How to Increase Your Website Conversion Rate (3 Free Fixes Anyone Can Make)

The Vault Unlocked

Sahil Patel, CEO of Spiralyze, reveals how most websites are built on gut instinct rather than data, and explains how A/B testing—used in me

Key takeaways

• A/B testing is the scientific gold standard for validating what works, just like double-blind medical trials.
• Most websites are stuck in a 'pre-scientific' era of design based on templates, trends, or personal preference rather than data.

Main topics

• The scientific method applied to web design
• Why templates and gut feelings fail for B2B SaaS websites

Notable quotes

"Your best guess about what converts is still just a guess."

"We're measuring probably less than 100 to 200 years of recent history. That's how people took medicines, right?"

Conclusion

Sahil argues that businesses must move from intuition-based design to d

Transcript preview

Speaker 1 (0:00) Most founders run their website the way people used to take medicine before medicine was a science. Somebody's neighbor tried it, it worked, so you copy it and hope. My guest today doesn't play that game. Sahil is the CEO of Spiralize, and he built something called Predictive CRO, studying what's already won and lost across hundreds of companies before betting on the pattern. We get into why that happy stock photo on your homepage is probably costing you leads. why your actual product beats any lifestyle shot you've ever paid for, and a two-minute test using nothing but Google Translate that tells you if a stranger can even figure out what you sell. If you built your site off a template and a feeling, this one's going to sting. This is The Vault. Let's unlock it. Speaker 1 (0:54) So, Hale, welcome to the show. Thanks for having me. I'm really glad to be here. All the way from Spiralize the business itself. Now, I know before the show we were talking, I'm very excited because this is such in the marketing department of business and A-B testing. And I know that what you guys are doing is revolutionary and it's quite interesting. So I'm excited to get into it. Before we do, why don't Speaker 2 (1:22) you Speaker 1 (1:22) tell us a little bit about who you are and how we even got to this point? Speaker 2 (1:25) Yeah. So my name is Sahil. I'm the CEO at Spiralize. We're an A-B testing company. I'm the CEO. I'm a second time doing the CEO thing. So I've been doing different kind of SaaS tech type companies my whole career. And I live in Atlanta, Georgia. Speaker 1 (1:44) Okay. Well, I mean, how'd you get into... into Spiralize? Tell us where that came from. Speaker 2 (1:51) Yeah. So it's a fun story. So I am actually Speaker 1 (1:54) not the founder. I was the founder CEO of my previous Speaker 2 (1:57) company, which was a SaaS business. And I was actually an early customer of Spiralize. So around the time Spiralize was just getting off the ground, I started using the product. And I saw that it really worked. And it opened my eyes to the power of any... Let's see. the power of A-B testing as a way to improve the efficiency of your app, of your website, of your landing pages. That's how I became a true believer. I sold that business. I was thinking about what's the next thing I want to do. I had stayed in touch with the founder here. He had kept building the business organically, self-funded, was looking for a partner. It was great timing. It also was great because they... I lived in Atlanta already and Spiros is based in Atlanta. So it was a good arrangement. And... I had done the solo CEO founder thing wearing both hats. He had been doing it along the way. We both knew how hard that was. There was this mutual understanding of it. And we also had, there's some things he's really good at. There's a different set of things that I'm really good at. If we did them together, we thought that would work really well. And it's worked out really well. I've been doing it now for four years. Speaker 1 (3:08) And we are using words like A-B testing. And I'm just assuming there might be people listening that... Like what is even AB testing? So let's take a step back here and go, okay, let's talk about what is it? What are we even talking about right now? Let's take this. Speaker 2 (3:22) Let's even step away from like marketing websites, business. Let's just start with something probably everyone can relate to, which is when you have a new medicine, how do you make sure the medicine actually works? Maybe you have a cancer drug, cancer treatment. Speaker 2 (3:43) you could just guess. You could give