Ready, Test, Go. // Episode 4

What Elevators Can Tell Us About CX


Listen to this episode on:

Apple Podcasts

About This Episode

Automation expert and author Adonis Celestine talks about putting the customer journey first, and craft high-quality products that directly address their wants and needs.

Special Guest

Adonis Celestine

Testing expert and advocate Adonis Celestine is the Director of Automation at Applause. He has written three books on digital quality, including Quality Engineering: The Missing Key to Digital CX. He also speaks at conferences all throughout the EU.


(This transcript has been edited for brevity.)

DAVID CARTY: Where do you like to travel on vacation? Do you prefer the hustle and bustle of the city? Maybe do a little bit of bar hopping or take in a musical. How about traveling to a big sporting event or a concert with thousands of people in attendance? Well, you won’t find Adonis Celestine there. No, no. He prefers a quieter, more secluded type of vacation.

ADONIS CELESTINE: Yeah, I love traveling, mostly in Europe. So I love islands, especially in places where there is low human impact, where I don’t see anybody. Yeah, Canary Islands is one of my favorite ones, especially because of the nature there.

The entire island is like a volcanic eruption. And when you are there, especially during off-season, if you’re there on a beach, you see, like — next to, you see a huge pile of volcanic mountains and get this surreal feeling of the power of nature. I love those places. I like to walk around it and spend some time over there.

CARTY: Of course, even if you’re seeking the solace of nature, it doesn’t count as a vacation if you didn’t take any photographs. Pics, or it didn’t happen.

CELESTINE: What it is if you don’t have those social media likes next to you? Life is boring. I love mountains as well. So just like climbing a few — not a big climber, I am, but small mountains where we do a small bike up and then have a small picnic when we come down.

So I do it with my family. So my wife also likes to walk. So we go around to the holidays.

I love Croatia as well. They have a lot of small islands, unexplored islands. And some of them are really remote, which you can get on with a boat, like, maybe once or twice a day when there’s a boat service from the main island to a smaller one. So I love it.

It’s an amazing place. I would recommend you to go there. I think Europe itself is quite beautiful.

CARTY: For Adonis, these vacations offer a critical time to unplug, reduce the noise, and center himself, not to mention it’s also a great way to avoid any emergency work calls.

CELESTINE: I work a lot during my day, work on several stuff, including projects, company, my personal stuff. I love to speak. I like to write. I spend a lot of time on that. I also play hard. So on these vacations, I like to shut out myself, be away from the world. So these kind of vacations help me to kind of hit a reset button. But I can — after the holidays, I go back to work with a fresh mind with some fresh inspiration.

CARTY: This is the Ready, Test, Go. podcast brought to you by Applause. I’m David Carty.

Today, we’re speaking with seclusion seeker and automation expert Adonis Celestine. As the director of automation at Applause, nothing satisfies Adonis more than a happy customer.

As an advocate for test automation, he strives to make software testing easy, yet innovative for brands all around the world. Adonis has written three books on digital quality, including his most recent, which he published in September, called Quality Engineering: The Missing Key to Digital CX. He also speaks at conferences throughout the EU, but right now, he’s speaking with me.

All right, Adonis. Let’s start off with the broad strokes question. How do you define digital quality, especially as it relates to the customer experience?

CELESTINE: Yeah for me, from my perspective, quality is really hard to define. It depends on who you ask. Like, if you ask my kid, who likes to play Roblox, they know — they would never care about quality. It’s like, it’s in a game. Like, they’re climbing stairs. And after like 10, 20 steps, there are no steps anymore, and they don’t care. It’s just a feeling, the experience they get. That is what matters to them.

Same in the business side as well. It depends on who you ask. You ask a business guy what is quality, this is all the requirements. They write it up.

And if you ask a quality engineer what quality is, it’s all about test cases that he has written down. But in the end, quality is a perspective. It cannot be properly defined. And that perspective can change based on the experience of the user. And that’s why I see quality and digital experience are something very similar to each other.

CARTY: Now, Adonis, the title of this episode is “What Elevators Can Tell Us About CX.” So, Adonis Celestine, what can elevators tell us about CX? I understand there’s an example that points back to the Industrial Revolution.

