Transcription
Hello and welcome back to another episode of the Collective Clicks podcast. This is your host, Brandon Bateman, and today I'm joined by Garrett Craigan from my team once again. This is part two of our three-part series. I don't even know what we're calling this thing, Garrett. It's basically like the three most important things in PPC, sure? Yeah, something like that. And it's all about conversion.
So, Garrett, why is conversion important? Wow, that's a big question. I would say that it's the biggest lever you can pull in impacting your cost per lead and your cost per deal. It's how much those leads cost because you've already paid for the click. So this is the place where you can bring down those costs or really make them way too high.
And if you're listening to this and you're thinking, "Well, I don't care about conversion rate; I care about ROI," let's explain why this matters. A lot of people measure PPC cost per lead pretty often, and some people, based on that and thinking paper leads may be a similar channel, start to think that they're paying for their leads in Google Ads. Which, actually, is it true? Yes, but it's true in the same way that on direct mail you might have a cost per lead, but you might have a certain cost to send a postcard. Just because you got one less lead doesn't mean it cost you less, right?
So pay-per-click marketing is literally that—it's pay-per-click. If a click is the item that we're selling and everything that we're buying, then everything that happens after a click is really important for getting to our return on investment. What a lot of people don't realize—like Garrett, I know you've seen this, and I know I've seen this—our clients will sometimes get upset around things like, "I got this lead that was outside of my target locations" or something. "I spent all this money for this stuff." And every time I see that, I think, "Okay, that's a big problem." I also think, "Well, what about the 80% of our budget that's wasted on clicks that I convert that you never get a lead to know that there was a problem?" That's the much bigger problem we want to tackle here, right? Even more than low-quality leads sometimes.
So we brainstormed together and we got seven things together that impact conversion rate that are pretty important. If you haven't heard any of our concepts before, I can almost guarantee you most of these are things you haven't heard before because I'm sick of hearing when people talk about conversion optimization like the same two things you hear about over and over and over again. And all the other things that we find make a difference that we just don't hear about.
So some of those common things are going to be on this list, but we'll also talk about some stuff that we found that is a little bit less known. So let's get into it. The first thing is yours, Garrett. It's Hick's Law. What is Hick's Law?
So Hick's Law is the principle where the more choices that a person has, or the more options that a person has, the longer it's going to take them to come to a choice. So in practice, it's the idea that if you want a person to make a choice quickly, give them one option. In a page, for instance, or a website, it's the idea of giving them a button that says, "Get a cash offer," not four different options to view offers or watch testimonials or join our newsletter or watch this video. It's just, "Here's the button, do this," and make that as simple and clear as possible.
Yep, totally agree. I think it's so easy to forget this because when you're building a landing page, sometimes you're thinking, "I want this to apply to someone in this circumstance, and I want it to apply to someone in that circumstance." We start adding in all this stuff that, without that stuff, you're actually better off because you're trying to speak to your audience. I think who is it that gives that analogy where it's basically like this: just say I hand you a bowling ball. Can you hold it? Yes. Let's just say I hand you another bowling ball. Can you hold it? Yes. Let's just say I give you a third bowling ball. Can you hold it? Probably not, and you probably dropped all three of them.
It's a similar circumstance with the messaging on landing pages, which is something new people need to know. And that actually leads super well into the next point here: not everybody needs the same landing page, and that's something that we have to recognize.
So if you're thinking, "I'm trying to speak to all these different people in all these different situations; I can't fit it all on my landing page," let me give a couple of examples of when you might need different pages for different purposes. One prime example is that a website and a landing page are two very different things that have different purposes. A lot of people would actually end up using their website as their landing page when they shouldn't.
For example, let's think of some of the characteristics that a website has. It usually has lots of words on it. Why do you put lots of words on your website? Well, it's because for SEO. If you had really thin content on a webpage, you're going to have a hard time making search engines happy. Versus if you have tons of copy on it, you're going to have an easier time. So what's that going to do? It's going to make it easier to rank, and it's also simultaneously going to make it probably very low converting.
