Data Analytics Chat

How To Solve Business/Consumer Problems Using AI

• Ben Parker

Share episode

Send us a text

Becoming an AI and Data Leader: Insights with Bilal Am – Transforming Businesses with AI

In this episode of Data Analytics Chat, host Ben Parker interviews Bilal Ahmed, Chief AI and Data Officer at Altech, about his career journey and invaluable insights into the world of data and AI. Bilal shares his experiences from his extensive career, highlighting pivotal moments, challenges, and the transition from a technical to a leadership role. He discusses the importance of believing in oneself, taking calculated risks, and the role of persistence and resilience in achieving success. They delve into solving business and consumer problems using AI, emphasizing the importance of asking the right questions and having the right data. Bilal also touches upon the hype around generative AI and the need for educating businesses on realistic expectations. This insightful conversation is a must-listen for anyone in the data industry looking to leverage AI for real-world applications. Don't miss out on the valuable advice and practical strategies shared in this episode!

00:00 Introduction and Personal Journey
00:12 Believing in Yourself and Embracing Failure
01:14 Career Highlights and Future Plans
03:17 Transitioning from Technical to Leadership Roles
04:49 Learning and Mentorship
06:54 Key Advice for Innovation and Change Management
07:53 Stakeholder Management and Career Setbacks
13:08 AI in Business: Misconceptions and Realities
35:09 Balancing Quick Wins with Long-Term Solutions
37:42 Conclusion and Final Thoughts

Thank you for listening!

bilal:

I acted as operations, I acted as finance, I acted as marketing or product. And this experience really helped me shape me into a leader I am right now. Yeah. I think first of all, you have to believe in yourself. Yes. You can fall short, but whenever you want to innovate, whenever you want to learn. Failure can be the part of the cycle. But the thing is you have to keep the iteration small. Slowing down just because of the fear of failure is not gonna work moving forward. The world is changing at a very fast pace. Outcome over output. Listen to your customers so you make sure you are giving them the value. persistence and resilience. That is, I would say, the key ingredient towards success and moving forward. Business value first, start with a problem that if solved, create clear business value. Whether that is cost saving, revenue growth, or better customer experience, AI should be means not the starting point.

ben parker:

Welcome to Data Analytics Chat, the podcast where we discuss the world of data AI and the CRI shaping it. I'm your host, Ben Parker, bringing real stories, expert insights, and practical advice to help you in the data industry today. I'm excited to welcome Billow Am. Chief AI and Data Officer at Altech. In this episode, we'll explore his exciting career journey, pivotal moments that have shaped his career path, and discuss how to solve business slash consumer problems using ai. Bilal, welcome to the podcast. Thank you, Ben. Thanks for having me. I'm looking forward to this. So do you wanna start off with introducing your career, Joe?

bilal:

Okay, perfect. I will keep it brief. So I have around more than two decades of commercial experience. The first 10 years I have been working mostly with corporates and I can say that my first 10 years were really like a tech expert in the field of data and ai. And then in 2015, early I shifted to Germany and since then I have become a business data leader. The journey have been very reward full. I have been working with some of the key names like Rocket Internet, then Vox, which was the insurance, unicorn Scout 24, the biggest real estate platform in Germany. And as of now working with Altech. But just to add, I will be also leaving Altech toward a very exciting opportunity. I'm planning to start my own consulting services. I cannot share more. I'm shaping it up with my co-founders as of now. But I really want to add value at scale. That's why my consulting and advisory services is a mission toward that.

ben parker:

Amazing. And obviously all the best for your future endeavors. So I guess, obviously thank you. Obviously you come from a technical background then how do get, how do you find,'cause obviously there's a lot of people in that space. Either feel their fear, the move from technical to the more leadership business type position or they feel they've not got enough training or et cetera. How did you go around gaining that knowledge? Because also you've gone to another country as well, so it must have been a lot of massive change.

bilal:

