Polished Geek Founder & CEO Deb Cinkus was honored to be invited as a special guest expert on the Fix The Convince marketing podcast (episode 4, February 2020).
Deb joins host Paul Mosenson of NuSpark Consulting to talk about Conversion Rate Optimization (CRO) for your website: What CRO is, why it matters, and how to get started if you are new to CRO.
Listen to the podcast now and read the full transcript below.
Fix the Convince. Welcome to the Fix the Convince Podcast. Here is your host, NuSpark founder, Paul Mosenson.
Paul Mosenson (00:08):
Hey, welcome to the podcast, Fix the Convince. I'm Paul Mosenson. We have some interesting topics today and we're going to talk about this concept called conversion rate optimization. You've heard of it. I know you have. Here's the thing. You're spending all this money on this brand-new, beautiful website, multiple pages, tracking analytics and you're spending all this dough on SEO and advertising and media and content marketing, et cetera, et cetera. There's our marketing plan. We have everything listed here except one thing. Isn't the goal to drive leads in sales on your website? Right? How do you know you're getting this many prospects to click and submit as you should be?
Paul Mosenson (01:00):
Well, there's one thing that you have to keep in mind here in this whole digital marketing realm is CRO, conversion rate optimization. It's easier said than done. There's an art and a science behind it for sure, like any marketing strategy. And this really needs to be not taken lightly. Right? It's really important to do everything you can to drive a prospect to a form or a purchase button to click submit or whatever your CTA is to generate leads and sales to your company. If you're not optimized for conversions, they're going to leave and go with a competitor. You don't want that. Do you?
Paul Mosenson (01:49):
So there's a science behind this and that's why we're talking about CRO today with my special guest. And she is Deb Cinkus and she is the founder and CEO of Polished Geek, which is a boutique digital marketing and web development agency, headquartered in North Carolina. They've been around since 2009. Deb and her team at Polished Geek help clients improve their website and marketing ROI, with consulting and expert implementation services in marketing automation, website personalization and, of course, conversion rate optimization. Deb is a CXL Institute Certified Optimizer, a credential awarded to those who successfully complete the CRO industry's most rigorous training and certification program for conversion rate optimization professionals. So she's a specialist and we're glad to have her on the show today. Hi, Deb.
Deb Cinkus (02:45):
Hi, Paul. Thanks for having me.
Paul Mosenson (02:47):
Sure. No worries and thanks for being available. And hopefully, we get lots of listeners and conversions. Right?
Deb Cinkus (02:55):
Paul Mosenson (02:55):
That's what we're talking about today. So I'm going to ask a number of questions and I look forward to some good conversation. A lot of executives may have heard of a CRO, but how do you define CRO exactly? What does it really mean to you?
Deb Cinkus (03:14):
Well, there's the short answer and the long answer. So the short answer is that, in a driven, scientific way, look at what you're getting as far as conversions on your website or your marketing, whether it's email marketing or something like that. You really analyze everything and you apply a scientific process to figuring out what you could possibly improve and then taking steps to improve it and then seeing where you are with that. How did it work? Did it not work? Then taking what you learned and putting it back into that process again to continuously drive the metrics in the direction you want them to go. That's the short answer.
Deb Cinkus (03:53):
The long answer is there's about six different specific areas of analysis we do whenever you're going to do a real CRO program and really get into the meat of it. And I'd be happy to go through what those all are. There is the web analytics, of course. That's kind of the foundational thing. You can't improve what you can't measure. We all know that. So with analytics, we can show that there's goals and trackable events and that you know what people are clicking on and you have a baseline. You have to have that first.
Deb Cinkus (04:28):
And then the next thing that you do is you do some technical analysis. Look at the performance of your website, speed, are there a lot of broken links laying around? And one of the things that we always find when we do this is we call dead ends or funnels to nowhere where you actually get someone to convert. You get them to fill out a form or you get them to download a lead magnet. And then the thank you page is basically, "Thanks a lot, bye-bye." You don't give them the next thing that they should maybe be interested in. You don't engage them. You don't keep them on the site. So that's some of what we look for in technical analysis.