So that's good for SEO but not good for PPC. Also, you might have links to all the other pages on the website. Why? Because you need an internal linking structure for SEO. But do you need that for PPC? You probably don't, right? So that's where one of the most clear divisions here is you need a separate landing page for PPC than you do for your website traffic, and that's a really important piece to getting this to work for the months.
Another example here is even with air between channels. We've found the kind of landing page you want for YouTube advertising versus the kind of landing page for search advertising. The difference in terms of the message you're talking with each other is at different stages of the funnel.
And then lastly, and this is where people very rarely get this quite right, is just even within a certain channel. People just assume, "I'm going to have a PPC landing page," when the reality is it's going to be some people that are searching for certain things and other people who are searching for other things. I can tell you what the average person does here is they're just going to put on their landing page stuff that applies to everybody. What that does is it makes it so your conversion rates lower when you'd be better off if you had different landing pages for each of these subsets of what people could be searching.
It's kind of like when you're doing mailers. It doesn't make sense to send the same postcard to somebody facing pre-foreclosure as you would send to somebody in probate, but people do that anyways, right? And then their message is either really relevant to only half their list or it's not relevant to either because you're trying to speak to too many things.
So with the landing page, you have to do this so that you have separate landing pages for each thing. One way that you can do that to make it easier, if you're thinking this is wildly complicated, is we use something called dynamic landing page content. The way that this works is we'll control the content on the landing page based on the URL parameters. When we do it that way, it actually makes it so that we can have a single landing page, and then the content looks different to the person depending on where they came from. That actually increases conversion without increasing the complexity at all.
So I know that was a lot to unpack for number two there about not necessarily always having the same landing page for everybody. We found that we have a lot better success when you match very closely what they're looking for and even use the same words that they're using to explain what you're explaining. That's how you minimize the risk of somebody clicking on your landing page or clicking on your ad, getting to your landing page, and then thinking, "That's not what I was looking for."
The third one here is form structure. So let's talk about different ways that you can do this. I can tell you that you're not alone in wanting to get as much information from the potential seller as possible, right? Because the more information we have, the better we can qualify that lead. However, the way that we get that information matters a lot because it turns out if you just ask everything that you want in that first form, then you're going to end up with a low conversion rate because you're like smacking these people as they enter the landing page with this big form that looks like this big commitment.
What we found is if you can get that in smaller chunks, then you're much more likely to get the response that you need from the seller. So there are a few ways to think about this. I like to think about this in terms of two stages. The first stage—this is what you show in the initial form—should just be the minimum amount that you need to consider that this is a lead. In our case, where we usually find that is your address and the phone number.
Now, some people say address only, and we've tested it. The thing is, the address-only leads aren't that useful because you have to skip trace them anyways, and you might as well be going to a colder list. They actually convert to contact information at a lower rate than when we do address and phone number in the first step. We have the best success when we just look for the minimum that we can ask for at the very beginning. That tends to be the address and the phone number, and then we ask for the rest, right?
So that's where you'll usually see this two-step structure, and that tends to work really well. You still get the contact information from the beginning. Now, believe it or not, the leads that only fill out the beginning of the contact information—we tested it. Everybody assumes that they're the worst leads, and they are, but they're only marginally worse—very, very little. We're talking like within 5% of the leads that fill everything out.
Also, we used to work with a structure where we had step one, and then we had step two of the form. Step two asked them a bunch of questions. Step one just asked them for the two. We had about 50% of the people that would fill out step two of the form. When we changed that so that step two isn't actually just step two, it's more like one step per additional question, we elongated it, I think, into eight or nine questions. We got 80% of the people that filled out the first step to actually go all the way to the end.
So the moral of the story here is the better you can do at asking for everything you need at the beginning while making that as little as possible, and then giving small incremental steps all along the way, generally you get a better conversion rate during that, and you get a higher rate, which you get everybody.