CELESTINE: Yeah, that’s indeed, because elevators have and quality and experience of something in common. During the Industrial Revolution, when — after the Industrial Revolution, when the passenger lifts were introduced, people were really afraid to take the lift. They were feeling claustrophobic, and they were complaining that the lifts were really slow. And they are even better if they walk up the stairs, which is much faster.

So they did several things, like bringing up the speed of the elevators, making it really fast and usable. But still, people didn’t like it. And then a smart engineer came up with an idea.

So what he did, instead of approaching it technically, he put on some mirrors on the elevators. So the mirrors not only gave them a feeling of there is more room to avoid claustrophobia, but it also kept them occupied. So people were using it for all kinds of stuff, like adjusting their hair, putting on makeup or lipstick. So they did not notice the passage of time, which gave them a feeling, OK, it is quite better.

You can also see the same in your digital applications. You see this progress bar when you are opening a web app, or you see all the spinners. All these are like putting these mirrors on your elevators — to make the customer feel that your application is faster and more efficient. So that’s why for a digital experience, it’s not always about technical stuff. It’s about managing this kind of stuff, which brings up — elevates the experience of the user which is very important.

CARTY: Now, if any elevator engineers happen to stumble upon this episode, might I suggest a minibar or an espresso machine? That would be my next innovation. Just a suggestion.


CARTY: That one’s for free. You can take that one.

So I want to get back to something you said before, Adonis. So, you were talking about business requirements, user requirements. These typically define the test cases, right? So, how far do these go toward actually addressing user needs and concerns?

CELESTINE: Yeah, it depends on project to project. I have seen projects where you have these user requirements or so-called business requirements which have been worked on for 12 months in an Agile way. We develop something, and then we go to the customer just to find out that there is actually no market for that product itself. So that’s why it’s very important to have some design thinking mindset where we first agree on an idea, make small prototypes, take it to the market, validate with the right customers and users, and then build on top of it. So it’s an iterative method with valid user feedback. So that is where I see, like, a gap in the current development methodology where the user requirements are somebody’s thought, or someone higher up in the organization thought this is what the customer wants, which can be far from reality in some cases.

CARTY: Right, so it sounds like there are definitely some user-centric gaps there that really aren’t covered by these user requirements and test cases. Can we drill down on that a little bit more? What other kinds of gaps do you see in this kind of situation?

CELESTINE: I can give you an example of a ketchup bottle, for example, from Heinz. So you have this upright bottle which is like a very nice product. And you also have this ketchup bottle with the down-facing lid, which is more popular. If you look at the market analysis from Heinz, the bottles with opening at the bottom sold much more than the ketchup bottle with the lid on the top. So you see two distinctive designs here. One was more product-centric, which was based on requirements. User requirements, based on functionality — tested, probably, on the test cases derived out of it. And the other one is more like what the actual user wants.

While the ketchup is — Heinz ketchup is a great product, if you have struggled to get it out of it, especially when it is cold and when it is, like, only the few ketchup left in the bottle, that is where the user-centric approach comes into picture. So I see a lot of gaps, And companies which have managed to identify this and make revenue or make an iterative development in their products have been really successful

CARTY: Yeah, that’s a great example. So you ultimately propose a shift from a product-centric approach to a user-centric approach to development. Now, this all makes sense, but it’s probably no small effort. So, what sorts of challenges can businesses expect to encounter as they make this kind of shift?

CELESTINE: You’re right about it. It’s more easily said than done. Of course, the first thing to find out is, like, what does your customer actually need or want? That’s a $1 billion question which companies are trying to solve. But it is also not always what the customer wants. As Henry Ford said it, if you ask my customer back in those days what they want, they would have probably replied like, ‘OK, I need a faster horse.’ They don’t see the need for a car there. So it is not always about what the customer wants, but it is about balancing what you think. There is a market for a product and what the actual customer wants. So balancing this and getting this outside-in perspective into your design process, into your development process, is the key.

CARTY: So there’s a happy medium there, right? I mean, this is also very different from the days of old. I mean, you can go and make that monetary investment to gather that input from your customers, right? So is it a matter of some of these companies just not spending that money the right way or not valuing the customer’s opinions enough? What’s your perspective on that?

CELESTINE: Yeah, it’s like — it’s a process. So we have to try it out first. It’s not always easy. So there is always some kind of resistance to change.