Deb Cinkus (05:06):
And then there's mouse tracking. There's things like looking at heat maps, how far down people scroll on the site and where they click on the site. You can use your session recordings, which anonymously records. Just basically, you can sit and watch how the mouse moves around the screen and where people went and where they click. That's very interesting. It's kind of funny sometimes when we watch some of those recordings. We see what we call "click rage" or mouse rage. Have you ever been using a website Paul, and you get frustrated? Have you ever noticed that, involuntarily, you're kind of moving the mouse around rapidly in the area where you're frustrated? Actually you can see that in the recordings.
Paul Mosenson (05:48):
You know where that comes? I know a lot of companies do it, is when they're trying to close out those popup boxes for lead generation and conversion. Right?
Deb Cinkus (05:57):
Paul Mosenson (05:57):
So it's like a damned if you do, damned if you don't.
Deb Cinkus (06:00):
Yes. That happens, but you'll see people ... you can tell they're confused, by the movement of the mouse, around form or around maybe a headline statement or something like that. So some stuff like that that we can see. So that's some of the mouse tracking that we look at. And we do something we call heuristic analysis. That's the official CRO term for it, but basically it's a site walkthrough by an expert. Someone is walking through the website, putting themselves in the shoes of the customer and saying, "Would I be motivated by this? Am I confused by this? Are there things on this page that don't make sense or are actually in conflict with what I would expect? How difficult is it for me to find what I came for?" So it's someone who's been doing CRO for a long time. They've looked at a lot of sites. They've studied best practices and such. And so they begin to look at the site from that perspective.
Deb Cinkus (06:53):
And then there's user testing, which is ... You take like third-party people. You don't take the established customers. You don't take the staff at the client site. This is separate from those. And this is where you get maybe eight to 10 people who sit down and you give them a couple of defined tasks and you get their feedback. You get them to verbalize and say, "Well, I didn't expect that here. I would've thought this would be here." You get them to give you their honest feedback and it's interesting. You don't have to do this with 100 people. You can do this with 10 and get so much valuable information. So that's really good for the top of the funnel. People who are coming to the website, what's the first impressions, what are the things that people tend to think.
Deb Cinkus (07:41):
And then the last part of the analysis is, actually, more of the mid and bottom of funnel and that's where you're doing qualitative surveys and analysis. You might put a survey widget on the website and ask people, "Did you find what you were looking for today?" Right? "Did you achieve what you wanted today?" We try to stay away from that clichéd one of, "How do you like our website?" Because you're not going to get as much valuable feedback. You can put stuff like that on there. Even people who maybe don't want to do that or maybe they want to do that and also pursue other things, there's usually a lot of qualitative information you can get. It might be they use an online chat, you can go through the chat logs and then you can do some analysis on what are the themes you're hearing? What are the objections you're seeing? What are the things people are asking?
Paul Mosenson (08:34):
Well, there's a lot of elements here and you got into the weeds, but it's interesting and just really tells the audience there's a lot involved here, but what's the bottom line is the messaging that engages you to go further. You look at all of that. You look at the buyer personas and there's so much involved, starting with the messaging and then all the way through call-to-action buttons and engagement buttons because you want to drive people from interest to desire to action. Right?
Deb Cinkus (09:09):
Paul Mosenson (09:09):
And we go through all of that, scientifically and a little bit of art, to optimize that whole journey, I guess we'll call that, to, "I'm here on your website. Do you want me to leave or do you want me to stay?" Right? Isn't that the whole idea?
Deb Cinkus (09:26):
Paul Mosenson (09:27):
Everything has to be convincing audiences to click on that submit button.
Deb Cinkus (09:36):
Right. Check out and buy that item. So that's why you look at things from all these different angles. Six of those things that I briefly explained you, look at all of them. And you pull them together because they give you a very good picture of all the things that you could possibly do that will drive that behavior better. So some emotional marketing feedback that you can get qualitative stuff. It's a little bit of psychology. Yeah, some of it's technical, but you get so much input from looking at it from all those distinct angles and then you end up with this huge bank of ideas. And these are all the things that you can begin to try, whether you actually do formal AB testing or whether you're just going to try changing it on your site and then measuring it the next month and seeing if it made an impact in the direction that you want.
