The Art of Conversion Optimization
with Ryan Thomas
Ever wondered how conversion optimization works? My friend Ryan Thomas is here to explain it to you in full, granular detail.
Ryan is a strategist at CXL Agency, where he helps his clients build scalable, profit-generating CRO programs. In this episode, he shares strategies, techniques, and insights from the cutting edge of web optimization.
“Whether an experiment wins or loses, there are important learnings… you’re learning the customer’s mindset and the context of the business.” – Ryan Thomas
In the Podcast:
- Ryan’s background and how he got started in CRO
- What conversion rate optimization actually involves
- Common misunderstandings about CRO
- What Ryan’s day to day optimization process looks like
- Useful optimization tools
- The biggest optimization levers in your sales funnel
- Skills all CRO specialists should have
- Resources to learn more about CRO
- The future of web optimization and how it’s evolving
Links and Resources:
- Zoho PageSense
- Ryan Thomas’ Linkedin
- CXL Institute
- “Thinking Fast and Slow” by Daniel Kahneman
- “Lean Startup” by Steve Blanks
- “The Four Steps to the Epiphany” by Steve Blanks
Brad: Today I’m talking with Ryan Thomas. Ryan is a senior strategist at Peep Laja’s CXL Agency, where he helps ecommerce Legion and sass companies grow through conversion research and experimentation. So Ryan is very much in the trenches and on the cutting edge of conversion optimization, and I’m super excited to have him on the show today. Ryan, how’s it going, man?
Ryan Thomas: Oh, good. Yeah, thanks for having me. definitely excited to talk about CRM. It’s a really interesting topic. And I always enjoy just chatting with people and, you know, teaching people what I would know, because it’s like, there’s so many people in the industry that helped me along the way. And it seems in general, like an industry where people are really, like, willing to share their experience and to kind of like, be really, like, open and transparent about the process and, and everything that’s involved. Awesome.
Brad: Yeah, definitely excited to learn more about what you do and kind of your story. So maybe that’s a good place to start. Ryan, tell me a little bit about your background, and how you got into CRM in the first place.
Ryan Thomas: Sure, yeah. So my background is kind of like a mix of, you know, computer science programming, development kind of stuff. And then like the business and marketing side of things. So like a university, I studied a computer science business option program. So it was basically like, both of those, I would kind of bounce back and forth, like focus on the computer science stuff for a while and then get super into the business and take like all the business courses that I could and, and then get back into the programming side of things. So you know, in a way that kind of set me up pretty good for CRM, but that was much further down the line. So during university, I got involved with a company called Mosaic, which did tactical marketing programs for big brands, fortune 500 companies.
Okay, um, mostly, I was working for beer companies, actually. So like Budweiser, Bud Light was one of the main brands that I worked for. And also some like lesser known Canadian ones like Coke, and he was a big one and mad. Yeah. Nice. Didn’t think you know about that one. Good. Yep. And then some more like international ones, like Stella Artois, and stuff, they were all kind of under the same umbrella of companies. So it was kind of cool to work with them. And the cool thing about it was like, this wasn’t just like boardroom theoretical marketing. This was like, front lines, like the customer experience, like I would add, you know, community events, like music festivals, or you know, rodeos or soccer tournaments, stuff like that. And then also, like in liquor stores, or, you know, pubs and clubs, like doing actual, like promotions, where they’re trying to create, like a fun and positive and memorable experience for the actual customers.
So, right also, like a pretty good lead into CRM, where the customer experiences completely centric to the whole thing.
Brad: Yeah, yeah, definitely. So let’s talk about CSL. Yeah, conversion optimization agency, started by Pep Leia, who’s one of the, you know, thought leaders, I guess you could say, in the realm of CRM, whenever you start working at cx L, and tell me a bit about what you do, kind of on a day to day basis there.
Ryan Thomas: Yeah. So working with CSL was just about three years ago. And what led to that was actually taking the CX CRM certification course with you actually, which is how we met so people might not know that. And before that, it was like a much longer period of just like, following Pep as he was building the business. So like, I think when I started following him shortly after he started blogging about conversion optimization. And, you know, I kind of slowly learned some of the concepts involved in CRM, but I never really had like a high volume website where I could apply any of that stuff. So then when it came time to like, actually want to specialize in CRM, and then you know, the course came up and it was like a special coaching version of the course where we got to learn from Pep directly, rather than just you know, watching online courses and stuff like that.
So I thought it was a good opportunity to bring my kind of, you know, loose assortment of skills that are all really good for CRM into actually learning how to do it properly. And then that led into actually getting a role as an optimizer with the agency side of the business. So pretty, pretty lucky there. Not a lot of people get to go from zero to 60 So quickly in the industry,
Brad: pretty good payoff for joining a course.
Ryan Thomas: Yeah, yeah. If you look at the ROI, I think the course was no Canadian dollars, like 20 $500 or something like that. So compare that to that the annual salary of a full time optimizer. Pretty big multiple.
