, and this is, I think, a scary example, but it’s relatively true. We make a boat-load of money usually on a handful of ads. We do. The ROI’s just insane. But as we begin to scale these things up, literally our profit margin starts drying up, drying up and then sometimes we’re literally, quite often we’re going negative just for customer acquisition.
And it’s one thing to go negative for customer acquisition. It’s another thing to tack another seven, or eight, or even five percent onto that. I mean, that really makes it where I’m working to pay somebody else at that point.
Brad Geddes: And a lot of our customers are already using Marin or Kenshoo, or DoubleClick, you know, name whoever your bid system is, alright? So they can’t absorb that price very well, or if they’re an agency, they can’t sell that at some price point.
So, it’s- I think we’re going to see the rise, and you probably see this a lot. We’re going to see the rise in, I think, the next two to three years of complimentary just get stuff done platform systems where you know, it’s- Marin or Kenshoo, I mean, they’re good systems. They don’t automate ad testing.
And if you’re using Google CPA Bid System where you run the rules, you don’t need a third party platform outside of maybe a client reporting or something, or you don’t need them for bidding. So, I think you’re going to see some software rise up that are handling everything but the bidding aspects.
Charles: Oh, I totally agree with you and I really think CPA period has changed the game. When you look at it from the standpoint of okay, do I want to sit here and try to figure this out and micromanage it, or do I just want to say get me users at an acquired user base of x, y, or z? I think that really is a huge game-changer.
Brad Geddes: Yeah. I agree. For lead-gen, that’s great. Or even Google’s rudimentary bid rules for eCommerce – they’re not perfect. Right? But you’ve got to ask yourself. Okay. What’s better – me to hand five percent of spend to a third party, or me to use these really basic tools that are three percent off my margins, or are ether six percent off? Right? And that kind of tells you which one you probably want to use. But they’re good enough for, not everybody, but for most people.
Charles: Oh and I think you just hit on a good point.I mean, and I’m even going to say maybe 95 percent of the people will be fine using those. I think then, you know, you do have that small percentage of people that need everything. And if they could double whatever they have, they would. But that percentage is, I think, so relatively small compared to the overall universe of users. And I think that’s huge.
Just looking at your software, we came into it, we were pretty much using Google ads. We’d unfortunately had some Google problems. Finally, it took a while to get the account cleaned back up and fixed. But once we got it back up and running, I’m a happy, happy camper looking at some tools like this, because I feel like this is something that I need that is probably the best investment I can make when it comes to Google pay per click, hands down. I can’t think of another platform that would give me the bang for the buck that we’ve got here.
Brad Geddes: I’m glad to hear that. That’s the goal, right? So, our goal is always focus users on actions or make bulk actions simple. And so, we’ll roll in more and more features over the next coming years on that. I mean, we just rolled out the quality score tool, which now that you’ve got- because quality score’s all about either fixing your landing page, or organization, or testing ads.
I mean that’s all quality score comes down to is those three things. So we can do things like take a look at what’s your weighted quality score? Is your problem landing lages or ad related? And run it through an algorithm and say here’s the ad group that you need to fix the absolute most. Here’s the second most. And that’s again, let a machine do the math and let the user focus in.
Here’s my top quality score gainer. Here’s my top gain over here – whatever it happens to be. And I think that’s important because there’s too many things to do where you kind of look at your data all day long and are like, well, that was interesting. And if you every say that, you probably just wasted an hour of your life you’ll never get back. Right?.
Charles: I’ve looked at data before and it’s like- I’m embarrassed to even say this. My wife is an accountant. So, literally, I have my wife crunching out these massive spreadsheets. And she’s like, okay. Here’s your problems. And some days I look at them like, I know the problem, but I don’t know how to fix it. I’ve got 10 problems. Which one is giving me the biggest bang for my buck?
And often we look at it. I think most people spend their time tinkering on the macro- even the micro things, things that really almost make very little difference overall when the big problem, like you said, may be quality score. Maybe their landing page completely, utterly sucks. And I think that is a huge problem. What do you think is the biggest problem the average user has when it comes to just using Google in general?
Brad Geddes: So, I mean this is why people like you exist. The biggest problem is not understanding the basics of organization, keyword match types, and ads. I mean in reality, the amount of accounts we still see that are one ad group and 2,000 keywords, and all broad-matched words, and they’re terrible, right? Some people get better. Organization’s the first thing they get a handle on – how to actually do organization decently.
And next, it’s about learning about match types and actually, you’re managing search queries not keywords. Right? They’re just a proxy for search queries. And then you learn about the ads, and next it’s the ad extensions which, with the new desktop layout, are super, super important. And again, most accounts don’t have structure snippits let alone some of the other lesser known extentions.