it to a bunch of people and then say, did they get better? But you want to make sure it really works. So what do you do? You get 100 people. You give half of them what's called a placebo. It's a sugar pill. It's nothing. And you give half the people the actual drug and you measure. Speaker 2 (4:06) how many people got better. Speaker 1 (4:08) And what you want to make sure is that, say you had 100 people, 50 people, make sure Speaker 2 (4:12) men and women equally distributed, young people, old people, big people, small people, all of those things. Because you want to make sure that the difference wasn't because you put all the men in one group and all the women in the other, and it turns out the drug works really well for men and not so well for women or some other variable. That's called randomization. That is an A-B test. And it's the gold standard for measuring how effective something is doing. It's not the only way to do it. You can do it before and after. It's not as good. This is the gold standard. Something called this double-blind controlled experiment. Fancy words. A-B test. Same thing. Speaker 2 (4:50) So that is, now we accept that as the best way to know if a medicine works. If you went to the doctor today and she said, hey, I've got something for your cold. Speaker 2 (5:08) here, take this. And you looked at it and you're like, Speaker 1 (5:11) huh, Speaker 2 (5:12) I've never heard of this. Like I was going to take some Robitussin, Speaker 1 (5:15) but Speaker 2 (5:16) what is this? And the doctor's, oh, it's, it's really good. You might ask, well, how do I know it's good? Like, oh, I gave it to my neighbor. Worked great. Okay. Is your neighbor, they have the same symptoms as me? How do I, how much should I take? Oh, just take a swig. Okay. Like, is it effective at like two tablespoons, one tablespoon once a day? Like, well, I bought, trust me, I've been a doctor for 10 years. This will work. You would never do it. Okay. Or at least I don't think you would. You would hope not. Okay. This is now back in the early days of medicine. Okay. And by the way, that time wasn't that long ago. Okay. We're measuring. probably less than 100 to 200 years of recent history. That's how people took medicines, right? That's the days of patent medicine. People standing on a shoebox, hawking their Sawhill's patent cough syrup. No one would do that now. You would never take a cancer drug that hasn't been tested yet. Speaker 2 (6:31) Like there's the pre-scientific era to medicine and then we're down in scientific era to medicine. Yet for people's websites, they do that all the time. They do the equivalent of, I'm a doctor. I've done this for 10 years. I don't know. I gave it to my neighbor. Worked great for them. Should work for you. All the time. You look at websites. Why'd you do this on the website? Oh, it's best practice. What do you mean it's best practice? Well, I read about it on a blog. Okay. So it worked for one person. How do you know that works for you? Well, it looks good. CEO liked it. Speaker 1 (7:07) Yeah. Okay. Most of the decisions. Speaker 2 (7:09) I, I used a template. Speaker 2 (7:14) I picked the best. I'm not going to name brands because this is not the fault of any of the brands who do website templates. There's lots of really good ones out there. Some also not good ones. Speaker 1 (7:26) I Speaker 2 (7:26) looked at three templates. They said this was the best template for my landing page. This was the best looking homepage template. By the way, if you're Cyhull's Barbershop, that's probably fine. But if you're building an actual business where the website is the front door to your business, you may want to put a little bit more thought into it. Speaker 1 (7:50) Yeah. If you're an e-commerce brand, okay? If you're selling Speaker 2 (7:55) B2B SaaS and you're selling cybersecurity software and your average product probably sells for like $100,000. Every lead that comes... Every person that comes to your website and goes away, that's... really valuable you're millions of dollars you're losing Speaker 1 (8:13) so so Speaker 2 (8:15) what is ab testing is taking that same thing that we use in all of many other parts of our life and applying the taking websites from the pre-scientific gut instinct era to Speaker 1 (8:28) the Speaker 2 (8:28) scientific era there are that's a journey and for some companies it makes sense to do that for some companies it doesn't And it doesn't mean you can't make mistakes doing that. It's prone to mistakes. It's not perfect. But if your website is an