Brad: Yeah, maybe. Yeah. Congrats, man. So, Ryan, what kind of projects do you work on? Is there a particular kind of business vertical? Do you work with CSL?
Ryan Thomas: I’m sort of, I guess, yes, and no. So like, basically, the concepts that we try to apply our base are like scientifically backed, you know, processes and research and, and that sort of thing. So in that sense, it’s really portable to different industries and verticals and stuff. So there isn’t that much of a need to specialize. And so I haven’t very much in that sense. So it’s been a pretty good mix of e-commerce businesses, you know, Software as a Service, online learning, and just kind of more straight lead gen for, you know, more expensive products and stuff like that. But because of my technical background, with web development, and stuff like that, I’ve kind of just had a bit of a nice time working on more technical projects. So just like an exact example of Code Academy, which a lot of people are familiar with is, you know, a pretty big brand.
And like a really interesting project that we worked with for almost two years, or a year and a half, or something like that. So the things that had more to do with technical topics than I would tend to get put on those projects just because of that, my background.
Brad: So before we go any further, Ryan, let’s kind of define what CRM is, because I feel like there’s a lot of people who talk about CRM in different ways, and they’re slightly different perspectives and definitions of what it entails. So from, from your perspective, what does CRM actually mean? What would you define it?
Ryan Thomas: Yeah, I mean, the, the obvious definition is just, you know, the term itself, conversion rate optimization. So, you know, making changes to, to a website and sort of like a scientific way to maximize conversion rate. But I think that that term is even a bit outdated. And this kind of ties into, like, you know, where the industry is going and stuff like that. But I think it’s become less of such a niche thing, where you have just like this kind of solitary focus on incrementing, one metric, you know, make a tweak, and we want to get more transactions, we want to get more people filling out the lead form. So it’s, you know, as the discipline and the industry is maturing, the definition, I think is broadening more into, like, General experimentation.
So it’s still focused on like, on site optimization, like, we’re, we don’t really get involved with looking at, you know, the ads and, you know, SEO or, you know, so it’s, it’s narrowly focused in that sense. But it’s more encompassing the whole process, like a whole user experience, going through the various, you know, buying stages on the website, and then even to post conversion. So there’s So coming back to like, what’s the definition of sorrow? I would say it’s like, on site experimentation using like, proper, properly designed experiments that are properly planned out, using statistical methods, just, you know, the same methods that are used elsewhere in scientific research in order to create measurable improvements in the website performance.
Brad: Okay. What are some common misunderstandings you’ve seen out there that people have about CRM, going back to that kind of different perspectives that people have? Some people it’s kind of hazy what CRM stands for? So what are some of the misunderstandings you’ve seen that are common?
Ryan Thomas: Yeah, I think there’s still a lot of them. I think maybe the most, the most popular one is that it’s sort of related to even just the definition being so tightly focused is that we’re just focused on making these tiny little changes, like trying 50 different button colors. Like, I think it was an Amazon experiment where they did that. It’s like just making these tiny little tweaks or like meek tweaks, as they say, to try and get people to convert or using kind of, like manipulation and dark patterns to try and trick people into converting. So it’s, you know, properly done. experimentation is like, the incentives should be well aligned, as kind of like a bridge between the business incentives and customers incentives.
So like, you know, one of the best ways to improve the website’s performance is to make it a better experience for the customers. To make it easier and, you know, more fun or whatever for them to do what they already want to do, and avoid, you know, things like tricking them and get the friction out of the way and stuff like that. So, yeah, that’d be kind of like, the two main ones, I think, Okay.
Brad: Do you ever work with clients, and they sort of have like a preconceived notion about the kind of experiments they should do. And you have to be like, Whoa, back up, we have to start with the process, we have to start with the research. So I feel like that’s one thing that I’ve seen people think, like you said, buttons, they jump to, oh, we need to test the buttons, or we need to test XYZ. And it’s like putting the cart before the horse. So yeah, yeah, that must be a process. Right?
Ryan Thomas: Yeah, there’s, there’s a pretty common thing with clients. And it’s. So I think one of the main things is like the hippo problem, you know, the highest paid person’s opinion, tends to dictate things too much. And this happens, even when, you know, we get involved with the client, gone through an entire sales process to tell them exactly what we do, how we approach it. They’re paying good money for this program. And then once we get in there, there’s just people from different departments or from higher up just throwing ideas, and they’re like, we want to test this, we want to test this thing. And usually, it’s because they saw their competitor do it, you know, they’re like our competitors doing this one thing on the website, we want to do that as well. So at least they want to test it, like that’s a win in itself. They’re not just like, trying to roll out those changes.
But you know, the, the idea of either best practices or just copying competitors is sort of one of the main things that is difficult to work with. And so there’s often like a training process in the beginning of like, you know, trying to get like, because you want to balance that with, you know, maximizing the reach of the CR o program. So we want these ideas coming in from different departments and from different stakeholders, but we need to, like filter those through the process, and prioritize them and get them involved in just even that process of prioritization so that they understand why like, okay, sure, that’s an interesting idea. But we have these other ideas that are backed by research, or for whatever reason, are higher priority for us to run with right now?