There's also a side benefit here to where if you get just the information you need at the beginning, you can then send that to your CRM immediately. You can start calling the lead while they're filling out the rest of your form, and that's the best-case scenario because after they fill out your form, what are they likely to go do? They're likely to go to someone else's website and fill out their form too. But if you have these questions that keep them engaged on your website, and you can call them before they're done with that, then you're kind of holding them captive for a period of time.
So sometimes the way we think of it is we don't even really need that second step of the form. Of course, some of that information is nice to have, but it's partially just a play to keep people engaged on our website while we have a chance to call them and hopefully get a hold of them.
Yeah, so I have a background in behavior and motivation, and so I think about this in terms of two principles. The first one is the foot-in-the-door principle, and the other one is the sunk cost policy. So people think in terms of what's the most efficient use of my time and energy. If you get them in your form with a simple ask, like an address, they're going to feel like they have to get to the end of the form to see what's at the end of it because they've already got their foot in the door, right?
And so it's easier getting them all the way through with one question than giving them ten at once. The other one is they see going through and giving their address and their number as a sunk cost, and now they're like, "Well, I've already put in all this effort; I might as well get through." But we know that they could just stop if that's a sunk cost, right? But they see it as, "I'm already in this deep; let's get all the way through."
And those two principles are why the single step per question works well. It's less cognitive load for people, and they're already invested, so they just get through it because they're curious and because they see it as, "I've already invested all this time and energy into this form."
Yep, totally agreed. Yeah, that's a better way to explain the science.
So the next principle that we wanted to cover, I believe, was page speed. I feel like this is equally over-talked about but also under-emphasized at the same time. I think that everyone loves to look up their page speed and say either it's amazing or it's terrible, and that's the end of the world. It also is a very impactful metric because people are increasingly less patient and want instant responses, especially on mobile.
So having your page load quickly does have a pretty big impact on how well it converts. The best ways that I've seen to optimize this—the first one is what you use to build your site or your page does matter because there are some builders that are optimized to be faster than others. For us, what we use is Unbounce for our page builder, and it's designed to be light and fast.
The other factor is what is on the page matters. If you have big images, heavy videos, or if your form is embedded, all of those things that have to load before the page can load are going to delay how quickly the bulk of your page can load. So if you have images, make them as small of a file as you can without dropping down the image quality to a point where it's garbage. Avoid video as much as you can, or if you do have video, there are ways to have it load after the bulk of the page has loaded so it's not blocking the page from loading.
And that ties into the next piece, and that's to order your scripts on your page. This is a little bit technical, but there are ways to only fire your most important scripts before the page loads and delay other scripts that are nice to have but you don't have to have in there immediately for the page to load. We could probably do an entire episode on that concept, but the basic piece of it is like what is in forms—only fire up front what has to be fired up front. Only embed what has to be embedded, and only have as large of an image as you have to have, and that's going to help your page load as quickly as possible.
The other piece is people love using the Google PageSpeed Insights checker for page speed, but I found actually the best site to use to check your page speed is the GTmetrix page. It takes in more data points and gives you a better output of what to change than the Google PageSpeed page. So just a plug for using a better page speed tool than the Google one because it does miss a lot of points and tends to overinflate the impact of certain metrics.
So that's page speed.
Yeah, thank you for that, Garrett. I would add, just for anybody who's listening to this and thinking, "What's the big deal if my page takes one more second to load?" Turns out it's actually a pretty big deal. The way that this works, there are studies that show this, and I don't remember the exact numbers, but incrementally for how much longer your page takes to load, the conversion drops way off.
I just want you to think, if you're listening to this, have you ever clicked on an ad and then just given up because the page on the other end never loaded? You probably thought that, like, "Well, maybe their page must have been slow." That page that didn't load for me took like 30 seconds, and that's why I didn't go. My page is way faster than 30 seconds, but the reality is that the page speed very much depends on the speed of someone's network or their 3G or 5G.
3G sounds like I'm like 10 years ago, but whatever it is—LTE, right? If we're talking more modern, it depends on how fast that is at that exact moment. So sometimes people do have issues that cause it to load a lot slower, and that drops people off.