Paul Mosenson (10:29):
So when you're a high-level executive and you're listening to this show and, obviously, your big responsibility is growing revenue and growing sales and things like that, how would we communicate to clients about the importance of CRO? I know it can be, I guess, overwhelming, but it's important. Right? Because you have hungry mouths to feed. You want to generate more revenue for your company. Don't think the web designers know what they're doing. There's a lot of factors here. So what's your advice if you're convinced a company said, "Yeah, we need CRO. We need to include this as part of our marketing program," what's your advice to get started?
Deb Cinkus (11:23):
Well, I mean, you kind of hit in on the head in the intro, Paul. You did a very good job of summing that up. You're spending all this money everywhere else. You're investing in marketing automation, perhaps, email marketing, you're doing content marketing, you're investing in paid ads and you're doing all these things. You're spending all this money to bring people to the website and if you don't do the other part, which is work on improving that site, optimizing your conversion rate, getting more qualified leads. If you don't do that, you're throwing away a high proportion of that money. Now, it's that leaky bucket thing. That analogy of you have a bucket, it has a lot of holes in it. You keep pouring water in the top. That's your paid advertising and all of the marketing efforts that you're doing, but you've got a lot of leaks because your conversion rate optimization hasn't been done on your site. So you're paying all the money to bring them there and then you're losing them.
Deb Cinkus (12:18):
So it really is about optimizing, not just making your website better and more pleasant to use. It's not just about that. It's about making sure that every marketing dollar you're spending, you're getting maximum value out of it. And so if someone is new to this concept and you're thinking about doing this, one of the first things that I tell people is, well, we've got to make sure we've got some baseline metrics. Like I said earlier, you cannot improve what you can't measure. Right? So we've got to have the ability to track what people are doing on the site. We've got to have analytics. That's the first thing you're going to have to pay some attention to.
Deb Cinkus (12:58):
And a lot of websites we come across have been built without that in mind. They've been built to look awesome. They've been built to look great, but there's nothing being tracked, unless maybe you're talking about a checkout. That's a challenge. Right? You've got to get these things in place so that we can tell where are we now and then where do we want to go?
Paul Mosenson (13:21):
Well, when you're in B2B. It's a little more complicated, depending on your business model. Right? Because you have to define what a conversion is if you're going to have the benchmark at all. Obviously, the easy one is generating email addresses on a website, whether it's get a quote, download guide, whatever, a form. It's a form. Right?
Deb Cinkus (13:40):
Mm-hmm (affirmative). Right.
Paul Mosenson (13:42):
But with other tracking, you might ... What I do is we build in the email submission tracking because that's another engagement, if you want to treat that as a conversion versus a form or if you're a business that relies on phone calls, track that, as well. I think all those things, together, define what a, quote, "website conversion," is. And I think, for the audience out there, they have to understand what we're talking about here is you call it lead generation, but it's really inquiry generation for B2B. Right? Because you're giving your salespeople an opportunity to qualify those inquiries that come in. Because everything we talk about with website and landing pages, at the end of the day, whether it's get a quote or download guide or sign up for free assessent, whatever it is, they're really mostly at the inquiry stage until they become qualified and that's where nurturing and the other things come in.
Paul Mosenson (14:39):
But I just wanted to bring that out there because the word conversion could also be on the sales side, which is qualified lead to close. Right? Which, it's not really a CRO thing. More of a sales consultant type activity or sales management, but we are talking, in this conversation, a conversion which is a focus on what happens on a website.
Deb Cinkus (15:05):
Right. Exactly. And the formal terminology is micro and macro conversion. The macro conversion might be the big stuff you actually really want. "Give me your email address. Fill in this form. Convert and give this lead to my sales team." That's a macro conversion. That's what you're aiming for. There's a bunch of little micro conversions that happen first. Right? A person might watch a video on your website. You can track what percentage of the video they watched. They might download a free whitepaper that doesn't require... it's not gated. It doesn't require them to give an email. There are things on the site that you still also want to track that we call micro conversions, which are little steps of taking them in the direction you want them to go, little things that happen before they turn around and fill in that form and give you their email address.