Brad: Yeah, you know exactly what you mean. experienced that myself. Mainly, what I create for my clients is copywriting. Yeah, like a sales page, landing page, whatever it is. And I share it in the form of a Google Doc, right. And sometimes, it’ll be early in the process, and they want to see what I’ve created. And I’m like, sure, just to give him a sense, even though I’m still working on it’s still in process. And there will be notes and edits from like, for other people who weren’t even in the original conversation on the slide. Yeah. What am I supposed to do with that? Yeah, so yeah, it’s always a fun, fun and interesting part of client work is dealing with that.
Ryan Thomas: Totally. Yeah, had a similar case where we were putting together the plan to run a customer survey. And, you know, we have a lot of like, like, a lot of really good reasons for the way that we ask questions and which questions we ask. And it’s not just stuff we came up with, like this is drawing from like the experience of very skilled people in the field. You know, people that like, say LSA, or to like, you know, she focuses on that almost exclusively like how to, like ask questions for qualitative research, how to do user testing, like all this kind of stuff. And, you know, so we learn from these people, and we understand the psychology and we have specific things we try to avoid doing, like avoid leading the customer too much, trying to get open ended feedback.
And, you know, so we put together our survey document, and then random people that we’ve never even heard of in the company are making suggestions like, Oh, you should change this to this question. And anyway, in some cases, they probably know what they’re doing there. Maybe they are actually UX researcher and stuff, but they’re still detached from the process of like, our goals, you know, the thought process that went into crafting those questions in the first place. And then so then we’re just in a Google comment thread trying to explain Well, actually, we do this, you know, somebody you’ve never even met.
Brad: Yeah, and it ends up making everything take twice as long as it shouldn’t. But that’s just the nature of the beast, right? client work? Yep. serene. Let’s talk about your process. I know cx l has a framework that they use, they even share that path is shared that widely on the internet. And I’m sure you probably have your own more in depth more maybe customize day to day process that that you use with clients. So I’d love to learn a little bit more about that. Like, where do you start? What’s step one, what’s step two, And what’s kind of the high level overview of what your CRO framework looks like?
Ryan Thomas: Yeah, so, I mean, it starts, it always starts with research. But you can even kind of back it up a bit from there. So like, the beginning of the client engagement, before you start doing any research is you need to, like, understand what the client’s goals are. And like, you know, where is there? What’s their experience with their experimentation? What’s the status of their experimentation program currently? How mature is it? You know, how diversified? Is it within the organization? Like, what’s the culture of experimentation, like? So that’s, like, accomplished through, like a client intake survey, and then like the CRM maturity audit. And so those are kind of the things that we take into account in advance, just to even figure out how to structure the program, like, should we do, you know, just research for this for this company? Or is it like research and redesign of some pages or epic to do experimentation.
So there’s, like, you know, important work to be done, you can’t just assume that you can jump into research and testing. Although many people do that some people just jump right into testing without research, which is, you know, much worse. But, you know, once we have those things figured out, and assuming that it’s like, our most typical engagement with the client is, is that like a round of research, like a deep dive into multiple data points, that is used to build out a testing roadmap for an experimentation program, like ongoing monthly retainer type of thing. So, you know, the research itself is what the, you know, CFL has published the framework for they call it research Excel. And it’s basically like this molecule with a bunch of different branches on it that represent the different research methods and data sources that we pulled together in order to get like sort of a unified list of insights. And sort of that process.
Brad: What are some of those sources? What are some specific points on that graph? That molecule?
Ryan Thomas: Yeah, so it’s like a balance of qualitative and quantitative research. And both are kind of equally important. And they work really well together. So on the quantitative side of things, you have like analytics, so like, there’s kind of two, or maybe even three, depending on how you look at it different aspects to just analytics research. So the one would be like a technical tracking type of focus, like, it’s actually like, not uncommon to get pretty far into a research project, and then realize that something that you thought was user behavior was actually just a tracking anomaly. So something is wrong with the way that the analytics is set up, and it’s making it look like people are accessing this page tons of times, but you know, maybe the snippet was loaded twice on that page. So numbers are just being inflated or something like that.
So it’s important to look at that early on, like, have a good look at the tracking and do and do an audit just to make sure that things are set up properly. And then also that they’re making as much use out of their analytics platform as they should be. So there’s like a lot of features in in Google Analytics that are very powerful. But you know, out of the box, when you just install the snippet, none of that is set up for you. So that’s kind of like phase one, analytics is tracking technical issues. And then there’s conversion focused analytics audit. So this is like a more thorough process of like, looking at the user journey from, you know, where they came from, to landing, the behavior that happens along the way, like, you know, different events that can happen interacting with different elements, using Site Search, that kind of thing.