Yes. Actually, that's a very nice question. The question I asked myself is, after the first 10 years when having a very tech leader, what exactly it's keeping me to the level where I am and what do I need to go further? And the answer was the domain knowledge of the business knowledge, because I have been very tech focused and I was always delivering an output, but the relationship from output to outcome was missing. The answer was listening to the customer, listening to your business stakeholder. And what really helped me was my experience with Rocket Internet. I started a journey with them when we were only nine and 10 people, and I would say then I became from a data leader to a the company wide leader where I acted as operations, I acted as finance, I acted as marketing or product. And this experience really helped me shape me into a leader I am right now. So I would give the very high credit to this experience in Rocket Internet to what I did that helped me to transform.

ben parker:

Brilliant. And then, so was that just like learning on the job then? Or did you get mentors or was it external reading, learning? How did you go about obtaining this knowledge?

bilal:

I think it was a blend of multiple things. I will give the highest credit to learning on the job. And then second of all is curiosity, because you need to be curious to learn a bit more on your own. Everything will not come in a play to toward you. I also, in the meanwhile, did some certification, like project management and planning. When required, then I had a mentor. I would, we cannot underestimate the value that they add to us. I have been reading a lot, so I think it's a mix of everything, but where you convert all that knowledge is on the job. So whatever you are learning, information gathering, I was practicing it in the in the field and constantly sharpening it. And I would also like to add, because if you look into my profile, I have worked with different domains. So over the time, I have also solely built these transferable skills. Which helped me when I go pick up my next challenge, even if it is in a different domain.

ben parker:

Brilliant. I like that. And it's yeah, I think it's, you obviously when you do that, it's, you're building your confidence, aren't you in yourself. And you, I guess you're getting that belief that, look, I can do this. Which is, I think, a massive thing.

bilal:

Yeah. I think first of all, you have to believe in yourself. Yes. You can fall short, but whenever you want to innovate, whenever you want to learn. Failure can be the part of the cycle. But the thing is you have to keep the iteration small. You have to make sure you are learning quickly out of it and finding the insights that kept you behind and constantly evolve. It's the same thing what we are trying to do for companies. We are trying to find insights and make them recover out of their failures, and I applied the same mechanism in my personal career growth.

ben parker:

Brilliant. And then oly, as you progress, what's been the sort of key. Advice that people have asked you from progressing?

bilal:

My advice. Okay. Let me, give me, there are different points that really help me. There is not one advice I would say. One is innovations require risks, so we should be taking calculated risks. We should not be afraid of failures. I'm not saying failure is very good. We should not be afraid of that. We should apply the right methodologies to cover our risks. So we avoid failures. But even if we avoid failure, we have to make sure the cycle is small. We are constantly learning, adapting and innovating. Slowing down just because of the fear of failure is not gonna work moving forward. The world is changing at a very fast pace. Change management is a second pillar. Be ready for change. Unlearning is a key problem I have seen around in people. We all need to unlearn things, how we have been doing it. It's an advice for companies also, and relearn how things should be done in this age of AI era. Other thing, which I would like to mention is stakeholder management. It is a lot of people who are coming from technical background sometimes undermine it. So stakeholder management has become the key focus for data and AI success from the very beginning. How do you understand them? How do you tell them the value will look like so you don't end up over promising them and then under delivering it. Then also how do you constantly communicate with them, telling them where we are and what should they expect and all that. So these are, I would say, the key points I would like to share.

ben parker:

Brilliant.

bilal:

That's

ben parker:

great advice there. So Linda, obviously you mentioned about the position at Rocky in created impact. Has there been any other sort of defining key moments that have you feel have shaped your career?

bilal:

Yes. There are multiple first let me take a step back and say, let me tell you why I mentioned Rocket, especially. The first 10 years I was working as a tech leader, working mostly with the corporates. And then Rocket was my first company when I started to work for a company which was trying to build up. So it really changed my perspective. So it was a big change from that. But then if I look into every job or the challenge, I. It really added value to me and I added value back to them. We can take an example of Vox insurance, which was digital unicorn in the field. When I joined them, one of the key challenge was their growth and they were growing with the help of m and a. So that really helped me understand the business perspective of it. When you are growing for an m and a, what exactly need to be done? Then I was also preparing them for IPO readiness. So that was also a good learning for me. There was a lot of regulatory stuff that need to be taken care of, preparing them for that. And the other thing where, what I would like to mention it is more on the technical stuff, but also enabling the stakeholders We really innovated and learned different operational model that need to be built for scale. Before I joined, they were working in the central operational model. But as we were going into the platform approach and we wanted to scale, we wanted to, we started to adapt data mesh in a cross-functional way. I wouldn't say we took the data mesh a hundred percent additive, but we adapted it to our needs and then we were really able to build a data platform, I would say state of an art, which are. Consumers were able to use in a very self-service manner. So I'm still proud of that delivery, that how we enable the whole company with the foundational data platform, that they were able to answer most of the questions themselves and able to answer how exactly the strategy should be built on top of that.

ben parker:

Amazing. I like that. So then throughout your career, have you faced major career setbacks?

bilal:

I would say we all go through setbacks of some kind of failures, where we learn and all that. So if I reflect back, I would say one thing I would really like to mention and I would like new the candidates in the field to avoid this, is the, like stakeholder management when I was working in my first decade. I delivered a solution which was technically perfect, but it somehow missed what stakeholders actually needed. So I learned that understanding people need is just as important as solving the problem. But the good thing is I learned it quite early in my car career, and that has become the foundation in the second pillar of my career. So I would really advise this to every new person coming into the field is. Outcome over output. Listen to your customers so you make sure you are giving them the value. You're not giving them a technical solution which is not able to unsolved or unlock them. And then I would not say failure. I would say learnings. In the last, in 11 years, I have LED companies, which really did grow at a scale. We inno, we innovated a lot of solutions. And this comes with some kind of risk, we should be able to adapt it. So I would say we have developed at a very fast pace. Some of the features did not work as they were supposed to, but we never said, oh, we failed. We actually learned out of it. We quickly adapted by learning what were the inside wide network and gave a better solution for our consumer. And this is, I would say, has become the foundation of keeping the operation small. If there are smaller setbacks, make sure they are coming to you very early, for example, in few weeks, not in months or years, because then you cannot recover and you have burnt a lot of money for the company.

ben parker:

Yeah, no, I mean it's, you need to obviously deliver to what your client needs, isn't it really? It's key, but it also, I guess on the flip side, at the minute, it's obviously a lot of uncertainty about generative AI and obviously a lot of stakeholders are pushing for it, when really there's probably not a need. How would you go about overcoming them sort of hurdles?

bilal:

So yes we are in a hype cycle and I would say there are two things I will adapt. First of all. Education. We need to educate our business partner what current AI is possible and capable of, so they can think from that perspective. By empowering them with this, I think we can bring them or we can meet MA match at a common ground. And the second thing is I would also really focus on context. A lot of time business leaders or our co-partners from the business have a very generic idea. F really we should go down deeper with them to one level of why. Second level of why and how we do it. Defining what exactly they want to do, what will be the inputs, how exactly the success will look like. Then commit to it. So I think this all goes down to, to educating them right now and then understanding their problem to two or three levels in depth so you, we have clear context from the problem to what consumer pain point it is solving and how exactly we will calculate the ROI of that, because if we are not going into that discussion right now. Unfortunately, this project will fail in six months or in the first quarter. And then these questions will arise. So I think these questions should all shift left before initiating a project.

ben parker:

Yeah, and it's, it's fascinating the hype, isn't it? I think you, you need like a hype for data foundations, don't you? Like it just needs, it needs something to. That's where you get your value from, isn't it? If you're getting like a certain percentage of food gen, AI can't do everything. You need your data, don't you intact? Yeah.

bilal:

