May 3, 2023

Guest Host Eric Melchor of the Innovators Can Laugh Postcast

Guest Host Eric Melchor of the Innovators Can Laugh Postcast
Spotify podcast player badge
Google Podcasts podcast player badge
Overcast podcast player badge
Castro podcast player badge
PocketCasts podcast player badge
RSS Feed podcast player badge
Apple Podcasts podcast player badge

If you are interested in conversion rate optimization, this is the episode for you. Khalid Saleh is the founder and CEO of CRO agency Invesp. He has worked with brands like eBay, 3M, and the Discovery Channel. In this conversation, he shares what is a reasonable budget for CRO, why he avoids looking at homepages when first evaluating a site, what you should focus on to get quick wins when it comes to increasing conversions, and more.

You will also discover:

  • a reasonable budget for CRO testing and what that includes
  • Why Google Universal Analytics is sunsetting
  • what you should focus on to get quick wins when it comes to CRO

Listen to the Innovators Can Laugh Podcast HERE --->


Dana Dowdell 00:05

Hey, there is your favorite podcast host, Dana Russ is your second favorite. We can accept that you so much for tuning into It's Just Business. Occasionally, Russ and I like to take a little bit of time off from podcasting. But we don't want to leave your podcast feed empty because we love you. So as we've done before, we are highlighting the podcast of one of our previous guests. So it will fill the spot of your normal Wednesday podcast drop. And we'll give you a little bit of insight into a different creator. And we hope you enjoy it. We'll catch you on the next episode of it's just business. 


Khalid Saleh 00:44

Sometimes companies come to us and they say, this is the page this is the moneymaker. This is what we need to focus on. And I always tell every company, I tell them, You're bringing a tank to a battle. Don't use this as an eye. Let us look at the whole side because you will be amazed at what we discover. 


Eric Melchor 01:00

That's Khalid Saleh, the founder and CEO of CRO agency Invesp. He has worked with brands like eBay, 3M, and the Discovery Channel. So, if you're interested in conversion rate optimization, this is the episode for you. Khalid shares what is a reasonable budget for conversion rate optimization, why he avoids looking at the homepage, when first evaluating a site, what you should focus on to get quick wins when it comes to increasing conversions and more. I'd love to hear your thoughts on this episode. As you listen, you can find me on LinkedIn at Eric Melchor. Tag me say hello, let me know that you're listening. And now let's talk with Khalid.


Eric Melchor 01:41

You created one of the first companies to focus on experimentation and conversion rate optimization in the US. How did they come about what led you into this business? 


Khalid Saleh 01:51

Oh man, I wish there was an elaborate plan to create the first conversion rate optimization agency. So in a previous life, I used to be a software architect for mostly on E commerce enterprise ecommerce projects. One of the last big projects I did in 2005 was for Motorola. And it was as a software architect, it was a dream job. Basically, they came in it was September, they said we have to launch the site by December. And they hired a large team about 120 developers, it was three software architects, I was one of them. And whatever we asked for they said, Yes, we want to do this technology want to do that? Yes, yes. As long as you can deliver by December, of course, we can deliver by December $50 million project of that the hardware was 15,000,030 $5 million for software. Absolutely incredible. And I had I had interest in marketing, but I never really delved deep into it. And I still remember thinking to myself, like, oh, man, those guys are spending so much money, what are they going to get the visitors to their site, that was the only concern that I had. I still remember talking to them about the SEO, I didn't know a whole lot of SEO about SEO, I was starting to ask some friends. And I'm like, Ah, they don't even understand SEO. Man, this is gonna be a disaster. Long story short, we launched we didn't launch in December, we launched in January. And they didn't have an SEO problem. I mean, the site's flooded to the point that's in the first day in the first six hours, the servers went down. And we had we had some beefy servers back then 16 Sun microsystem servers clustered and just could not handle the traffic. So traffic was not an issue. We bring back the system, or we're online, and I'm like, Oh, great. So I guess they figured out the marketing piece, no big deal. Well, by the end of January, and for that humongous investment that they've made, I think we've had about 10 orders? And am I be generous by saying 10 orders for that?


Eric Melchor 03:42

Where was this traffic coming from? Was there commercials or something? Got it. What was the awareness?