important place where you do business in some way with your customers or potential customers, this is a much better, it's a really good way to make your website your number one salesperson. Speaker 1 (8:53) And we're talking about optimization here. So we're talking about you have your website, which is, let's call it your A. And then what you're doing is you're just testing little things here or there on a variable, on a B variable and sending traffic to both to seeing what converts. You don't have to limit it to small changes. Speaker 1 (9:21) You could make a big change. You can Speaker 2 (9:23) make Speaker 1 (9:23) big changes. Yeah, you can make big changes. If you're getting ready to Speaker 2 (9:27) run, have you run a marathon or a half marathon Speaker 1 (9:30) ever? I have ran a half marathon once and I promise I would never do it again. My body's not meant to run long distance. Speaker 2 (9:38) I'm in the same. My wife ran once and she did great. But just watching her train said, I have other things that I want to do. Okay. When you run. Speaker 1 (9:50) What did you do to get ready? I did the, there was like a protocol. This was a while back, but like I trained with a group and you do your runs and you, you know, you do your start with your two K's or you work your way up and then all the way up till you get to where you got to go. Yeah. So Speaker 2 (10:04) there's a, there's a training regimen. There's a training regimen Speaker 1 (10:07) for sure. There's Speaker 2 (10:08) probably Speaker 1 (10:08) a rest regimen. Speaker 2 (10:10) If you run 10 miles every day, you're not going to... That's actually worse. Long Speaker 1 (10:14) distance, short distance, speed. There was quite a bit of work. Yes. Speaker 2 (10:17) There may have been a dietary component. Okay. Now, depending on how serious you are, maybe you didn't need to make drastic changes, but probably some advice told you to avoid some foods. Probably go easy on the hamburgers and milkshakes and maybe go a little bit more on the healthy lean protein. Okay. Now... One way to get ready for a marathon is to only do one of those things. Speaker 2 (10:42) And then you could say, my time got better. And I know with certainty, it was because I wore the right kind of shoes. And then the next marathon, maybe you'll add compression socks. And the next one, you'll eat chicken five times a week instead of burgers. You could do that. And then every time you got this pretty high confidence level, by the end of the day, what gets you there a lot faster? You should combine those things. Eat lean protein, wear good socks, wear good shoes and train and get enough rest. Now, will you know that 10 % of your reduction in your race time came from the socks? I'm just making shit up. Another 5 % came from eating lean protein. Speaker 1 (11:26) No, you'll never know that. Speaker 2 (11:28) The same is true in A-B testing. And you can get away with it, I would say, particularly on websites. Speaker 2 (11:36) Particularly on a B2B website. First, I just want to be clear. The stakes are much lower. We're not injecting a cancer-treating drug into your body. Speaker 1 (11:45) Yes. Speaker 2 (11:45) The stakes are really high and you want to go one variable at a time there. Absolutely. We're talking about, can you get someone to fill in a lead generation form or click? I'm ready to start. I'm going to try that software for seven days first, see if it works. So the stakes are lower. The other thing is, in a... For our audience out there, I'm going to use some acronyms, B2B, D2C, just business to business. If I'm selling software to other corporations, that's business to business. If I'm selling shoes on the internet, direct to consumer. Direct Speaker 1 (12:18) to consumer, you Speaker 2 (12:19) have all this traffic. How many millions of people are maybe going to Nike.com to buy shoes? A lot. So you can quickly get a sample size, which means maybe you run it for a week and there's millions of people coming. You can very quickly detect there's a difference between the performance of two versions of pages. On the B2B side, how many people are shopping for a new cybersecurity software today? Some. In fact, there's quite a few, but not as many are shopping for shoes. So one way to get results quicker is to make big changes. You don't want to just do a bunch of random shit. But it is an acceptable trade-off to maybe you change the message on your homepage and the image that goes with it. And why? Because usually the messaging, what's called the copywriting,