And then all the way through to the actual conversion, or exits if they exit before that. And you know, finding drop off points is a really kind of good initial thing to look at, to drill into more to understand where you can make improvements that will actually move the needle and increase conversions. So that was a sort of long winded explanation of analytics. But other quantitative sources are things like heat maps, scroll maps, much tracking. So people don’t really think of that as quantitative, because you’re not really looking at numbers. But it is like a bunch of aggregated data on Yeah, where people are clicking and stuff. So it’s like in practice, it’s almost more qualitative, because you look at a heat map and you’re like, Okay, people are spending more time over here, like, you know, this thing is drawing their attention and their clicks and stuff like that.
Brad: Do you guys have a particular tool that you use?,
Ryan Thomas: um, well, hot jar is the most common because it’s pretty good. pretty inexpensive, and, you know, easy to get up and running and stuff like that. But we don’t worry. We try to be as tool agnostic as possible. And kind of have to be like when you’re in an agency setting and working with dozens of different clients who have varying tool stacks and No, we want them to own their tools. So even if we’re recommending something, we just try to act as like a neutral advocate, we’re not like, you know, we don’t have a horse in the race or anything. So, you know, we can work with anything but hot jar is by far the most widespread one. Next, most popular probably be like, who like if they’re using that as their testing platform? Yeah, then that also has pretty good heat map heat map functionality baked in.
Okay. And also, you can like, if you’re running a test, you can see heat map comparative between control and variant, which is like really useful is like, getting for getting more insights into a test. Yeah. While we’re on the topic, though, page sense. So like Zoho, they have a testing tool called page sense. And it also has heat map functionality. And it’s actually more sophisticated than I think any other one that I’ve seen. So it’s like, a lot of heatmap tools are just like, create, like a pixel grid image, like they actually use just like a base, basically a screenshot or a snapshot of the page, and then just overlay the click data on top of it. Yeah. But it’s pretty limiting. So like, you can’t see it. So you have like a drop down menu or hover effects and stuff like that, like you really can’t tell if they’re interacting with things or with those things, or how they’re interacting with them. But where’s like page sense. And there’s other tools that have done this, but I can’t remember them off the top of my head.
The it’s actually like the HTML element that is like the clicks are associated with. So like, there’s, there’s even like a browser extension where you can look at the live site, and then you see the click data overlaid on something. So if you open up a menu to see the clicks on every single menu item, and stuff like that, that’s super cool. Actually, I discovered Zoho pretty recently. They’re actually an Austin company. Oh, yeah. They have an office there. Yeah. And
Brad: I signed up for one of their services for social media. So they’ve got like a whole suite of different apps for marketing and optimization and stuff.
Ryan Thomas: Yeah, yeah. I know one of the product managers, and he’s based in Chennai, India. And they’re like one of the biggest offices in that city. And they have over 5000 people just in that one office, like working on on product development and
Brad: Oh, wow. Global. Yeah. pretty huge. Even host their own, like conferences and stuff like that. Nice. Yeah. So Ryan, what are some of the biggest leavers when it comes to optimization? Where’s usually I know, all websites, and all sales funnels are unique. But where are some of the best places to start? If somebody wants to begin optimizing, initiate the process?
Ryan Thomas: Yeah, so it’s, you know, when answering a question like this, it’s tempting just to jump straight to like, what are some actual things you can change to to get some, some quick wins, and I’ll touch on that in a bit. But it really comes down to like, the best way to have an impact is to actually understand the customer and the context. And so that, again, ties back to, to the research that you do. And, you know, I went into like, quantitative quite a bit, but the analytics and quantitative side of things can only tell you what is happening on the website. So if you want to understand the why of it, that’s why you need the qualitative side of things. So like customer surveys, on site polls, user testing, like this is these are the kind of methods that you need to fill in the rest of the story.
And like, actually, understand as much as you can, like, what’s going on in the customer’s mind, like, understand their actual motivations? You know, why are they wanting to do this in the first place? Why did they come to this website? What are they? What problem in their life are they trying to solve here? And then what are the roadblocks that happened to them along the way, like both, you know, physical, physical, like, you know, still digital, but I mean, like, friction points, like something is broken, or there’s something that’s, you know, just not working properly. It’s like a weird customer flow, you know, so friction types of things.
Brad: Yeah. Like the UX actual Yeah. interaction with the site itself.
Ryan Thomas: Yeah, exactly. And then the more psychological aspects, so like, what are the things that maybe they find confusing, or what are their hesitations? And, you know, fears, uncertainties, fears, uncertainties, doubts, I think, is the, the common way of putting it. Yeah, the feds so. So that’s really like, if you can come to understand that, like, understand what’s happening on the website, understand why these things are happening, why the customers are making the decisions, what’s holding them back. That’s, you know, that’s gonna set you up for success. Like that’s, you know, there might be a one some little things where you can, you know, do some quick runs, like, one example is a lot of sites have sliders like carousels on the homepage. So be aware. Yeah, it’s the most important real estate on the website.