Okay. Let me answer this one also, add some of my comments to this one. There is no doubt about that. There is a massive. But this massive hype gives us massive opportunity also. And then we have to build a framework around this, how to make sure we can get the value out of it instead of being trapped into the hype or get into the fear of form, of fear, of missing out. And with this hype, there is also a myth of silver bullet. AI is a silver bullet. At least the business leaders think that we have to come out of that. And going down to my last point, which I said. It is really about the context and the human oversight also that needs to be merged into that.

ben parker:

Yeah. So it's gonna be ing fascinating times ahead, so we'll see how

bilal:

it is. It is.

ben parker:

Sorry. Because Yeah, because people don't, obviously I know it's so much hype from Gen ar, but people don't wanna miss out and things like that, but sometimes it's, yeah, you need to have the right business case, don't you?

bilal:

Yes, the right business case and the right data that I forgot one point from your last question is you were touching data platform, so business case is very important because gene solutions can be very expensive as they scale. So we have to really move left where we are developing. All of this business case will fit into the R-A-R-O-I topic that I mentioned. If you ask me, I. We'll say, we will see a lot of failed gen AI solution in next six to 12 months, because right now most of us are chasing Gen ai without looking into ROI, without looking into, we have the right data or not because garbage in, garbage out. So the quality or the output of the NAS solution wouldn't be good if our data isn't good. And if we see these kind of failures. It'll again raise the importance of the platform. And when I say platform, I wouldn't mention about the tech scalability. Those are all enablement layer. There's those will be required. But the two key pillars which I would really like to focus will be the first will be the data quality. And when I say data quality, let's not boil the ocean, not, let's not try to fix everything. At least fix it for the key use cases so we can move fast forward. And the second would be seman layer. Gen AI and AI offers us so many possibilities, and to get the best out of those possibilities, we have to fix this semantic layer so our AI can understand our data very well.

ben parker:

Yeah, no, and I mean we did have obviously still work in recruitment, like the last couple of years there has been more incr like increasing volume on the data engineering side. But I think you just need to, it needs to just be ramped up again, doesn't it? Just to get businesses more aligned with their data foundations.

bilal:

I would say yes. But then again, in a very balanced way.

ben parker:

Yeah.

bilal:

I'm just sharing some like failures that I have seen Sometime the resources are so skewed toward data engineering and platforming, then we are not spending enough sourcing on the last mile, which is insights and analytical solution or air solutions on top of that, because that is going to transform the data into value. So I would say we need to have a very balanced distribution, so we need platforms. But I would say when we are talking about platform, let's not focus so much on only the engineering part of the platform. My, my focus would be more on the data side of it, which is the data quality data governance and seman layer. And when I mentioned the data governance, let's not take it, we want to slow down everything. Build this data governance in a layer. So that also help us to innovate. It does not block people otherwise. If we look right now, data governance is taken as a blocker for most of us. And in that era where we are moving, where general we have access to everything, we have to make sure governance is there throughout the journey and we have to respect the regulations European regulation and international regulation. So we need to build this friendly governance layer in between.

ben parker:

Yeah, it's just, I dunno if things with data governance is, it's just not that glamorous is it for companies. But it's so crucial to get done. It's, I mean we, I, for us, we see companies are starting to invest more and more, but it's a lot more needs to be done, doesn't it, across the whole industry.

bilal:

Yes, it is not glamorous, but it is required because of the, I would say regulation. We need to respect that also, and I think I believe into that also. So we are not exploiting anyone's personal data, neither our solutions are exploiting it. Also, we want to make sure. The solution that we are building on top of that are explainable and they're not biased. So I think we start, need to start start to respect that, not just because of the regulation, because of our values and belief also.

ben parker:

Yeah. Completely agree. Okay. And then moving to the next question, what skills have helped you stand out in the industry? Do you feel.

bilal:

Okay. Yes. This is a very personal thing. Everyone have their different, for me, there were multiple things, but mentioning some of them, curiosity and learning. I have been very curious always around data, business challenges, business knowledge, and constantly learning for them, not waiting for them to come toward me. So that was one of the key trait I have found out that listening. To our customers, to our colleagues or trends around us often give me the insight that others miss. So that also helped me a lot. The key, I'm a very keen listener and a calm listener. And then I would say I have worked across different areas and domains that also help me be successful because I bring transferable skills. Then the mindset I really appreciate the problem solving mindset. I bring that also onto the table that help us to accelerate and move forward. And the last one, but the, not the least one, and this has become the most important in the last couple of years and will be, is persistence and resilience. That is, I would say, the key ingredient towards success and moving forward.

ben parker:

Brilliant and some great skills mentioned there. So then let's move on to the data topic. So we're gonna look at how to solve business slash consumer problems using ai. So what, in your opinion, what's the biggest misconception businesses have about how AI can solve real consumer or operational problems?

bilal:

Yes. We did touch on this slightly a bit before, but I would say silver bullet myth. This is a big problem that I see around many of us. Think AI is a magic button that instantly fixed problem. You press it, everything will be fixed. In reality, AI is only as good as the data context and the human oversight behind it because we are in the very early journey of it. We have to make sure we are having a good blend of all these three. Again, I'm repeating data, a very clear context, not a high level, because high level context does not bring solution. That is just general strategy, and then human oversight. We have to make sure human insight is incorporated into it, and then reinforcement will constantly keep on improving it.

ben parker:

So with this obviously. A lack of knowledge or misunderstanding. Is it more businesses, I know some Es that do hackathon events where they bring the business and tech together. Is it, is that sort of the sort of things businesses needs to be constantly doing now to gain this knowledge?

bilal:

Yes. I would say there need to be happening multiple things at the same point of time. They should generally be at a board level or a management room level training, which is high level, so they know what to clearly expect because the expectation need to be matched, what can be delivered as of now. But on the second and the third layer, we need to be working in a very cross-functional. We need to educate our product managers with the possibilities. What can happen, what cannot happen to make it happen. Context is required. Collect the context. After collecting the context, we need to figure out whether the data is there or not. Otherwise, we need to acquire the data to keep them into their into the journey. So I would say, as I have been moving very close to the business in the last 10, 20 years, now the business has to move closer to the possibilities of AI and the data also. And we need to meet in the middle.

ben parker:

So it's the communication and change management piece, which is the challenge. A hundred percent, yes. Cool. And that's apparent because obviously it's easier said than done, isn't it really? It's you, when you change, obviously, when you, whenever you change, obviously you can deal with resistance. Yes. You got fear, you got FI mean, you got a lot better. Happens. So it's not easy and I understand that businesses is it's not

bilal:

easy, Ben. I agree to that. It's not easy, but I think we need to start talking about it and carving the way toward it because what is happening if we will not talk about it and if data and AI teams will start to say yes to everything, yes, they will survive six months, but there will be massive amount of failures coming in. So as we always build business or data strategies and then we say, okay, what we will achieve in six months, 12 months, 18 months, in a very similar way, we need to have a AI native data digitization strategy in which we will say, okay, how exactly we will enable not only data and AI teams, how will we will enable the whole company? And in that enablement education is one of the key pillar, then tools and technology and other things. So we need to have a proper strategy instead of just hitting one or two use cases out of fomo.

ben parker:

Yeah, no, I agree. I think it's, I think that's in the future, that's gonna be the next big wave, isn't it? Like educating the businesses. Some are doing it, but I guess again, more needs to be done. I think it's, I think we need, it gets to that point where, oh, we need to do this now. That's when it's gonna happen. Yeah. Business new needs to be a bit more proactive in regards to this.

bilal:

Yeah. You are right. Some are doing it and I was reading an article that most of the the other companies will start to do it after this hype cycle unfolds a bit where there will be a lot of failures. People will start to reflect into it. I think it's our leaders responsibility to bring this visibility into our organization and spread this knowledge so we can integrate this as early as possible. And don't burn the money for our companies and create those failures, because that comes with some disappointment also. So I think we all leaders need to stand for it and create this awareness. Starting from the boardroom, I would say.