Deb Cinkus (15:50):
So making sure that you can track all of that because it's not just about that one thing. And the other thing, too, is that you want to think of quality versus quantity. No matter how good your sales team is, they only have so much time. Right? Addressing here B2B lead gen like you just did. They only have so much time. So giving them five times as many leads aiming for quantity might not actually help your business if all those extra leads are low-quality leads and they're not really well-motivated, they're not really a good fit. It's not about just getting more people to fill in the form. It's getting more of the right people to fill in the form. And that's one of the reasons why we track those little micro conversions that show us the behaviors that tend to be linked to: this is a good prospective customer. This is someone that is very likely to want to buy from us.
Paul Mosenson (16:46):
Right. So we talked a lot about B2B, but of course eCommerce is a whole different thing, but a lot of it is similar, at the same time. It's still persuading people to perform an action. Any of your thoughts there regarding B2B and eCommerce?
Deb Cinkus (17:07):
Well, eCommerce CRO is certainly easier to measure. I mean, it's very obvious. You begin a CRO program and after a month or two or three of doing that program, have the number of orders gone up? Has the average order value gone up? The number of repeat customers. You've got that immediate quantitative thing of, "I have the exact numbers from the checkout." B2B lead gen is more fuzzy. Some of the customers we work with have sales cycles that go on for months. Maybe they're selling an item that's very expensive. It costs $20 thousand or more. It can take awhile for them to actually close that deal. It's harder, too, because a lot of times lead gen, B2B businesses, don't necessarily want to share their sales figures. So it's not as obvious as what you get on an eCommerce site.
Deb Cinkus (18:02):
The process is still the same. I mean, we still look at all the same sort of things. We still approach it with the same methodology. We still approach it with the same mindset. Doing analysis and research, getting into the mind of the customer and finding different things to drive whatever the metric is we're trying to improve. The end result of if we drive sales is a little bit harder to measure as a CRO agency, anyway. It's harder for us to measure when we don't have eCommerce checkout, the immediate gratification here of, "Hey, look. Sales are up."
Paul Mosenson (18:36):
Yeah. It goes back to eCommerce. There's obviously a number of elements. You talk about micro conversions and macro. Do you go into cart abandonment? There's different elements. There's ...
Deb Cinkus (18:51):
Paul Mosenson (18:51):
... product to cart and cart to purchase. Right?
Deb Cinkus (18:54):
Paul Mosenson (18:55):
And optimizing those little steps. And also, you talked about average order value. So that, to me, would mean cross sell, upsell, or optimizing the site to make sure people are buying the right products. But, go ahead.
Deb Cinkus (19:10):
Right. No, you're exactly right. I mean, and that's one reason why you have to make sure that, when you're doing CRO, you don't focus on just one metric. Don't only look at increasing the number of orders. It's kind of, again, back to that quantity versus quality, because I've seen programs where someone puts all their effort into increasing the number of orders and they do all these things that result profitably and more people putting something in their cart quickly and moving straight to checkout and getting to the checkout quickly because they're not encouraged to actually browse around. You don't have any upsells, like you just mentioned. Average order value ends up going down. You have to look at a whole bunch of metrics. There are plenty of examples online of successful CRO tests where one metric went up, but a metric that is more fundamental to the success of the business, like revenue or average order value or repeat orders, actually, goes down.
Deb Cinkus (20:11):
So you have to be able to look at all these things and not just be, with blinders on, looking at one metric because it all has to fit together. That's one reason why you want to make sure that you approach CRO not just with all of that techy stuff I talked about earlier in the beginning, but with strategy. You've really got to understand the business. You've got to understand how the business is measured. What are the things that you want to achieve? Those are very important. It's not about just getting in and doing the techy part.
Paul Mosenson (20:40):
Well, you have to have a ... You talked about benchmarks and part of that is, especially with eCommerce, but B2B as well, is what's the value of a customer, average revenue per transaction, those kinds of things. Because I know I have clients who kind of ideate, kind of get stuck on the concept of offers and, for instance, free shipping, which is expected, but also may cut into your margin. Right?
Deb Cinkus (21:14):
Paul Mosenson (21:16):
And so all these kinds of things are decisions. Right? And can be tests. Right?
Deb Cinkus (21:22):
Paul Mosenson (21:24):
It's funny because we talk about emotion in CRO. Right? And a lot of it's like data, data, data, this and that, but people buy with emotion and that's a whole nother, bigger topic, but at the end of the day you'll buy something which makes you feel good. And if that means you got a discount even though you could've afforded the regular price, it's still the difference between purchase and bye bye. Right?