And for whatever reason, the company can’t decide what’s the highest priority thing, but they’re, so they’re like, let’s just put five things there and just have them slide out of the way after a couple seconds. Yeah, and so it’s, you know, it’s distracting. And, you know, the data from these types of things ends up showing that like, the more things you put there, the less you know, it’s totally logical, the less likely anyone is to see any of it or notice any of it, because they just can’t consume all that information. So like one of the, one of the most reliable tests to do is to replace that with just static hero area that is focused on the actual value proposition of product or service or whatever it is, and listing and user benefits written in customer centric reader centric copy, not, you know, we do this, it’s, you know, or just like, bland, descriptive product copy, which is what everyone seems to go with these days, which I’m sure you as a copywriter see that all the time?
Brad: Pretty much the default? Yeah. It’s out there. Yeah.
Ryan Thomas: Yeah. But even that, like, I’m calling that a quick win, you know, ditch the slider, go with value prop. But in order to know what to write there, and in order to know which end benefits are important to the customer, you need to have done that research first. So pretty much always comes back to that.
Brad: Yeah, I totally agree with that. I mean, I don’t do CRM, but there’s a lot of overlap, I would say, with what’s the kind of research that you do for CRM and the research that I do? For copywriting. I take kind of a research oriented approach. It’s at least like any project that I do. At least half the time is going to be spent on research and nice. Yeah, what I’ve found is the most effective kind of writing the most effective kind of messaging is the messaging that your market is already using. Yeah, having to invent creative prose, creative words, you know, to show some kind of skill or verb as a writer, it’s like, You’re, you’re echoing back what the people are already saying. Because, yeah, find the places where they’re having the conversations already on the internet. Yeah, this is like forums and testimonials, surveys, like you guys do.
Yeah, long as you can find where those conversations are happening. You can see, like, how they feel about their product or their problems and their challenges and the solutions that are out there. So yeah, just like you’re saying, it’s really about approaching Echo, echoing what the markets are already saying, and not so much. inventing, copying and measuring out of the ether.
Ryan Thomas: Yeah, yeah, exactly. A lot of, I don’t know, maybe novice copywriters. Or maybe that’s not fair. But it seems like a lot of copywriters are trying to focus too much on being like witty, or, you know, clever and stuff like that. But yeah, like you said, like, find out the words that the customers actually use to describe their problem or to describe the solution. And it’s like they’re giving you the answer. And that’s one of the reasons that we look for open ended feedback when we’re doing customer surveys, because then we can get, you know, not just voice of customer to figure out what they will say without having to choose from a list of options. But you know, what words do they actually use? And? Yeah, super powerful. Yeah,
Brad: yeah, I think there is, this is a distinction that I think I kind of make these days that I ever recently kind of realized or discovered myself is that sort of transparent, mirroring type language that you use that you find through surveys and things like that, you want to use that absolutely on the front end. And that is kind of the focus of the front end, like a landing page or a web page. But then on like, the back end, so to speak, the back end, more like email marketing, you can sort of get more creative, you can get more like, yeah, you know, storytelling, and you can kind of have fun with it more, and you still want to hit on the themes, right? That talk is about in this kind of conversation. But you can get more creative with it.
And like email marketing, or social media, but the front-facing stuff like a web page, that’s where you want to have that kind of transparent copy that is pulled directly from research, from surveys from online forums and things like that. That’s what I found. That’s kind of how I see things more these days.
Ryan Thomas: Yeah, makes sense. Totally. Yeah. And it’s like no more, like emails or more long form copy like you have, you know, this is kind of a challenge everywhere, but you also need the reader engagement, which, you know, you can’t just Like, you know, dispassionately list a few of the words they said like it, you know, you need to also like, you know, this is a person, they have an attention span they have, you know, interest and stuff. So there’s, you know, obviously a lot more to it. But
Brad: yeah, yeah, that’s definitely the foundation is that research? Mm hmm. So Ryan, when it comes to the type of person that is good at CRM, or, you know, is inclined to to become a CRM specialist, are there certain kinds of skill sets or mindsets? What do those kinds of people have?
Ryan Thomas: You? Yeah, I would, I wouldn’t say that. There’s just one I’ve, I’ve seen and worked with people from like, lots of different backgrounds who are very good optimizers. I say like, the two categories of those are, like one sort of, like technical side of things, which is me, mostly, I like to think I have diverse skill sets as well. But you know, I come from a technical background. I’m quite familiar with web development, and the technologies that create the web. So I understand how websites are built. And just like, you know, paired with enough knowledge of like the other side of things, which are called say, like psychology, more like UX and even design kind of skill set. So I think like, no matter which one you’re in, you need some of the other one.
But I’ve seen people do really well, who are from, you know, both of those kinds of backgrounds. And a lot of it is, I mean, either way, you need to also have sort of an analytical mindset in this is like, you know, shifting more into the mindset question rather than the skill set question. But it kind of, it can be a skill that you develop as well. Especially, like, say, the statistics, like it’s, you can separate out optimization from statistics, and you don’t have to be like a PhD in stats, to be good at optimization, like you, you just need to know enough, like, you need to kind of know the basic statistical parameters, what they mean, how to interpret them. But you know, that’s something you can even just build as you go, you don’t need to, like, already be super good at that.