Khalid Saleh 03:48

It's Motorola, everybody knows Motorola, everybody comes. And if you think about back in 2006, Motorola was even a bigger name than what it is nowadays. They had at that point in time launched the Moto Q, which was the competitor of the palm pilots, there were no iPhones back then. So it was a really big deal. And everybody was expecting it. I mean, here's the funny part. Just to tell you back then, because we're talking about what's now 2016. By 1617 years ago, one of the big sources of revenue on the site was people buying ringtones via email is like key, my ringtone for your phone. That was kind of a big deal. And I actually that was a portion that I kept from myself to code because it's software architecture, you get to pick and choose. And people say, Oh, is that complicated? I'm like, well, there were so many systems that we have to integrate with to be able to display the ringtones and do the shopping and send the orders and do the fulfillment electronically. I mean, this all sounds silly. Nowadays, when you say it's a month later with 10 orders, guess what, many people lost their jobs, people basically, Motorola was not happy with the end result that we're not looking at. It's not on the consulting agency that they've hired, but they're like, What the heck, we've invested so much money and this was supposed to be kind of our launch into E commerce. It was a miserable failure, and I still remember going through the sides with my team trying to analyze what's happening, what's going on again, back then no real analytics. So it was just like us logging things into the server log and trying to see where people are stopping. At the end, we figured out the issue, it was all too late. And my wife at that point in time, was starting an agency and she really didn't know what she wants to focus on. And were like, hey, well, you know, maybe this is an interesting problem. I wonder how many companies actually deal with the same problem. And that's how invest was born. It's kind of small thing. And in all honesty, investment issue was supposed to be something on the side for my wife to just work on, get whatever clients that she can get. I ended up quitting my job and invest became one of the larger, more successful CRO agencies worldwide. 


Eric Melchor 05:50

All right, good story. Now, do you recall what the issues were with Motorola? Do you remember? 


Khalid Saleh 05:54

Oh, yes, I remember very well. It was a promo code. It was a promo code on the checkout. And we had it so prominently displayed, and at that point in time promo codes was were also very brand new. So again, you were dealing with problems back then that sounds really funny nowadays, but it's like, oh, you can actually get a promo code and instead of paying 199 for the Moto Q, you can get it maybe for like, you know, 129 we were pushing, we're telling people to leave the sites and you know, sometimes you can find the promo code sometimes you don't find the promo code. It was just poor usability and it's one of those funny things because ecommerce at that point, time had existed for a while. But almost every new e commerce website that came online had to reinvent the wheel. We didn't learn from each other correct is that oh, here are the mistakes that you need to avoid. No, no, no, we got to do it because we are going to do it so perfect. And we make the same mistakes again, and again, reinventing the wheel or just making the same mistakes, I guess two different sides of the same coin. 


Eric Melchor 06:51

Yeah, you know what else sounds funny when you mentioned ringtones? Because 2006 people were spending 10 $20 on ringtones and 2023. Everybody's got their phone on silent


Khalid Saleh 07:05

amazing like sometimes you think it's like people actually spend money on this on like a ton of money. It was kind of a made for like the E commerce operation Motorola that was like a big portion of the revenue at least it was anticipated to be to be Yeah, well, yeah. 


Eric Melchor 07:18

Okay. Now, what services does invest, provide for clients? 


Khalid Saleh 07:22

Couple things that we've added recently. So we're typically turn key, but we're finding out that there are more companies that have the developments and the design resources, but they need more of strategic CRO guidance, where we do the analysis, come up with the ideas and help their teams guide their teams to do the AB test implementation. So that's something new that we just started offering in 2023. Okay, great.


Eric Melchor 07:45

Now, what type of clients do you service? Is it predominantly b2b, or is it b2c as well? 


Khalid Saleh 07:50

Mostly, I would say 95% of the companies work with our b2c very focused on E commerce, but we are getting a ton more b2b. And b2b is always it's kind of an interesting from an agency perspective, as its b2b clients are an interesting animal, especially in the E commerce space. They don't necessarily have many conversions in order for us to run solid AB testing experimentation, but at the same time, the order value is 100 or 1000 times larger than the b2c clients. So it's kind of an interesting dilemma. How do you run CRO for b2b companies 


Eric Melchor 08:23

now for a b2c company? What would you say for a brand big brand that's maybe doing? I don't know, at least 50,000 visitors a month? What would be a reasonable budget for testing and optimization? 