Deb Cinkus (21:53):
Mm-hmm (affirmative). Oh, yeah. Absolutely. You're right. You mentioned things like free shipping. And you're right. Free shipping has been shown repeatedly to be a big motivator for people checking out. Actually, it's funny. You can do tests and you can take a product that costs $10 that has a $5 shipping charge. And you'll sell less of that than if you have a $15 product with free shipping. It's just the way it goes. And people want to feel like they got something and people really resent paying for shipping, nowadays. So that's one of the fundamental things to test, but it also just depends on the type of product and how fast people tend to want it and stuff like that, too. There's also some product where people are willing to pay for shipping and they're actually comparing it more as a commodity price.
Deb Cinkus (22:46):
So every business is different. So there's some things that are kind of like truisms that we think we all know, but you never do know until you test it. You never know for sure until you look what's going to work for this business and their products and their customers.
Paul Mosenson (23:02):
Yeah, for sure. I agree. I mean, everything is individual. Right? And sometimes, if you want to position yourself as ... let's just use the word upscale, or whatever you want to use. Right? We have value here. We don't need to discount. Our product is good. If you need to buy us because of our approach and how we solve problems, things like that. But at the end of the day, it's good to test, as well. I know the other thing I just wanted to bring up ... I know we're getting into weeds, but here's an example of probably a popular test you do. And again, going back to eCommerce, but also it could be with B2B commerce, is the concept os testing dollars off versus percent off. Right?
Deb Cinkus (23:51):
Mm-hmm (affirmative). Right.
Paul Mosenson (23:52):
And what does that mean to people? Example.
Deb Cinkus (23:55):
Yeah. That is also a very common test. And there's no hard and fast rule, again, how well it works. Sometimes people will respond more to they have $100 product and I'm giving 10% off. Sometimes people will respond better to seeing it presented as $100 item, 10% off today. Some people will respond better to $100 item, save $10 today. Same amount of money, but different audiences respond differently. So you've got to test and see which one works.
Paul Mosenson (24:35):
Yeah, for sure. Yeah. I mean, it's people's mindsets. Right?
Deb Cinkus (24:41):
Paul Mosenson (24:42):
And how they work to feel like they're getting the deal. CRO, as you mentioned earlier, covers so many different things. You've got the ... whether it's user experience, engagement, tracking conversions and B2B. We talk about web forms and testimonials and proof points. We do all these crazy things. It's all about, really, trust and proof. Right?
Deb Cinkus (25:11):
Paul Mosenson (25:12):
More than anything. If you want to do offer, you throw that, as well. Trust, proof, offer. Oh, I like that. That might be a new slogan I can trademark. Right? But there's a lot ...
Deb Cinkus (25:27):
You heard it here first, on the podcast, live.
Paul Mosenson (25:29):
That's right. Right. Fix The Convince, that's what we're talking about here. But with all these different elements and stuff, what's the process? How do you decide what to do first? And to me, I tell people, let's get this done first, before I even start marketing. But is there a process of what comes first? Or is it customized based on the client?
Deb Cinkus (25:56):
Well, it certainly is ... I mean, all of this is customized based on the client. What's the current situation? What is it you're trying to achieve? What kind of business model do they have? That sort of thing. It's always customized to the client. And to go back to a structured process, this is where we introduce CRO as a science and an art. And it absolutely is both. The science helps you, again, with that prioritization. What do we do first? Now, I did mention earlier that web analytics is the fundamental. We have to do that first. We have to make sure that's all there and we can track everything because, otherwise, how do we know whether or not what we're doing is even helping?
Paul Mosenson (26:35):
Right. Setting benchmarks.
Deb Cinkus (26:37):
Yeah. You've got to have those benchmarks. Absolutely, Paul. And let's say you've done all of that analysis in all of these different areas that I've talked about; the technical analysis, your click and heat maps, interviewing prospective customers and you've looked through your chat logs for trends. You've done all those things. You're going to end up with this huge list of ideas. "Ooh, we could try this. We could change that. We should make this more obvious. We should have a page for this." There's just a ton of things that you end up coming up with. You end up with this big bank of ideas. And you have to figure out, well, which ones do we do first? Which ones are most likely going to provide bang for the buck? Because you can't do everything at one time.