But yeah, in terms of like, mindset, I’d say one of the biggest things is being able to put your own kind of biases and preconceptions aside. And, you know, look at what the actual research and data are telling you. Even if it goes completely against what you expected, you know, you should be happy to be to be shown to be wrong, you know, because like, either way, if an experiment wins, or loses, there’s important learnings there, and you’re learning the customer’s mindset, and the context of the business and how all these kind of factors relate to each other. And it can be like, you know, I can work with clients for months and months, and then I’ll just like, sit down and think about it for a bit, and then get to an even deeper level of understanding like, it’s, you know, it’s kind of important not to just be going through the motions.
What’s really important to be good at is actually understanding what’s going on like the website, the customer, the business, and like so that can you know, that you can keep getting more and more of that as you go. So being able to have the mindset to be prepared for that is really important.
Brad: Yeah, hundred percent. It’s like the, like the scientific method. Mm hmm.
Ryan Thomas: Yeah. Exactly. And then also, there’s another sort of facet to it, which is like not related to the doing of the optimization itself. But it’s like the soft skills around, working with people in different contexts. So like, you know, communication, you know, transparency, accountability, those kinds of things, being able to navigate the politics of your own organization, if you’re like an in house Yarrow, or if you’re in the agency, setting, all all the different clients that you’re going to work with, like everyone’s going to have different, you know, or charts, different team setups, different sort of cultures, when it comes to like experimentation, like a lot of them will just like AI experimentation as an idea.
Brad: But then when it really comes down to it, they’re still super resistant to like letting you run tests, which happens way more frequently than most people would think. You know, like,they hire c XL and then say, hey, wait, actually, we don’t think we want to run an experiment.
Ryan Thomas: Yeah, exactly. Sometimes it’s literally like it takes a long time even to get anything off the ground. And I’d say like, if I had to pinpoint it, the one is kind of the management or executive level people just switching directions wanting to try out something new. And you know, dictating the flow too much and not giving the the point person like are we always have like a point of contact, who’s basically the project manager on the client side. So the most successful programs are when that person is empowered to, like make the call of like, yeah, we can run this experiment, or not, or whatever. Not like they’re just in an interface for a committee that is ill defined. And yeah,
Brad: find off by four people every time.
Ryan Thomas: Yep. And then the other kind of common Roadblock, that that’s tough to navigate is designers. So which I can understand, like, if if you have like a designer who has created the website, you know, that it’s like their baby, they put so much work into this and effort and, you know, they’ve, you know, done a lot of their own research, they have good reasons, they created a whole, you know, branding, like visual identity, and all this kind of stuff. And so, you know, I can understand where they’re coming from. But they can be really, if they have too much pull in the experimentation program. And if they don’t come from an experimentation, background or mindset themselves, then they’re going to be super resistant to anybody messing with their baby. So it can take a long time to just just to just be like, okay, we’re just running an experiment, like, I know, it doesn’t look exactly how you want it to, it’s not pixel perfect.
But we’re just going to try this and then see if there’s a business case, to putting more resources into it. And like, you know, and then once, if you can get through that cycle a few times of like, sure, this maybe looks a little bit ugly, you know, it doesn’t look how you want, but the customers really respond to it, you know, we have like a nice boost and conversion rate here. Now it’s time to implement it. And they can do that final polishing when implemented. So if you can get through that cycle, and get them to kind of trust that like, yeah, they’re still going to have the final say, when it comes to what gets permanently put on the website, then you can kind of gain some traction. But it’s, you know, these kind of sort of politics, internal and external, it’s a big part of the success of the program.
And so like having the skills to be able to deal with these situations, lots of different kinds of personalities, people have different motivations and different objections and stuff. Like, that’s super important as a skill set, or mindset.
Brad: Yeah, definitely. And you do, you’re involved in the sales process, too, right. Like, Frank. Yeah, on. So those sorts of soft skills are crucial. I mean, it’s, you know, for me, it’s being able to sell it, is having that ability.
Ryan Thomas: Yeah, and it’s also really interesting, doing both, like not just being sales or just being the practitioner, because when you’re, if you’re the one selling the project, and then you’re also going to have to do the work. I think it’s a good way to kind of keep you honest, as a salesperson, you’re not going to sell them something they don’t need, because then when you’re delivering that you’re going to be the one being like, Well, you know, we’re not getting results or, or whatever. But on the line, you can’t oversell it. Yeah, exactly. Most salesmen do not implicate myself here. Anything. In the past life and past career.
Brad: Yeah. It’s an interesting dynamic in that world.
Ryan Thomas: Yeah, I think it’s, it’s good if incentives are aligned. And if you’re the same person, then they’re just automatically aligned. So yeah, you avoid that sort of separation. And then, you know, the other side of things is, every experiment you do, it’s, it’s another sales process. Like, it’d be nice if it was so easy. Like, I sometimes envy people who do in house experimentation, where, you know, they just have the authority to be like, this is what we’re doing, you know, we’re running this experiment, and they get to make the call. And, you know, because for me, like every experiment, I pitch, you know, I have to communicate the background.