Khalid Saleh 08:35

That's a good question. So I always think of the budget as two components. There's the software side, and then there's the human resource side. So if you think about the software side, well, I need to do some sort of, I need to have some AV testing software, just at a bare minimum, correct. But I need some sort of user behavioral analytics, software, heatmaps session recording on sites, polling, exit, intense pop ups, and things of that nature. So you need to budget for that. And in all honesty, you can spend anywhere, let's say you have 50,000 visitors, you can spend anywhere from almost nothing, because you're correct. And you can be spending up to 10 $15,000 a month, are you getting a lot more for the 10 or 15,000, you do get a lot more advanced features. 80% of the time, and people don't like it, when I say this 80% of the time, you might not use those features. They are good. But I always tell people on the car, how much how often do you use them? And the best example that I can give is Google Analytics universal versus GA four. That's the big talk in the community nowadays. And people say, Well, you know, Google took away all these features from Universal that we used to have. And we're used to having Google universal now we don't have them in GA four. Well, there's a reason for that. Because when Google looked at the features that people use them, there was a ton of features in Google Analytics universal, they found out that 80% of the people did not use those features. So they said, Oh, hold on. So we're collecting all this data. It's costing us so much money and this is free. Correct. Most people are not paying for Google Analytics less remove them. So that's one piece, which is the software piece now on the human resource. So in order for you to run any type of testing, you will need somebody who will basically do the analysis Correct. Figure out oh, what needs to be fixed on the website? Where are the revenue opportunities? Where do we have leaks in the funnel? So that's one, that person also after the discover the problem, they need to come up with a design on how to fix that problem. So you might get lucky. And it's rare correct to find those unicorns somebody who's able to figure out the problem and think, come up with a design. And then we are a CRO company. But I've seen this way too many times, we always AB test anything that we've rolled out to our website, I cannot tell you how many times I look at something and I'm like, Oh, yes, there is a problem here. Oh, yes, this solution, this new design is gonna fix it. 100% and we roll it out, and we tested it, it's like, oh, 50,000 visitors 25,000 people are gonna see the old design, the broken design, when he thought 25,000 people are gonna see the new design, and we're gonna measure the impact. And I've been humbled way too many times where people like the broken design, I'm like, really people? How about this is so much better. So in order for you to run any type of AB testing, you need a front end developer, if you're doing kind of so that they'll be able to code that test. So you're not just taking the fix and just deploying it on your website. So at a minimum, you might need two to three resources correct to run that. Now. If you're running those as full time, people, full time staff in your company, they might be costing you anywhere now in the US probably 30 to $40,000 a month, you can hire agencies to do that work. And agencies are oh gosh, like, you know, they're all over the place, you can pay probably as little as $5,000. And you can pay as much as $30,000 $35,000. So it's all over the place. 


Eric Melchor 11:40

Now, if you're interviewing an agency, what are some key questions or maybe just an expert CRO expert? What are some key questions that you should ask to make sure that this person is knowledgeable and probably the right person for the role? 