So my perspective is, like, just try it out. Invite your customers probably for your product backlog meeting or when you have developed a nice prototype. Just show it to them to see, like, how do they think about it, get that feedback into your process, and then iteratively develop it. I think that’s a good way to approach — to bring in this user perspective into your process.

CARTY: And if you still have some opposition when you’re trying to make this shift, there will always be naysayers or people that don’t want to change their ways of working or thinking about a product. How do you proactively deal with that opposition? Because it’s going to pop up with any kind of change that you make, right?

CELESTINE: Yeah, of course, all this process that I talk about requires a real mindset change itself. And that is not easy to happen, easy to make it within an organization. So it’s all about giving them a taste.

For example, when I was working for a bank, we invited a few of our customers to our office. We just put them in a room and asked them to use our application with some minimum instructions. And like, we were, like, waiting out — it’s like an interrogation room.

So we could watch what they are doing. They cannot see what we are up to. So with the way that they were navigating the application as developers and testers in an Agile team, that gave us a lot of insights. That gave us a lot of feeling about, ah, that’s a different perspective. I never thought about it. I never thought about a few things even my own application, which I have been building for the last five years or so. So, those insights, if they can get a taste of it, probably, like, that resistance from people who are willing who are not willing to change will go down a bit because they have seen it in action. They have seen it working. They get a perspective out of it.

CARTY: Right, and to help emphasize the point, from your perspective, what are some brands that succeed with an aggressively user-centric approach? Anybody come to mind?

CELESTINE: I think I have a lot of practice. So if you look at the modern tech companies, they are all about digital experience. Let’s take the example of Netflix, Uber, Airbnb, All these companies have some kind of like X factor which makes them fly, and that X factor is the CX factor, in my customer experience, factor. So if you look at Uber, Uber is nothing about — Uber don’t own any cars by themselves. They are nothing about travel industry itself. It’s a tech company which revolutionized how travel needs to happen. I can give an example.

As I said in the beginning, I travel a lot. So before the Uber days, whenever I go book a taxi to the airport, the tester in me kicks in. So I used to doubt everything. Did I book it correctly? Did they register my name properly? What if the driver’s car breaks down? All those things creates anxiety. For me, I don’t care if Uber is expensive or not. That anxiety has been taken away with the use of Uber. For me, that’s a good experience of a service. So all the tech companies that we see today are focusing on customer experience, and that has been their key success.

CARTY: But I bet the tester in you still checks the app five times to ensure that you put the right address in there, right?

CELESTINE: Yeah, of course.

CARTY: I can relate to that 100%. So I’m with you.

CELESTINE: Yeah, that’s why Uber has an app which gives you some real time insights. They know exactly — they acknowledge that they have received your request. They tell you exactly where the driver is. All these things reduces those kinds of anxiety.

CARTY: Absolutely. Now, I’d imagine narrowing down the customer journey isn’t easy for everybody, right? So what are some ways to verify and refine how you’re developing for and testing against a customer journey?

CELESTINE: Yeah, there are three different ways. The most common one is, like, to first build out some kind of personas to identify who are your end customers, what kind of customers you have, segmented based on the age group or based on some kind of segmentation, and then work on those journeys, like how would those customers use my app, and build your application for those personas and also test your application for those personas. That will give you a good insight about what your customer is doing and also to ensure that that journey is seamlessly well.

And there is also the second way where it’s like a trial and error way, which we call the A/B testing in the quality world where we send out a feature, see if it is working— for example, in the US elections, back days when President Obama was there, they used this A/B testing to find out which of the website was attracting more hits from the common public. So they had a website with a nice US flag on it. They also had a website with a family picture of Obama on it. Of course, the one with the family picture attracted a lot of people. So that was a way for the people to find out, OK, this picture works quite well. You can do the same with products and features. We can build something, show it to a larger group of people to see which one works, and then go about that path.

And then there is also the third way, which is more a technical way. Use AI machine learning to also synthetically simulate how a user journey would look like. So for example, in flight situations, nobody can actually test an emergency situation. It’s all done through simulation. So those journeys and stuff could also be simulated and developed based on that and also tested for quality based on simulation.

CARTY: Yeah, whether you use AI or whether you use A/B testing, these customer journeys are dynamic, right? Everything from technical changes to geopolitical to economic factors, these can all change how a customer interacts with your brand and your product. So how do you provide a reliable, high quality experience if and when that journey evolves?