Like, here’s the research that went into it. Here’s why we think this is a good idea. Here’s your whites an opportunity. Here’s the overall strategy that this tactic fits it fits into. So yeah, a lot of sales. Maybe I should have said sales is the other pillar of the skill set. But that makes a good optimizer.
Brad: Yeah, yeah. For any marketer, sales is part of it. So Ryan, what are some of the best resources out there for somebody who wants to learn more about CRM, whether that’s books courses, online content?
Ryan Thomas: Yeah, I guess I’m a little bit biased, since I learned it from cx L. Well, I guess I would say I’m biased because I work for cx L. But I guess, you know, looking at it the other way around is like I work for CXL, because I learned from CXL. So you know, I initially learned from, you know, pep in the blogging and as he built it, you know, built the CXL Institute, you know, learned a lot from that. And then that led into wanting to take the course. And so that was like a really good resource, like, I don’t know of a better like, CRM certification program out there. And then, you know, the CXL Institute has grown since I started like quite a bit. And they’ve been adding really high quality content from World Class practitioners in different disciplines and not specific to this aerospace at all.
Brad: Amazing How many courses he’s published on the website? Yeah, on the platform, like, there’s so many. And there’s more like, every week, it seems like.
Ryan Thomas: Yep. Yeah, exactly. So yeah, like, you know, definitely have to mention that as a good resource. There’s lots of other ones out there, though, obviously, like, I don’t have personal experience with them. But there’s lots of like, even some of the same practitioners that do see Excel courses, they do their own courses, or for other platforms and stuff. So, you know, generally speaking, online courses are really good. Because this is not the type of stuff you can learn any university program, you know, like, you can learn some good, like general marketing concepts and business and, and stuff like that psychology and university, which is great to have as a background, but when you’re trying to do like leading edge, stuff like this, then you need stuff that was published, you know, within the last year or two from people who are actually doing it, not just, you know, people talking about it from a theoretical perspective.
So, um, but yeah, other than, like, the online courses, I didn’t personally learn a lot of my CRM stuff from books, although I did have a general interest in psychology, you know, for a long time. But I think like the one that was pretty influential, there was like Thinking Fast and Slow, by Daniel Kahneman. Just like just a really good kind of, like, concise listing of a lot of the, like, cognitive biases, and you know, psychological patterns and stuff like that, which is, you know, you don’t want to misuse these obviously, like, it’s, it’s pretty easy. The more you know about psychology, the more you can sort of get adventure into dark pattern territory and try to trick people. But the same concepts can be used in a white hat sort of way to just, you know, help people accomplish what they’re trying to accomplish on websites.
And otherwise, in terms of books, also not specific to CRM, but I found a lot of value in Steve blanks work like lean startup and four steps to the epiphany. And, yeah, it’s like, you know, pretty well known in the marketing world. But that, for me, that was like a really good lead into CRM, because it’s like, one of the concepts that he talks about quite a bit is that like, getting outside the building concept, so like, you know, you have your team, you have all your ideas, you know, you do your brainstorming, and you know, even research and stuff like that, but until you actually get your offer out in front of people, and so you see how they react to it. You know, it’s just, it’s just a bunch of like, you know, theoretical stuff. And so I see, like, online experimentation as like, just a microcosm of that concept. Because like, if you’re just a web designer, web developer, and you just get to build something and launch it, and then that’s it.
You know, you don’t have this kind of extra constraint like actually having to test it. So when you’re, you know, formulating ideas and like a more granular way, putting them out there and experiment, then, you know, you get your answer really quick, you’re getting outside the building, the customers themselves are telling you, you know, what they think of your idea. And you do that over and over and over again. So it’s So yeah, those kind of general startup ideas I found were like a good kind of basis for experimentation just because of that kind of interplay.
Brad: Yeah, yeah. Very cool. I prefer working with clients who test what I send them to test the copy, not all, probably a minority of the clients that I work with do. But I like knowing, you know, knowing that it’s working, and then it’s having an effect. Suddenly, they’re totally happy just to like, get the copy, get the work and publish it. And they say, hey, it looks great. We’re gonna roll with this. But the people who are always like innovating, trying new things, experimenting, that’s when you really learn, and that’s when your understanding of marketing gets better and better. That’s, that’s what excites me. So
Ryan Thomas: yeah, and like copy experiments from my experience tend to be pretty good. Like if I had to choose, you know, copy versus like design, go with copy every time like if you can, like, like we discussed before, tweak the copy, use the words they use, understand where they’re coming from their motivations, like those tend to be really good experiments. And even like, just something as simple as button copy, if I had to pick like one kind of low hanging fruit experiment to run. It’s thinking about the button copy because like, so many websites will just go with, you know, get started. And no matter what stage they’re at in the funnel, it’s like on the homepage, okay, get started, gives you no idea of how many steps are after that. What is the next step? You know, this like, so uninformative. So
Brad: yeah, good point. Good point. Right? What do you think the future of CR o looks like? How is it changing?