Khalid Saleh 11:51

So one of the questions that I always like to ask even when I hire CROs for my company, I'll ask them about the different projects that they worked on. And one of the very basic questions that I asked them like, well, so I have a site with 50,000 visitors how and I want to run AB experiments on this website, because I found that there's a problem on the homepage, and I want to run AB experiments. How many variations do can I run in this AB experiment that I have? And buy a variation? So we have the original site design? This is the new home? The original homepage? Correct? This is the broken homepage, and I'm going to come up with multiple designs to fix that broken design. You can come up with one design two designs, five designs, 100 designs correct. There's no limits, really, because I always say you bring me two designers, you lock them up in a room, you tell them here's a problem, they'll come up with so many different designs to fix the problem. So how many designs do I run against that original design? And I'm very careful to listen to what they say if they asked me well, so how many visitors you have? And I told him 50,000 visitors, I know that I'm talking to somebody who does not understand statistics and statistics is very important in AB experimentation. The right question. And this is the question that I'm looking for from them is the question, how many conversions do you have? Because really, whenever you run experimentation is based on the number of conversions not based on the number of visitors. So that's important. So if the question they asked me, correct that CRO is asking me or that CRO agency is asking me tells me how much they know about CRO if they're asking about visitors, I'm like, oh, man, you're just started and conversion rate optimization and experimentation. If they ask, well, how many conversions you have? I'm like, okay, now I have somebody who I can have a conversation with other questions. I always ask if somebody has been doing conversion optimization, if somebody has been doing digital marketing, correct. And everybody has many good successful stories that they can share. I'm not interested in successful stories. What I ask is like, tell me about some of the failures you've had. If you've done this long enough, you probably will have the scars to talk about them and talk to me what happened in those because I am interested in understanding, okay, what are the reasons that some of these CRO projects that you've ran have actually failed? So that's always an important thing for me to understand. I focus a lot on the analysis that the person does before the test. How do you come up with dissing ideas, I interview people where we're actually we just hired two new senior CROs. So I always ask that question again. How do you come up with testing ideas? And I'm careful also to understand or they just come in with if somebody tells me Well, yeah, is this so obvious? There's so many so many things broken on the website is this I'm looking at? Nothing is so obvious, otherwise, most. I mean, ecommerce companies are run by intelligent people. And nowadays in 2023, everybody has the experience. So if you're just saying, Well, yeah, we should test this. We should test that on the cloud. Stop, stop. You need to do the homework. The testing piece is actually the easy piece correct? Like you know, figuring out how you code that you can hire a developer and designer come up with ways to fix the problems. Identifying the problems is very critical, even harder. And this is the piece that most people tend to ignore. What type of analysis do you do after an experiment finishes? Because think about this every AB experiment. That you run has hypothesis behind it, I found something broken. And I think if I make XYZ change, I'm going to fix that problem then on the website. Great, I love that. But typically, in order for you to fix this problem, you come up with three or four different designs. Correct is like a design one design to design three, they're all attempting to fix the same problem. Great. Once you find a most experiments that one design actually fixes the problem create our hypothesis was correct. Here's the tough question that most people try and skip on. Why didn't design two and three, fix the problem? What was wrong in those designs that actually made them lose? And a B experiment? That's a tough one? Correct? Because now you need to actually stop and think and say, Hmm, okay, well, we know the hypothesis was correct in one design, but in the other designs that where it failed, I wonder why this why this happened. I'll give you an example, which until now, I'm looking at trying to do the analysis on the team shipping date. Nowadays, everybody's struggling with shipping dates correct is like do we show that, hey, this is one of the product ships, we're working with one company, large ecommerce company. And we said, you know what, we maybe need to add clarity to visitors to tell them when the product is going to ship out. So we're going to show the shipping dates on the product pages, we're going to show the shipping page and shipping dates on the in the cart page. And then we're going to show the shipping dates in the checkout three different places. So this is a rather complex experiment, because it's running on multiple pages, this experiment. And because they have the transcript through multiple iterations, at some point had 27 different variations, because you can show it's different places different pages. At the end, the winner, the winner design was showing the shipping dates on the product page, not showing it at all on the cart page and showing it in the checkout less prominently than when we showed it in the product page. It's just fascinating to me. And now you're sitting is like okay, so I have a winner. This is great. So what's the theory now? So we actually need to be very clear with people on the product pages. We don't need to overwhelm them in the cart page, because yeah, too much information, too much information. And in the checkout, we also need to be clear, but not to the point where we scare them off that analysis, really looking at the data and trying to understand the story that the data tells you. It's just fascinating to me. 


Eric Melchor 17:08

Yeah, no, that's pretty important. I can't remember the exact stat but I remember reading somewhere that if the shipping information is not clear, like when the expected delivery date is going to come in a large percentage of people will drop off at that point. After a quick break Khalid and I discuss what they look at when evaluating a website in the key areas that have the highest impact on conversions. E commerce and DTC brands generally have a lot to deal with, especially now that customer acquisition costs are rising optimonk can help grow your email list like never before with conversational pop ups personalize your returning visitors experience with dynamic product recommendations guide your visitors to the right products based on their stage of awareness and create personalized experiences for each individual customer offer one website personalization platform and optimize grows with your business whether you are a solopreneur or doing $20 million in annual revenue optimizer is built for scale add a variety of integrations, including Shopify in Klaviyo. It's why more than 40,000 e commerce brands use optimonk to treat visitors like people, not traffic Plus, our personalization workshops set you up for success on day one. So you can identify the biggest website personalization opportunities that will give you the best ROI. There's no better time to start implementing website personalization. Get started for today, that's Welcome back to my conversation with Khalid. I wanted to dive deeper into how can lead and his team evaluate a site. So I asked him how his team approaches optimization. And what are some quick wins when it comes to CRO?