Deb Cinkus (27:13):
And so people who are formally trained in CRO have a prioritization framework that they use. And I won't go through the whole thing here, but I can tell you that what we do is we have very objective criteria and we have a whole set of questions that we kind of rank every single idea against. And some of those questions might be things like, how easy is this going to be to implement on the site? How fast can we actually get this done and actually see the change? Is this something that's going to take a day or two, or is it going to take a week? How difficult is it? How obvious is this change going to be to a user?
Deb Cinkus (27:50):
So say we've seen in our heat maps, for instance, and our scroll maps, that only 25% of the people get down to the bottom of the page. And if we're talking about making a change near the bottom of the page, we need to be realistic about the fact that only 25% of the people are probably going to see that. The things that are above the fold have higher priority than things below the fold. [NOTE: "above the fold" = the very top portion of a webpage, visible before a user has to scroll] And another thing that we certainly look at is, do we think this is going to reduce friction on this site and make it easier for people to do business with us? Is it more likely to increase their motivation so we rank against that? And then as an agency, and I'm sure you can sympathize with this, Paul, there's a lot of times you have awesome ideas. Because they require resources and it's your client, your client's busy, they're running their business, they're doing other things, they have a lot of internal projects, sometimes it can't necessarily produce the content that you suggest go on that site quickly. Maybe they can't produce that lead magnet idea that you had.
Deb Cinkus (28:47):
So one of the other ranking factors that we personally use is, does this particular idea require much effort from the client themselves? And if it doesn't, it's a higher rank than if it does. Because you know - time. Right? What can we as an agency run off and do, versus we have to have the client dedicate some resources and time to producing something. And of course, we can have resources to do that, as well. It's not like that can't be an additional service, but that's not generally ... The idea isn't to come in with CRO and then start adding a bunch of costs. The idea is to help, as quickly as possible with a scientific approach, prove their lead generation for their sales on their eCommerce site.
Paul Mosenson (29:33):
I mean, you bring up a point, though, of the tools and everything, what's involved. Right? Because even for B2B, optimizing conversions ... Well, you know what? Maybe if your call to action is a webinar or whitepaper, well, guess what? Maybe it's not problem solving enough. Maybe it's the power of the message that is not driving as any leads if you're a lead gen advertiser. Right?
Deb Cinkus (30:01):
Paul Mosenson (30:02):
And it goes back to the whole thing with buyer personas. It's a whole nother story. And targeting the needs of the buyer and this and that. Listen. If that's the situation, well, my team and NuSpark, we build those content assets and update landing pages. Right? And maybe you're more features than benefits and you need to get more benefits. That's all part of the content part of the CRO, which is a whole nother story, but it's all critical. It all works together. Again, going back to the decision to click the submit button or not. Right?
Deb Cinkus (30:38):
Mm-hmm (affirmative). You're right. You're right. There's missing elements that you need to have on the site, things that increase motivation. Those are the things that generally take longer. Sometimes it's simple stuff. It's changing wording on a button, but most of the time it's much deeper than that. So that prioritization framework I mentioned just helps us put some objective criteria around, where do we start first? And the stuff that scores the highest is the stuff we start with. And we just keep working our way down. And as you continue to do CRO, you continue to generate ideas, you continue to come up with better things, and you just keep adding them into that bank of ideas and you keep ranking them and making sure that you're always working on the thing that you believe is going to give you the best bang for your buck.
Paul Mosenson (31:23):
Yep. We're kind of winding down here and I'm going to ask you another question because everybody brings this up and they throw it out, like a term that is common, but may be misunderstood or not: AB testing. Right?
Deb Cinkus (31:38):
Paul Mosenson (31:40):
Why don't you explain that and examples of that in a little more detail about how that works? Obviously, with benchmarks and things like that, but just give an overview of that, if you don't mind.
Deb Cinkus (31:52):
Okay. All right. Well, first thing to know is CRO does not necessarily require AB testing. You don't have to do AB testing. I'm not saying it's not a great idea. It is a good idea, but you don't have to have it. Whereas, you cannot do good AB testing without the CRO research because, otherwise, you're just randomly picking ideas and just testing stuff for the fun of it. Let's change this color of the button. It isn't structured and it's not going to get you results. You're just kind of throwing things against the wall and see what sticks. So you have to have CRO research to do AB testing.