Ryan Thomas: Um, yeah, it’s I think it’s definitely maturing. So like, even in the few years that I’ve been actively involved in CR O, it’s gotten a lot more well known. Like, it’s not just such like a niche thing that hardly anybody knows about. So it’s, it’s getting more like widespread awareness. And acceptance is an important part of the marketing process. It’s becoming more integrated into the businesses, you know, marketing growth teams as a whole, it’s not so much like there’s a CRM doing experimentation, and everybody else is doing their thing. It’s more like experimentation as part of the like, the normal risk management process of like, Hey, we have a new feature we want to launch or this change, we want to do, let’s put this into experiment, gauge that effect and then circle back as a feedback loop.
So I think that’s like a really positive thing. That’s, that’s been, you know, getting more and more prevalent, and like the clients that I work with and elsewhere. Otherwise, I think it’s evolving in. So it’s, like I alluded to earlier, it’s not just like, let’s focus on just the conversion rate. It’s more like running experiments, and then looking at more down funnel long term metrics. So you know, if you have like a subscription product or something and you want to experiment, maybe you can get lots of people to convert right away. But then that increases refund rates, or, you know, the people that’s subscribed in this experiment, they only continued for two months, and then quit or something like that. So the more you can fill out the picture, when it comes to the metrics that you’re analyzing, the more effective you can be as an optimizer.
And so I see a lot more like more of that. And there’s even like, like Microsoft launched their XP platform, which looks at like hundreds of metrics. So like, they just take it to a whole other level, like not being focused on one thing. I mean, you need, you know, the opposite side of that is like, you need to know what’s the most important metric? Because otherwise you just get bogged down in this, like, endless debate of like, well, we increase this, but this went down. And Is that good? stuff? Like, the signal for the overwhelming amount of data? Yeah, yeah, exactly. Yeah. And then like, I’ve seen that, especially in, say, subscription products with a premium type of model where it’s like, okay, we can get more signups. But fewer of the signups converted into paid or converted into a trial, and then more of the people who did the trial actually converted into paid or something like that.
So it’s like it happens, like quite a bit actually, where it’s like, one thing goes up, the other thing goes down, the other thing goes up. And sure, the one thing is the dollars. But getting more free signups increases the people that you can remark on it to. And so it gets pretty complex. So it’s, it’s good to invest a lot of thought in that upfront, basically. Yeah. Yeah, and other otherwise, like trends and zero, it’s kind of hard to leave out the AI side of things. So there’s a lot more like tools out there that incorporate AR, they use machine learning to, to run experiments and stuff like that, like there’s entire platforms where, you know, you just plug in like MBT like multivariate testing, you plug in a whole bunch of variables, you know, let’s try these different layouts combined with these different value props and these buttons, or whatever. And then just let the AI kind of sorted out and try out a bunch of different combinations and shift traffic towards crusherasia.
Brad: variations and runs traffic to them.
Ryan Thomas: Yeah, exactly. Which, and I don’t see it as I don’t see it as like, as valuable as a lot of people do. I think like I, it seems to me that a lot of the AI tools are kind of like shiny, new toys for the executives to get excited about and throw some money at. But, you know, it really comes down to it like you, you need the learnings. And even if you think of it as an AI thing, where it’s trying so many different combinations, it might be making decisions off of just like a handful of conversions. And deciding, well, this one did better, let’s ship more traffic to it. But if you actually like, you know, package some of these things together as a proper experiment that you ran as well, of course, then you know, a lot of those fluctuations will kind of even out or, you know, trends can reverse some going the wrong way.
Brad: That’s a good point. You’re, you’re gonna miss out on that level of insight and understanding like the why, okay, yeah, this word was converted better, but what was the hypothesis? What was the reason behind them, if you just outsource it to an, you know, computer system and you’re never gonna have that level of insight?
Ryan Thomas: Yeah, and a lot of them are even just designed that way that there isn’t like an experiment that you run and then finish and then and then learn from, like, their ideas. Like you just plug in these factors and it’ll just keep trying things and you just leave it running like it’s You know, so I think there’s a place for that I think it’s most valuable in the context of like, say, like, finding the local maxima, rather than, like, you know, so like, I would, I would focus on like a human process of like actual research and experimentation to find the kind of like macro changes that you can make. And then when you want to just like get the last little bit out of it, then plug in a few more variations into the AI system and just let it run to see, you know, do that final kind of level of optimization. Yeah, but yeah, I don’t see it replacing what I do anytime soon.
Brad: Yeah. Cool. Ryan. That’s all the questions that I had for you. Joy talk, I learned a lot, man. Is there anywhere that people can go to connect with you to learn more about what it is that you do?
Ryan Thomas: Yeah, I’m basically just LinkedIn. Yeah, just find me on LinkedIn. I’m always happy to chat about CRM stuff for digital marketing in general. I don’t really have like a personal website set up at the moment. So that’s kind of where I’m most active.
Brad: Okay, sounds good. Thanks, again, enjoyed it. And hey, I’ll talk to you soon. Ryan. Take care.
Ryan Thomas: Sounds good. Thanks for having me. All right.