Eric Melchor 18:46

So you were talking about, you know, experiments on the product page in other pages. So when a client approaches you, and you're designing these experiments, I think when most people think about conversion rate optimization, they're thinking of landing pages and home pages. Do you feel like wait a minute, there's other pages, there's other key pages and other parts of the website that actually are just as important or maybe even more important? 


Khalid Saleh 19:10

Yeah, so one of the things that we've learned over the years, sometimes companies come to us and they say, this is the page. This is the moneymaker. This is what we need to focus on. And I always tell every company, I tell them, You're bringing a tank to a battle. Don't use this as a knife. Let us look at the whole side because you will be amazed at what we discover. Usually when we're evaluating we will look at the wholesale sites and come up with issues that we find on the website and then we prioritize them. So that includes the landing pages the homepage, the collection page for an e commerce website, the product page, cart and checkout. One of the key areas that we evaluate is how close a particular page is to the funnel to kind of like the conversion points. The closer you are to the conversion point, the higher the impact this page will have on your bottom line. Most companies when they run experiments should they think, hey, I ran an AB test on my homepage, it shows that this new design has a 10% lift and conversions. This is great. That means I'm gonna see a 10% increase in my revenue. And I tell them no, you'll be disappointed. They're like, what's why it's an A B experiments is 10%. I tell them, when you only optimized for the visitors who go that through that homepage, and in VISTA who didn't go through the homepage, there is no impact on them. That's the reason the closer a page is to the end conversion point, the end points, the more impact it has on your bottom at 10%. On your product pages, for example, for an e commerce website, or 10%, lift in your cart pages at 10%. In the site's revenue, for the most part, a 10% on the homepage might be half a percent and the site revenue. That's the reason one of the goals. One of the things I tell my team, I tell them I avoid home pages when we first start experimentations tell me why tell them because their political battles, everybody looks at the homepage, all verify our logic I'm so excited might be might be only like a one or 2% of the traffic that goes through town like it's so like, even if you have a 10% lift there, the final impact on the site revenue is so minimal, that it doesn't really matter. Don't fight these battles, establish your track record, establish how good of a process you have. And then you can get to that homepage. So the closer a page to the conversion, the more important it is the more visitors percentage of visitors that go through the page, the more important it is. And this is one of 18 different metrics that we evaluate before we say we're going to run a test on a page, we're


Eric Melchor 21:27

nowadays with visitors coming from maybe affiliate sites, maybe you're working with influencers and maybe those influencers have different landing pages. So they can be coming from everywhere, maybe directly to the product page, in some instances. So true. Now I'm curious Khalid, because you have so much experience when it comes to optimization and looking at different brands, when you run experiments, obviously, you're testing different designs, maybe one design is about the copy, maybe another design is the actual CTA button, maybe another design is an image on the page or something like that. And your experience is it pretty split between those three things in regards to hey, it's not always the copy, or it's not always the CTA button. But maybe it's like 33% for each of those three things or my way off. 