Deb Cinkus (32:28):
If you have low traffic on your website or maybe you just don't get enough visitors that you can get some statistically solid AB test results, you can still AB test. You're just going to have to make some assumptions from it. Or you can still do an awful lot of CRO improvements measuring your metrics against your baseline without doing formal AB testing. What formal AB testing is, for those who might not know, is basically presenting a portion of your traffic with another version of your site or page, whatever. So let's say a hundred people come to the homepage. If you're AB testing the homepage, fifty [percent] of those people are going to see the one you have today, the one that's generally available to the public and what Google sees. And 50% of them are going to see another variation of the homepage. They're going to see it maybe with different wording, they're going to see it with maybe different calls to action, maybe they're going to see even a different image or headline, different stuff.
Deb Cinkus (33:30):
Then what you do is, you measure, in detail, everything that happens on that page. We set up what's called goals in the AB testing tool. When we track those goals, we even track how engaged people stay on the site after they've seen the test version versus the control version of your existing site. And AB testing just gives you insights into the behavior of your customers and the parts of the site that are working in a way you don't get any other way. Because you are measuring things on a really granular level during the test and it's randomly assigned by the computer which group someone goes in, you get something that's really scientifically valid, as long as you test long enough, with enough people.
Deb Cinkus (34:14):
And here's an example of what I mean by CRO without AB testing and CRO with AB testing. So let's say I'm doing some CRO research and I come up with an idea and I've decided I'm going to change my site. And I do that and, 60 days later, I go back and I look at my analytics and I have driven my form completions up. I'm just going to pick a number out of the air, 4%. Not a very big number, but let's say 60 days later form completions are up 4%. Here's the thing. I changed my site entirely for everyone. So if I change a bunch of things on the pages or whatever, do I really know what drove that 4%? How do I know that some parts of what I changed didn't increase the positive behavior of my prospective leads and customers by 10%, then another part [of the page] actually was negative, minus 6%? And so I ended up with 4%.
Paul Mosenson (35:11):
But Deb, isn't that why we just test one thing at a time?
Deb Cinkus (35:14):
Well, you don't have to just test one thing at a time. In fact, that's pretty slow. You have to have a lot of volume to do that. If you have enough traffic volume, if you are Booking.com, if you are Amazon, if you are all these big players in the industry that have tens of thousands of people on their site every day, yes, you can test one thing at a time. You're going to get results really fast and you can do that.
Deb Cinkus (35:37):
The kind of businesses we tend to work with don't have that high volume of traffic. We tend to test a set of things all together and then measure every single thing on the page so that we can actually get results and knowledge faster out of that test, so that we can tell, hey, this part contributed a positive percent. This part was negative 6[%]. Let's keep the positive and let's rethink what we can do with this part that didn't do so well. So you just take that and you just reiterate into another test. You say, "We're going to keep the part that worked and not the part that didn't."
Deb Cinkus (36:11):
Without AB testing, I still made a positive direction. I still increased my form conversion, but I don't really know exactly why. I know something changed that people liked. That's still good. And if you don't have enough traffic to do AB testing, that's not wrong.
Paul Mosenson (36:29):
No, we just layer in best practices and then the learnings about what can be implemented.
Deb Cinkus (36:40):
Very, very true. Now, I know that some people in your audience might be listening to this and saying, "Hey, I'm really interested in doing this and I've heard so much about AB testing. Should I make sure I just hire someone who advertises themselves as AB testing firm?" And my advice would be, if you talk to a company that says that, first week of working with you, they're going to start running AB tests on your site and they aren't going to do any research first and they aren't going to do any prioritization and they aren't going to sit down and really learn about your business and your strategy and work up a plan, I would say run because AB testing in a silo, without that knowledge and analysis and research and just throwing things at the wall and seeing what's sticking is just random, and people getting excited about the tool. And that's not strategic and that's not going to drive your business in the right direction.
Paul Mosenson (37:39):
Yeah. I hear you. Now, listen, we all know there's tools out there, free tools like Google Optimize and, "Oh, cool. I can play around with my headline and see what happens." Right?