So there are many studies that prove that even small improvements in page speed dramatically improve your conversion rate. It's the saddest thing ever to pay for a click and then have that person never see your landing page because it didn't load. Even if that happens to 5% of your traffic, that's 5% of your budget that you wasted, right? That's significant.
The other thing here is it actually hits you twice. If you have a slower load speed, not only do you convert less traffic, but also it's going to affect your quality score in the platform. The way it affects your quality score is that quality score is built up with three things. One of those three things is landing page experience, and one of the components that affects landing page experience is page speed.
So if Google detects that you have a slower page speed, what's going to happen is, compared to your competition, Google is going to consider your page a worse experience. It's going to make it so that you are required to bid higher for the same traffic than your competition does. Or the opposite could be true. If you have an excellent loading page speed, then Google is going to make it so that you don't have to bid quite as aggressively in order to have great placements with your ads.
So if you think about it, if you were to just model out what happened to the page speed with low page speed, you're going to convert less, and then additionally, your clicks are going to cost more. So maybe you're converting 10% worse, and you're paying 10% more cost per click. You just have a 20% hit to your campaigns based on what could be a half-second difference in page speed. So it makes a massive, massive difference.
All right, I'll talk about the next one. That was number four: page speed. Number five is bid strategy, and this might sound a little bit out of place in a landing page conversation. Bid strategy is wildly important. If we use the same keywords, the same targeting, everything with a manual cost-per-click strategy versus with our highly trained real-time bidding strategy—TCP is the one that we tend to use the most—we find that the traffic will convert at double the rate.
So why is it that you have the same target, yet you have twice the conversion even with the same landing page just based on the strategy of how you're bidding? The difference is that some strategies are a lot smarter, and they actually take into account tens of thousands of data points that Google has at each individual decision.
So a way that did not just the keyword that that person searched. A lot of times when we see people that think that landing page is their issue, it turns out they're actually converting fine for the traffic that they're sending to it. They're just sending not the best quality traffic to it, so that's causing them to have a lower conversion rate.
So this one's pretty quick, but I guess I will say that landing page conversion comes down to traffic quality, and traffic quality, even with the same targeting, could be different based on what bid strategy you're using. One of the best things we've ever done for our conversion rate is just use a bid strategy that helps us laser target the exact right traffic because if you're sending the exact right traffic to your page, there's so much more.
Yeah, no, I agree. It's interesting when I see a problem with how well a page is converting. Before I look at, "Do I change the button or the form or the order of the things?" I look at, "Did anything change in the traffic mix that was going to this page?" Because that does have a pretty hefty impact on how well it converts. You can have a perfect page, but send a bunch of garbage, like spam traffic, and it won't convert. I don't care how great the form is or your video that goes through your whole process.
So that's a missed key for sure. Cool. The next piece to go over is A/B testing. This is so important for being able to know definitively if what you are changing on your page actually is impacting performance or if it is just random outputs of leads or bad conversion.
What we mean when we say A/B testing is you make a duplicate of your page, make one change in the new version, and send half of your traffic to one and half to the other. It doesn't mean, "Let's have our page, let's change one thing, and see the before and after," because you're working with two different audiences at two different times, and other things have probably changed in the world, in your market, in your other campaigns that could be impacting how the before and after is performing.
That's why it's crucial to have those tests run at the same time to be able to definitively know if that change did or did not impact performance because you take out time as a variable from that comparison. An inexperienced marketer will say that very often your conversion rate goes way up or way down even if you change nothing.
So the fact that you're looking before or after you change something is kind of useless. You never actually know what made the difference. A/B testing, I think, is way underdone in this industry, and part of the reason it's really tough in this industry is because we just naturally have a sparsity of data. The leads are really expensive, and our budgets aren't massive.