I think it’s the basics are still the problem of most accounts. I mean again, that’s why you’ve got- you know Google’s got three million advertisers or something. And you know, there’s probably 100,000 good ones. And there’s probably a million being managed by resellers and the other 1.5 million or so, in that middle range, are really where the basics will give them a huge, huge lift. But it requires you know, four to six hours of time to learn the basics well, not the advanced stuff.
Charles: Oh, I think you hit it on the head. I see so many people. They have this concept of, I read a sales letter somewhere and I should immediately be able to put money into Google and it needs to spit my money out immediately in small, unmarked bills. And you realize that so many small businesses, first thing they do is they just open an account, give it a try, and wonder why it doesn’t work.
I’ve got a friend of mine who has a local business. And he goes, could you look at this? And I’m asking him what’s he doing? And he’s like man, we’re just spending a boat-load of money and I don’t know where it’s going. And as I looked at the account – it’s a company that has an extruder that makes screws.
So, just imagine him putting all the keywords that he thought were relevant to his business. Needless to say, I mean they were literally burning money for days. He goes, I don’t know what’s happening. It’s like well, you’ve kind of got this big, giant, broad match for screw extruder that just may not be the right target. I think the reality of it is so many people just jump in and they think it’s going to be easy to make money, which is wrong. At least I think it is.
Brad Geddes: Oh, it is. It’s terribly wrong. All right? I mean, in reality, to get a basic understanding of AdWords, you can do it in four to six hours. To get a great understanding could take months of time and actually just doing the work. But, it takes five minutes to open an account. It takes three hours to build a decent looking account for maybe 1,000 ad groups.
If you’re building an account with thousands of ad groups, we’ve had accounts that have taken us a month just to build. But that’s that layout. It’s just like building a house. Right? Like, what’s the end thing look like? What do we need to start with? When does this kind of wire need to be laid versus the flooring laid versus- you know, you don’t sheetrock something before you have wires in place.
And it’s that same kind of thing. And people just sort of jump to what they see as opposed to having a structure to create a proper account. Because it takes time and effort. It’s not just handed to you.
Charles: Oh my gosh! I think so many people look at like- they expected there would be a software to do it for them. It’s like- I don’t know if you’re familiar with something called Speed PPC, but back in the day, it did a lot of stuff. It’s still a pretty neat little software but at the end of the day, the software can only assist you.
If you don’t know what you’re doing, you’re still going to put 2,000 keywords in a group and just go well, you know, it should work. I’ve got one landing page. All these things are like kind of, sort of, remotely related to my site, and I’ve got the same bid for everything. Google should fix it for me.
And I think that’s the reality is the average user’s coming in just completely blind. They’ve read some B.S. sales letter on, whatever and it doesn’t work that way. It doesn’t even remotely work that way. And with the changes in Google- I absolutely love Google. And it’s hard to think that now we’ve got Google. We’ve got Display. We’ve got YouTube. We’ve got Gmail. We’ve got- I mean, literally, there’s no shortages of opportunities to use Google.
Brad Geddes: So, I think of AdWords as an aggregation of channels because you’ve got remarketing, shopping, search, remarketing for search, display buys, YouTube – and then you cross that with devices and suddenly- and Gmail, I mean you could just keep going there’s so many different ways.
So, I no longer think of AdWords as AdWords. I think it’s an interface to manage multiple channels in one place.
Charles: Yes! Absolutely. And I think it’s really interesting. I know this is slightly off-subject but I love talking to people when we get into attribution, and lift, and we’re doing multi-channel. I think the average consumer cannot really even envision that opportunity that’s even out there. They’re thinking, I did a search on Google. That’s where my ad should be. I think they’ve completely missed the boat. And I think most marketers have, too.
Brad Geddes: Oh, I totally agree. Totally agree. Right? And the problem with attribution is you deal with end data very quickly. So, in some cases, this is where marketing managers and that thought process is so important of, let’s at least design a few different customer journeys, market to the customer journey and see if we can change things.
If we don’t have the data to prove it, let’s at least- I mean, they’re called best practices for a reason. They’re best for most people, not necessarily for everybody but let’s at least start with a best practice, multi-channel type of system and at least lay out our ideal customer journeys – start somewhere. It may be wrong, but at least it’s a starting place.
Charles: Absolutely. Absolutely. Well, I don’t want to keep you online. I asked for 30 minutes. I just want to say thank you for your 30 minutes. Where can we find out more about you and more about your software?
Brad Geddes: Sure. So, you can find AdAlysis.com. And you can contact me there. Or I do hang out at Twitter at BGTheory. And if I’m social, it’s likely to be Twitter over anywhere else. Those are the two best places to usually find me, or any SMX conference pretty much in the world – you can find me at one of the SMX’s as well.
[Tweet “Powerful PPC Ad Testing & Management Made Simple http://adalysis.com”]