Alon Peleg - The ROI of artificial intelligence

And in today's discussion with Alon, We'll talk a little bit more practically, a little bit more hands on about how you can drive real value and get real returns from your AI initiatives and AI implementations. Alon, welcome to the show. Very happy to have you here and to have this conversation with you today.

Would you like to add anything to the intro before we dive into the meat of it?

Alon Peleg: You did it really well. Thank you very much. I came to Iola six months ago. Before that, I was a general manager both in Cisco, Intel, and Wix, and I believe that now it's like they started the amazing journey here in Iola.

Great team, great product and really excited to speak with you about AI today.

Tim Butara: Awesome. I'm sure that you'll share some success stories and some personal cool experiences, but we'll leave that for the end. I want to first get into, you know, the, maybe not success stories. So practical, not so hands on, not so personal stuff.

And right off the bat, my first question for you today is how can companies maximize the potential of their AI initiatives, AI usages while minimizing, you know, you know, down downsides, such as time wasting, for example.

Alon Peleg: Yeah, I think I have the perspective on how to do it right. At least I will try to share it today.

First of all. It's clear that you need to set up a business KPI, which are important and aligned to the company goals. That's first. So you need to be very specific what you want to achieve. Second, it's the team that will be involved in the process. Many times you can see that it's it start in only in the innovation team of the company.

They will do amazing job. They will explore the, the product, the technology, the outcome. But then when they meet the, the team that will need to actually use the solution across the company, then they found blockers that they didn't think, think about in advance. You know, and then it's a big waste and frustration instead of that.

What we see best is to create some kind of a steering team that includes innovation, people, people from the production, people from the operation, IT, if relevant, engineering, if relevant, people that can take decisions that they know that there is budget assigned to that So it's another blocker, right?

You, you started a POC, you started to run and then, okay, we want to scale, but okay, no one has the funding for, for that. So, so the best will be, okay, you form the team, you know, the KPI, you know what the budget is, and then you will also start with a pilot. With a clear KPI, involve all the right people, get the buy in on the results, and then scale.

Don't try to swallow more than you can, because it's a technology change. It's a mindset change. We know the technology is working. It's more around how to make it practical, and the few tips I gave you right now, that's Well, my experience key success factors

for, for such a journey.

Tim Butara: And yeah, you mentioned that budget can be one of the, of the main blockers, right?

What are some other important blockers that, that teams doing working on these initiatives should take into account here and should, should kind of plan for and tackle?

Alon Peleg: Yeah, very good question. First of all, you are not coming to work in a lab, right? When you start to work, there is an ecosystem and the ecosystem consists from legacy systems.

That are already deployed existing procedures that people are working like that for years and you don't want to disturb them You want to show that with ai you can once you can do those processes faster Second you can capture Data that was not captured accurately before or unstructured data that you can capture now.

You want to The to make sure that the employees that will use this technology will actually enjoy Because, hey, some of them think that you are coming to replace them. No, we are not coming to replace you. We are coming to enhance the way you work. And come with outcomes faster. So maybe Maybe we can now improve the process with you.

So you, like, we need to make sure that the group that will be using the technology will enjoy and see the advantages quickly. They should be part of the process, so they will be the supporters of the process when you scale.

Tim Butara: Mhm.

Alon Peleg: So

Tim Butara: this impact on employees is definitely one of the, one of the big things that you need to account for.

Right. So are there any other key considerations here or like key strategies for balancing, you know, maybe that these negative impacts, you know, with the advantages that it can bring, like besides, besides what we already covered.

Alon Peleg: It's mostly value. For people who wants to be impact to, to be part of creating something bigger.

Okay, even the simplest employee in the grocery store, right? If you can show him that There is a way to enrich the way he's doing things to create bigger impact on the company with technology And to show him that hey, you are doing something that others are not doing you're actually Part of a revolution, which it's only the beginning.

People will feel more impactful, more important, more valuable, and they will join the, join the process. And we see in, and I will speak about it later. We see that the ROI in companies that are doing it right, involving the employees, following the pilot to scale, choosing the right stakeholder, thinking about how they want to end the process before they begin the process.