Khalid Saleh 22:11

So it's interesting, I think of them a little bit differently. I was say when you look at a site or a particular page, there's what I call the quick wins, something's so broken on this page, I'm like, okay, we can fix this, the client is going to be happy, everybody's going to be happy. And they're going to think they've made the best decision hiring us. And those could be anything in all honesty, it could be that they're sending traffic from Google ads, or Facebook ads, and they're making a promise in the ad and then somebody lands, hey, it gets 20% off and somebody lands on the page. And you don't mention that at all, you need to really mention that because you made a promise in your ad. And we need to tell people that you are in the in the right in the right place. So that that could be something as that sometimes the images don't communicate the value of the brand, sometimes the copy is so dull and boring. And I would say when it comes to quick wins, it's mostly around two things is mostly around usability and design. Those are the quick wins. And usually when you start a project, the first I would say four to six months, four months, five months, six months is around the quick wins, just hit them figure out what it is. And that's probably about I would say when you start a project 70% of the work that you're doing is around those quick wins. So again, usability issues and design issues. 30% is more strategic, and by strategic. Now you look at copy, and then you look at round positioning in the mind of visitors. Strategic testing is a lot harder, correct. And if you don't get it right away, compared to the quick wins, correct quick wins, like okay, here, we fix this, we fix that the quick wins, though the design and usability don't give you huge lifts, they might impact your site conversion rate by 20 25% 30%. After six months, if you're looking for something bigger, it's around brand, run, copy, those are the first six months now think about the next six months, we've already had all the quick wins, companies stay with us now on average, about 39 months, we've already hit all the quick wins, we fixed everything. That's when you move into something a lot more strategic. Again, now you're looking at the whole user experience around the website and you're looking at messaging that becomes about 80% of the work that you are doing work. So we finished the first 12 months, you know, doing strategic and branding, what do you do after that, then we look at what we call best templatized testing. So if you think about a cart page, or a product page, or even a homepage, at this point, there are specific design patterns for each of those pages. For a cart page, there's probably seven or eight different design patterns. So you know, we say you know, let's test some of those design patterns against the winners that we that we have. And it's an iterative cycle. Now one of the funny things that I mentioned to one of our team members was very frustrated. He's like a client spends like, every time their dev team touches the website, they break something and I told him, I mean, every time the dev team touches the website you have quick wins yeah, I guess they're keeping you. You know, is this as good as you're keeping your job this way? Why are you frustrated every time they release something, look at it and come up with something to test sometimes, by the way, and this perhaps is the one exception that we have. Sometimes we look at something that is so broke that we're like, we don't need an AB test. Here's a quick fix. Let's just go ahead and deploy this. We're working with a company SAS company, actually, and the CEO calls me he's like, I don't know what happened. your conversion rate dropped, like in a white almost at 85%. Absolutely horrible. Can you guys come in, and he was just really panicking. And I'm like, Guys, what happened? There like nothing really looking out. But yeah, we see the drop in conversion rates. First thing you do when you see a huge drop in conversion rates, forget about the analysis, forget about, like, all the work that you need to do is very simple. You actually go through the website, you know, and you actually subscribe just to see what's going on. Because 80% is very unusual. 


Eric Melchor 25:49

This is like the Motorola example, right? You learn from your experience, right? Let's just go through and see what happens.


Khalid Saleh 25:55

I select the right click no subscription level, create an account, okay, I'll create an account I put in my email. And I put in I have a standard password that I use quick password is just password does not rights. Okay, well, I use LastPass. It has like, you know, it can generate passwords, and I have a setup 12 characters fill password is not strong characters, and it failed. And I'm like, Okay, now I need to read. They're like no password policy. And they want this and I have to change it from LastPass. It's went through so I call this the I'm like, what happened? Why the strict? I mean, my first two attempts through LastPass without thinking did not meet your password policy. Like, wow, we had to hack like, you know, a couple weeks ago and our IT team, like your dev team deployed this, like, okay, yeah, they deploy something so secure, that I could not even go through and your conversion rates dropped. We don't need to be tested. I appreciate security. I used to be a software architect, but okay, you've like you know, you've killed you know, you've thrown the baby with the with the water. 


Eric Melchor 26:55

Yeah, yeah, problem solved right there. Now, for the audience I want I want to go a little bit deeper around the usability and design aspect. Because what I think that you could be referring to, I may be completely wrong, because when you say usability and design, I'm thinking where the call to action button is located? How easy is it for people to know what the next action item you want them to take? Is? Is that what you mean? 