Deb Cinkus (37:52):
Paul Mosenson (37:53):
But if you're not scientific with a benchmark and a plan and a strategy, the tool is just a tool.
Deb Cinkus (37:57):
Paul Mosenson (37:59):
Optimizely and all these other tools out there. It's not what the tool does. It's how you plan it and make recommendations on it with enough data. Otherwise, you're just playing games.
Deb Cinkus (38:12):
Right. Well, it's like if you hand a hammer and some nails and a saw to someone, if they're not a trained master carpenter, I don't think I'm going to get a very pretty table out of it. You've got to have the skills to wield the tools appropriately and you've got to know when to use each tool. And one of the other things that happen sometimes is when people start just AB testing with blinders on and they're just focused on, "Hey, we're just going to drive this one metric and we're going to do whatever it takes to drive this one metric," you may do that and may actually end up having side effects that hurt business, long term. So you could do things that maybe increase the number of orders in an eCommerce cart and your average order value goes down. You end up with more orders to satisfy, more operational overhead to satisfy those orders and get them out the door, but you're making less profit.
Deb Cinkus (39:08):
Now, it does need to be done as a wholistic program. And AB testing, you know, just dragging those tools out of the toolbox by themselves and putting them in the hands of someone maybe who doesn't know all of the approaches that need to happen to make sure that you are testing, measuring, optimizing in the right direction, now that's actually dangerous in some ways. So that's why I wanted to caution your users who are maybe like, "Hey, I'm excited by this topic," make sure that you have someone who's going to work with you strategically. That is the most important thing.
Paul Mosenson (39:42):
Yeah, for sure.
Deb Cinkus (39:43):
The strategy is there.
Paul Mosenson (39:44):
Strategy first. That's what we do.
Deb Cinkus (39:46):
Paul Mosenson (39:48):
Well, this is a good conversation. I just wanted to wind it up. Any last minute thoughts to an executive who's considering this? Obviously, their revenue and profit are at stake here, but anything else you want to comment on as far as a CRO, in general?
Deb Cinkus (40:11):
I would just say that the right time to do CRO is when you know that you're doing well. You're a business owner, a leader, or a marketer, you're already successful, you're already doing well, you're already doing a lot of the things that you should be doing. You're doing content marketing, marketing automation and these sort of things, you know you can do better, you're just not sure where to start. You're not sure exactly how to drive things in the right direction and move the needle. And this is what you bring in when you're ready. CRO is honestly not something for the impatient or the desperate. This is not the kind of thing you bring in when you're really dealing with some serious cash flow problems or something like that. It's not a magic bullet. It is an investment in your business to improve and to become more successful with a process, with a science, with an art. That's when it's the right time to do it, when you just know that you're doing well, but you also know that you can be doing better and you just are ready for that.
Paul Mosenson (41:19):
Yep, for sure. Well, Deb, thanks for joining me today. Really good conversation. Hope everybody hung in there because this is pretty serious stuff. You can't take it lightly. I mean, this is your business and, like any business, whether you're a CFO or CTO or whatever, what's people's jobs? Is to improve the company's efficiency. Right? In their department. Save money, build profit, different departments, sales, operation and, of course, marketing. And if you're not always optimizing your marketing, including websites, landing pages, order value, et cetera, you're missing the boat and you've just got to think about this. Think about it. And if you have enough data ... We all want to grow. Right? And grow further. This [Conversion Rate Optimization] is a tool that needs to be part of the plan.
Deb Cinkus (42:23):
I agree with you. It does need to be part of the plan. Totally.
Paul Mosenson (42:28):
All right. Well ...
Deb Cinkus (42:30):
Well, thanks, Paul.
Paul Mosenson (42:30):
I appreciate it.
Deb Cinkus (42:31):
I had fun this morning. Thank you very much. This was a great conversation and I appreciate you inviting me on as a guest and letting me share some of that perspective with your audience.
Paul Mosenson (42:43):
Sure. Well, thanks for listening, everyone...We'll look forward to the next show. All right. Thanks for listening. Bye.
Deb Cinkus (42:53):
Thanks for listening. Subscribe to get more marketing optimization insights. Fix the Convince.