We see that they are successful in that transition.

Tim Butara: So, so it's, it's a, one side of it is kind of this shift in mindset and, and helping People retain the value and maybe provide even more value. But on the other side, you have to probably balance this with some kind of training, some kind of learning focused initiatives, some kind of upskilling so that, you know, they're not just eager and motivated to take better advantage of these new technologies, but also have the adequate skills to get the most value out of these new technologies.

Alon Peleg: Yeah, so who is using today AI, right? Most of the people using AI today are in the industry are engineers, right? Most of the solutions are for engineers, but if we want to extend it beyond engineering, to go to the frontline workers, the tools themselves has to be very simple. It should be simple as WhatsApp.

Okay. We would not be able to go to like tens of thousands of employees and teach them how to use AI. The tools in order to break this barrier, the tools has to be very user friendly, very intuitive. You know, we don't need to teach anyone to use WhatsApp, right? Everyone knows how to use WhatsApp. It should be also for the coming AI tools and platforms that will serve beyond engineers, software engineers.

That's that's the key for success.

Tim Butara: Yeah, I think I think that was a key add on, right? It shouldn't just be user friendly for people who are very adept at technology and very tech savvy But also for like there's this term Citizen developers, I guess and i'm not sure i'm not sure if it's still trending But but I think it applies here, right?

Basically, it needs to be it needs to be user friendly It needs to be enabling not just actual engineers and developers but citizen developers as well Actually, it's probably also one of the One of the key aspects of being a citizen developer, right? It's, it's being able to use these new technologies.

So that's mostly AI in this case to kind of create the same level of compelling experiences that an engineer can do with being tech savvy.

Alon Peleg: Exactly. And for example, we see. It's working, meaning I see employees in the, in the manufacturing floor or on the grocery store or in the aviation industry, which started to use applications and that are replacing very legacy systems and they enjoy it because those applications are like super simple, easy to understand, and that's, that's critical for the success.

Tim Butara: Okay. So besides the user friendliness, besides getting your team on board and, and this focus on people and employees, what would be some other best practices and strategies and tactics for defining and evaluating a successful AI strategy that's able to, to drive ROI insights?

Alon Peleg: Meaning at the end of the day, what we are looking for to see that one of the outcome is real time insights.

If we don't succeed to show insights from the data we have collected that was not collected before or was collected before, but it took time till it was analyzed till management could understand trend. Then that's AI. Okay. People use the, use the solution data as structured data or new data was collected.

Inside immediately was shared with with relevant stakeholders in the, in the company and action were taken that will improve the ROI that will improve the processes. And this will show everyone that there is value.

Tim Butara: Yeah. Data is definitely important, but if I understand it correctly, it's not just data, but, but how you handle that data to, to, you know, get accurate insights and not them be maybe skewed in favor of, you know, whatever you want or, or them be, be poor or, or full of errors due to irresponsible data handling.

Alon Peleg: Yeah. So I will give you an example. Okay. In a jewelry manufacturer, right? Using a, an AI solution, suddenly production issues or defects were found on real time on a specific zone of the manufacturing from a specific machinery and then they can immediately take action and stop the manufacture process on that area till it will be fixed.

How much waste they reduced continuing to produce defect jewelries. By, by leveraging that technology, right? It's it was previous before that AI solution was used. It took them at least a shift till they understood that there was an issue with the process. Another like on a grocery store, when they have to measure temperature of the food regularly in order to make sure that they meet the standards and not selling hot meat, they are now can collect data regularly from hundreds of fridge in all the grocery stores, making sure that the standard of the food quality is, is always in the right level.

So this type of systems, that's the, that's the idea to provide you. Hey, I have all the data, everything is being captured. I have the real time insights. I'm taking action on the spot.

Tim Butara: Do you have any real life examples or stories that, that are kind of, that relate to, to these, to these more, more general examples that you can share with us?

Alon Peleg: Of course I would be happy to do that. So let's take, for example, delivery operation. Okay. You know, each one of us are getting a shipment we took from PS or whatever and then the package in many times are coming to us with problems instead of us starting to open tickets, open the computer, calling the support immediately when the delivery guy is on in front of you.