For the average company, I once calculated that if they want to do a proper test on the landing page, it usually takes about nine months to gather enough data to know that variant A works better than variant B. Most people just don't have the patience to wait nine months. They forget they even started the test nine months later, and they probably end up wanting to test something else partway through that nine months.
One of the things that we do that I appreciate is we use software to basically aggregate the data across all our landing pages. We can make it so that we can change one thing across all those landing pages and measure how that affects performance. What that does, when we're working with over 200 clients, is it gives us over 200 times the amount of data that we're getting on that landing page test.
What that does is it basically cuts down the amount of time it takes. So we're talking going from 200 days to do a split test to probably about one day to pull off that split test. That is a massive advantage because by the time you finish your one split test, we already did like a hundred. Now, we probably wouldn't do quite that many because getting data is almost our easiest part, but figuring out what to test and executing that takes a little bit longer.
But the point is that the massive advantage gets way better data, and that leads me to my very last point here, which is in the process of optimizing for conversion rate, one thing that we have to be very aware of is that we want to optimize for quality of leads, not necessarily volume of leads. This is one of the biggest mistakes people make in this industry.
It turns out making a pretty high-converting landing page is not that hard if you're willing to just say anything that it takes to get the lead, whether that brings a good kind of person or not. But if you ask the right things, you can create a situation where you're generating the highest quality possible.
I'll give you an example. There are some landing pages out there that basically make it sound like they're CarMax, and when you get to this landing page, a cash offer is going to pop up at the end. If you do that, like if you tell somebody you're going to get this instant cash offer with the click of a button, for example, if that wording sounds familiar, you might know where that's from.
Then what do you think is going to happen to the quality of those leads? They're probably going to be people that you call, and they weren't expecting your call. Versus if somebody fills out the form and they're very much expecting to get called by an acquisitions rep, then it's going to be a different kind of lead.
So I guess what I'm saying is—and this is true with anything that you test—you have to be cognizant of how it affects things down funnel. It's just like when you're testing ads, and you get the ad that everybody wants to click, but then that landing page traffic doesn't convert. Or when you have the landing page that produces tons of leads, but the leads don't convert down funnel. Or you have the appointment-setting process that sets tons of appointments, but the appointments don't convert down funnel.
You have to think about all those individual things, and it's really common to split test landing pages and for the thing that you're split testing based on to only be how many leads are coming out of this. If you do that, what you're naturally going to gravitate towards is whatever landing page generates the most leads, regardless of quality.
What you need to do is you need to split test which landing page, per unit of traffic that comes to it, produces the highest number of qualified leads. That's how we like to do our A/B testing. Now you understand why I say it takes probably nine months to get that data because if you're moving one step further down the funnel, it always takes more time to get that data.
But we find that there are a lot of different changes that can make to your page that appear to be good and actually hurt you. So you're going to be doing tests, and you're going to be saying, "Our marketing is awesome because this is working better and that's working better," and you keep on optimizing, doubling down on those things, and then your ROI is going down and down and down, and you don't understand why.
It's often because you're sacrificing performance down funnel for performance up funnel, and it's so common on landing pages.
Yeah, I agree. It's easy to say, "Look at all these leads! We have a great cost per lead; we're killing it! We just cut our cost per deal by X amount!" When really, you brought in a bunch of leads that don't know what they're doing or what they agreed to or even what you offer. They just saw an easy form and went through it because there wasn't enough friction, and there's a cost to that.
So yeah, that's why we always aim to optimize for cost per qualified lead or cost per contract because then even if there is a spike in one part of the funnel, that's not going to change what we see as a win unless it also changes our ultimate KPI, which is cost per contract or, at the very least, cost per qualified lead.
Yeah, we do go into some different points on the webinar that we did on this topic. This is point two of that webinar. We'll put the link to that in the show notes, and in there, because it's visual, we'll actually show you the landing pages and show you some examples of these different things. I think you'll find that valuable. You can also get the toolkit there that includes some other resources to help with this.
So for anybody listening, thank you for taking the time to listen to this slightly more technical episode that makes a massive difference in campaigns, and I will see you next week.