Khalid Saleh 27:22

Golly, sure. So when it comes to usability, there's the 10, what we call the MMG principles. So for example, system status, I'll give a simple example. You submit, you're expecting the system, the software, the site to respond to tell you, you know, we're checking this, we're doing that if it's going to take too long. So you see those, like, you know, like things that come up on the screen sometimes like, you know, circling that's is a system status, correct? Telling people that, hey, we're actually doing something as opposed to I'm not sure what's happening over here. Where is my browser status? When it comes to usability? For example, as I'm inputting, for example, and you see this quite a bit on airline sites, you're filling out some information and they're doing validation right away? Well, I see, you're telling me my email is invalid. Although I'm halfway through writing my email, give it give me a chance. Let me fill that out. Things such as you know, I'm interacting with the buttons and how I select buttons on the on the screen. Correct. So can I click this? So I'm selecting the size and I'm selecting the color and the arrangements of those. So that's, to me, that, to me is usability. And again, like, and if anybody, Google's the 10 an NG principles, they're very simple principles. And it's amazing to me how in 2023, we still tell people in the cube broken a simple usability issue. When it comes to design, there's two aspects to design. There's how we, as humans view design, correct, because design actually communicates a message to us, you can tell somebody that I have the best, most durable truck beds, for example, but maybe you're selling crock beds, but at the same time, the image corrects of the site has a car, it's not even talking about the truck. And I'm like, really, you're selling talk about it's very simple, okay, you can have an image of a truck bed, but it doesn't get the design of that image and how it's, it shows it can really communicate that wow, this looks like a really highly durable truck bed or skin say, Hmm, really, it looks very plasticky. And I'm not sure if I'm gonna buy it. So that's something very simple, but at the same time when it comes to design, correct, so design communicates a message and need to think about that message in addition to copy communicate message, but now we're just focus on design. But also when it comes to design, humans view does a page in a particular way. So it's either what we call a Z pattern, correct. We start from the upper left, we go to the bottom left, and then we go to the to the right, so that's one pattern or an E pattern, making sure that your text is flowing and your design is flowing correctly, where it meets the eye. That's part of design the location of a CTA, right. Where do you place that CTA I will say when it comes to CTA, I have what I call I don't speak Russian. So I have the Russian language test, but you can replace that with any language that you do not speak. Oh, Sophy on the classic, your webpage as it's designed right now, I'm going to replace orthotics and it's in a foreign language that you don't speak. Can I tell through pure design the primary city on the page? Or am I going to be confused? Because in a language that I do not understand, people always say like, how do you come up with this? Tell him well, somehow I don't speak Turkish, somehow, we end up opening an office in Turkey, and I would go visit Turkey and I had to open a business account, which meant I had to deal with Turkish banks. And those Turkish banks don't really appreciate the fact that some of their customers don't speak Turkish, that's meant I login to the Turkish bank, and I need to figure out how to actually log in and oh, gosh, the I don't know if there were AB testing, or they would just keep on changing the design just to challenge me every couple of weeks, because they will change things. And the buttons all look the same. And I'm like, oh, gosh, you know, and then you try and do Google Translate. And Google gets confused with their designs. And I always said, I'm like, you know, it needs to be very clear where to where to click on yesterday, my friend Dharma Shah, who runs inbound, and I've known the guys that hotspot from it was from CES, it was his first few months, I was seeing an ad for inbound. They're putting it everywhere on Facebook, it's coming up end of September. And it's funny because it said, I think inbound, I forgot 21 or something. And then it had the dates. So it had the title, and it has September, but then the days of the actual events were on the extreme other end. So it was not connected. Correct. So I was reading it. And it's funny the way I was reading, I'm like impound, and I'm September. I'm like, what is the date? And the numbers were so far? And this is this is a big company, correct? They're over multi billion dollar company. And I'm like, What are where are the dates? Which inbound is this? And it took me actually, five seconds. I'm like, Oh, here's the date. I don't know why they put the instead of saying September 22 to the 28th next to each other. They put September in one area and the 22nd to the 28th. And a completely different. So I actually took a screenshot and I message them on LinkedIn. I'm like, I'm not sure. Maybe it was intentional, by the way because it's really stopped me. Yeah. But maybe it stopped me because I care about HubSpot. And I use HubSpot. And I've known the guys there. It just stopped me completely. And I'm like, I don't know if the designer did this. Initially, I doubt that they did intentionally to stop people. I'm like, this is just poorly designed banner. And if you're spending money on Facebook, to drive people into your landing page, it needs to be very, very clear. And it needs to just think about the user I flow. I'll finish with other example. Some designers love to do two, three column correct on a page. So they go one column where the information is there. And they'll put the information, column one, column two, column three, and I tell them, Listen, I'm going down, I'm reading this is fine. Now you expect me to go to the left and then go to the middle and then go to the people don't like to do that. And when they when they see a design like that, they just ignore the section and they move on to the next section.


Eric Melchor 32:36

Khalid, thank you so much for coming on to the art of personalization. Where can people learn more about you and endless 


Khalid Saleh 32:43

so I hang out in the land of LinkedIn, some always on LinkedIn, I used to post a lot more regularly. So you can just search for Khalid Saleh on LinkedIn. Or you can check out our websites invest i n v SP is a p as a pilot people, lots of thanks, confused investment All right. Thank you so much, Khalid. Thank you. Thank you, Eric for having me. 


Eric Melchor 33:05

I had a great time chatting with Khalid. He has a very mythological approach when it comes to assessing a website. And because of his experience, he has a good understanding on where to start, what to focus on in the importance of post analysis when an experiment has been completed. If you want to learn more about complete, go to Links to all of this are in the show notes. Thank you colleagues for being on the show. If you liked this episode, send me a note on LinkedIn or leave us a review on Apple or Spotify. Thanks