CELESTINE: That’s a really good question because a lot of people in this customer experience digital world think like, OK, I made a customer journey, and then I’m done with it. But your customer journey also should go through your change management process. It is not a good journey map unless you keep changing it frequently as and when your customer persona’s profile changes.

So, awareness like, OK, you have — this is not maybe the journey document. It’s the first step. And then also updating it based on the feedback — every time that you get a feedback from your customer, you also keep updating your customer journey map. So, over a period of time, it will evolve. And yeah, that will help us to sustain the changes that are happening in the customer profile itself.

CARTY: Now let’s go back to using AI. How can organizations use AI to help assist with predicting customer behavior? And are there some common mistakes that some organizations make there?

CELESTINE: Yeah, I think AI and machine learning, when used at the right level, can bring in a lot of good experience. For example, when you’re shopping at Amazon, it will show you some kind of related products. That helps me a lot because I can compare the prices, and I can also see what is there in the market in their product range or in that category which helps me to make that buying decision. But that is kind of a personalization. Next time when I come and search for a similar product, it also recommends me, OK, these are the products that you look before. Probably these are the ones you wanted.

For all these to work for the analytics, you need data. And that data, I know it’s collected from me. I’m OK with it until a level. But there are lots of companies where, OK, if I’m discussing with my girlfriend or with my wife about buying something, the next — and even without any Googling or without getting into any of the internet stuff, the next day morning, if I open a website and if I get recommendations to buy that product, that gets really creepy. So AI and machine learning are useful if you keep the personalization to a particular level in such a way that experience is still really sweet. Once you—

CARTY: Even that can change over time too, right? We might become more open and embracing of these kinds of predictive behaviors over time too as people, right?

CELESTINE: Yep. Yeah, I think, yeah, if you look at the market, like, hyper-personalization is, like, what everybody is looking at. Nike, everybody, every company, digital company, they want to give that tailored experience. But hyper-personalization beyond the point can also be quite creepy and piss off customers. A lot of people don’t want to give away their data.

CARTY: Yeah, there’s definitely a fine line to walk there, to be sure. Now, Adonis, moving forward the way users interact with digital products is going to evolve even more, whether that means the metaverse or just some other version of an integrated digital experience. The way consumers interact with their digital worlds will probably look very different in the near future. So how does this evolution dial up the table stakes for businesses, and what should they be strategizing for over time?

CELESTINE: I think change is inevitable in the digital world. Even very technical people who call themselves very tech savvy, they could also find themselves in 20 years, like, completely out of the digital world. It’s changing so fast, especially in the metaverse world. But I think if you look at the bigger companies or the digital companies, they are already started to think about it. If you look at Nike or if you look at Gucci, for example, all fashion brands, retailers, they already have a store in the metaverse that people can buy stuff. So businesses that adapt to these changes, they will survive.

And others, of course — the Kodak example often comes into discussions in these kinds of scenarios. You have to move digital. You have to move forward, and you have to adapt new technology as you go on.

CARTY: All right, final sprint here, Adonis. I have a few more questions for you before I let you go. In one sentence, what does digital quality mean to you?

CELESTINE: For me, digital quality is all about experience and perspective.

CARTY: Simple. I love it. What will digital experiences look like five years from now?

CELESTINE: That’s hard to predict. Probably we need an AI engine to do that. But it will move on from a nice to have position that we have now into a necessity for businesses to survive.

CARTY: What is your favorite app to use in your downtime?

CELESTINE: I have a lot of favorite apps. I play a lot of mobile games. I play “War and Order.” That probably takes a lot of my time every day. But still, it’s like a stress buster for me.

CARTY: Absolutely. What’s something that you are hopeful for?

CELESTINE: It’s about — in the digital world, it’s about inclusivity and accessibility. As I said, a lot of people would be left out, even the technical people, in the future. How do we bring all these people in the metaverse world, for example? That’s a key for the future.

CARTY: That was the Adonis Celestine. If you happen to bump into him on vacation, say hi to him for me. Oh, actually, you know what? Better yet, just leave him alone. Let him enjoy nature, OK? Thank you for listening. Shout out to our producers Joe Stella and Samsu Sallah and graphic designer Karley Searles. Feel free to reach out at, and we will catch you next time.