They can, with the new system, they can open a case for you and with our product, the customer can speak to the application, say, okay, I got the wrong shipment or it's broken and the ticket will be automatically reported to the delivery team. You are now not wasting time on, on the back and forth.

The customer is more happy that. He's being heard and the ticket will be handled ASAP. That's one. In hospitality hotels that, you know, they have like audits. On the quality of the clean, cleanness of the room, right? So instead of with the legacy system cover three rooms per hour, now with our product or other product, they can be much more efficient, cover instead of three rooms per hour, they can cover 10 rooms per hour, making sure the level of, of the rooms are high, and the NPS and the customer satisfaction for those types of Of inspection hotels will be higher customer will be happy.

The, the, of course, the rating for the hotels will be higher. Another example from medicine and manufacturer where the quality control team, and they started to use our application and started to report the wait time during their shift. And they found out that out of a nine hours shift, they are actually waiting two hours for inspector for the cleaning department.

Two hours out of nine hours is a lot of waste. No one was capturing this what, what's the reason behind the fact that they are being delayed? No one, because it was very complex process. They had to, after the fact they had to go to a PC and start writing all the details with with AI. You can just.

Real time speak, explain what the issue, the data will be captured. And immediately they showed that they could save two hours per nine hour shift or the, in food manufacturing with AI, they did the analysis of the setup time. And they, for years, there were the order of the schedule. Of the of the manufacturer was okay.

I'm doing a b c a b c product. But with the data they collected now, they saw that. If they will reorganize the schedule and do B before A because the setup time will be much simpler. They will not waste time between one product to another. And now they added hours of production time to every day.

In the transportation, they have like fleets. Thousands of cars. Every Every day they are doing a one hour checklist to make sure every morning that every car is fixed, that they don't, they will not be stuck on the way. It took every, for every car, every morning, one hour. With AI, first of all, it will be much more efficient.

You can complete it maybe in five minutes. You can also start tracking trends. If you see that a specific car has the same issues again and again, or a group of cars. As a similar problems, you can prevent them. So it's faster. It's a, you can prevent the maintenance. You can decide not to validate everything if you are not finding any issue for a long time.

So all of that, like I gave you a few examples that we, we see today in

Tim Butara: the different industries. Yeah, I remember, I remember speaking about this huge benefit of time saving with Amir already. And also one interesting observation that I've had just now is, you know, all, all of the examples that you gave were basically about physical services or something physical.

And it seems to me like this is, you know, something digital is already optimized. Whereas something physical, that's where you can get real, real value out of AI because there's so much more, so much more room for optimization, I guess. So much room, you're

Alon Peleg: right. And there is so much room for new AI capabilities in those

Tim Butara: domains.

Well, this has been a really great conversation. I think that it was very, very valuable. Very, a lot of good practical insights, especially for anybody who hasn't heard, who hasn't tuned into the previous episode with Iola's Amir Haramati will leave a link in the show notes so that everybody listening can check that out as well because I think that the two conversations tie into each other really well.

But for anyone who would like to dig even deeper, who would maybe like to connect with you, learn more about you, what's the best way for them to reach you and to connect with you?

Alon Peleg: Yeah, I will be happy to speak with people who are interested email [email protected] or LinkedIn. My profile is there and really thank you for the time team.

It's it's a great platform and I will be happy to continue discussing other topics further.

Tim Butara: Thank you so much, Alon. It really also enjoyed our conversation today. Thank you so much for your time, for your insights, for the great conversation. And yeah, hopefully people listening right now will also want to connect with you and, and discuss more things with you more in depth.

Alon Peleg: Perfect.

Tim Butara: Have a good day. You too. And to all of our listeners, that's all for this episode. Have a great day, everyone, and stay safe.

Outro:
Thanks for tuning in. If you'd like to check out our other episodes, you can find all of them at agiledrop.com/podcast, as well as on all the most popular podcasting platforms. Make sure to subscribe so you don't miss any new episodes. And don't forget to share the podcast with your friends and colleagues.