ben parker:

Yeah, definitely. Okay. Do you wanna, could you share an example where AI has directly improved a consumer experience?

bilal:

Yes. I have so many examples around that. Let me share few of them over here. Let's take where we did it at a massive scale. I can give you an example of SP 24. So we know that in Germany, a lot of real estate listings, when they come, like most of them are like either empty rooms or the pictures are not in good quality. So what we did was we start, we created an AI solution, which was virtual staging experience where AI furnish the property according to the consumer taste in a single click. So I want to see this room in a modern furnished look. The solution will do in one click This, help them bring their imagination into reality and find their dream home faster. Now, what our consumers can do is, so they can say, okay, these are living room. This is a bedroom. In what way I can fix my furniture? How exactly this will look like Earlier people used to visit, maybe go to the May and plant a lot in their mind. They can see everything in reality. This ad helped us to convert much more people, spend more time on the listings, so we have really solved their problem that we help them to find the right place and do that place fit to their taste. This is one example and we have brought it at scale. Then the second problem was, I would say most of the search as of now is very traditional filter based. Some of the filters are on the first field, and then there is advanced filters. And then you have to find what you're looking for. I'm not talking about just real estate. Most of the filters are like this. We really took a step back and thought, okay, how we can change the life of our consumer that instead the consumer learns how our platform works. We go toward our consumer and learn what they need. And our solutions start to take care of them. So we introduce a natural language search where consumer can speak freely to the system, tell them what exactly they need, and then system will take care of it. For example, a very basic example, Hey, I'm looking for a room apartment with two rooms in Berlin. So it'll quickly do that. This is a basic example. Now moving toward an advanced example. Hey, I'm shipping from Munich to Berlin and I'm looking for a three to four room apartment or a house with schools close by. And right now this filter based search cannot do it. Our vision was to go in this direction where we start to understand more details about that family or that person, and then we can tailor our solution. Our approach was to go in that we have not only the real estate data, we have data about all the amenities around that surrounding when someone says he's looking for a school around, we know that this person have young kids, which mean we should also see if there are any sports related activities or gardens related across, and then we started to suggest them in Berlin. These are the areas where you can find good schools. These are the areas where you will also have, for example garden flows or whatever. So we started to match other characteristics, other dimensions, very close to them. And this was a good success and I'm sure the company's further building on this, but this gave us a lot of new potential. For example, as we are learning more about customer, the new possibilities that unlocked were. This person is shifting from Munich to Berlin. So in Munich, he will have either a partner to a house, which he will either put it on rent or sell. We can get this business from them, a new product where we can create, it's a new lead for us. Then if this person is shifting, we can give them moving service. Also, we have our partners, so this will not only improve their search, I would say this will improve their whole journey around this finding the new home for that. So I believe these are just few example I can share more AI if used right. This can transform the future of how we operate with systems as of now and very soon, AI will know about us much more than we know or we remember and will be able to tailor the solution or give us the recommendations at the right time and way better.

ben parker:

I like that. And I think it's getting more strategic, isn't it? I like your idea where you obviously asking about schools, so they've got young kids and then being at, using the strategic approach to think, oh, they're gonna like kids like sports or any kids clubs, things like that. Being a bit more using you. Your brain

bilal:

if you go brain, ex. Exactly. And the other thing which I would like to highlight again, because with this possibility we are not asking, but the user will just share what they need and they will give us information and cues, which are very helpful to give them the right solution. And with this very classical filter based approach, we are not able to get those views also.

ben parker:

Okay, brilliant. So how should companies identify the right problems for AI to solve? Because I know this is a big challenge.

bilal:

This is big and key challenge for all of us. I would say I will look at this from this perspective. Business value first, start with a problem that if solved, create clear business value. Whether that is cost saving, revenue growth, or better customer experience, AI should be means not the starting point. So there should be a problem. We should not say what AI can solve. We should look into the business value first kind of problem, and AI should come as a solution for that. The second could be, I'm sure a lot of companies already have this pain point mapping. They already have this. We could go to that board, look at where your teams or customer consistently hit bottlenecks or delays or some error. Those high friction areas are strong candidates for AI also. And then I would say before we really get into the implementation, we need to look into data feasibility because AI cannot give you solution if the data is not there. A good AI problem is one where we have. The data or where we can collect the data, and then we need to have a decent quality of that data to train the models. I would really suggest to go business value first. Go through the pain points and before we start to implement, look into the data quality. Do we have it? Can we acquire the data or to build the solution? That should not be the afterthought.

ben parker:

So quite the opposite to a lot of. Business right now where they're looking at tech first and then going for the business problem where really we should look boss, your business problem. Now then you can use the tech to solve where whatever you have, whatever your case is,

bilal:

a hundred percent, that's what I'm doing when someone tell, oh with ai, what can you solve? But I'm asking, what is your problem? Because in the end, we need to solve the key pain points where we can get a good ROI. We deliver good revenue to business. We give good experience to customer. Not just implementing a solution. Maybe we can implement a solution, but that is not the key pain point, which means then the business don't gain much enough and the solution remains on the shelf for a few weeks, a month, and then disappear.

ben parker:

Brilliant. So then. What's more important when using AI solve problems? Having the right data or asking the right questions?

bilal:

That, that's a good one. But I have a very clear take on this question first though, I'm a data leader, but Right. Question first. You can have all the data in the world, but if you're asking the wrong question, you will still get the wrong answer because the solution will not be built in the right direction. So I would summarize it like this. The question defines the direction, the data fuels the journey. So it's a very clear, right. Questions first, so we understand what we are billing. Then data comes.

ben parker:

Brilliant. And then obviously I know, especially in a minute, and everyone's looking for the, oh, quick, easy, quick wins. So how can we balance these quick wins with building long term sustainable solutions that deliver value?

bilal:

Yeah. This is generally challenge for every company also. So what I generally do is show value early. At the same point, build for scale. So let me explain a bit. So I prefer to start with small pilot to prove value. And when we do this pilot, we make sure somehow we can measure the outcome because that's why this pilot is there. But then design them from the architecture perspective that this can scale for the long term also. So we don't have to dispose of dispose the solution that we have built. So this early off. Outcome really gives the confidence to the business that we are going in the right direction. We can measure these things also. So we get good buy in for the long-term delivery tool. And also, I would say we cannot ignore either of them because early wins are required for confidence with the business. And if we acquire the good confidence, then with constant communication we can build things for the long-term strategy Also. For a good balance. I would say value chain view. So balance come from seeking quick wins, but as stepping stone is longer journey, not a separate side project. So what we are building for the small wins also, this should fit into the long term strategy. Of the company. And then also, as I said, we do the architecture in a smaller part. Also, it can be scaled, so it's a fine balance between both of them. But from my experience, I will tell you, if you only build for the long term, yes, which is important, and you are not showing signs of, I would say early wins. There are high chances of failure because of two things. One, lack of trust with the company. And once you are not deploying or delivering early wins, you might not be going on the right track. You're not getting that feedback loop. So these early wins are a good feedback loop also. We are going in the right direction.

ben parker:

Okay. And I guess it, you mentioned it builds trust across the whole business then, because Sly is, it's a massive layout, isn't it, for ai.

bilal:

Yes, data and AI is a trust business. True. And for that this leadership business leadership need to have trust on data and AI leadership also. And that exactly comes with this early wins and creative delivery to them.

ben parker:

Brilliant. Al you've been a pleasure to have on you provided some great insight obviously knowledge, really knowledgeable, and obviously I wish you all the best in your new business adventure. I look forward to hearing more about that and it's, yeah. Thank you for joining the podcast,

bilal:

Ben. Thanks again for having me. I love your podcast and the value you deliver, so I wish you all the success. And one last point for our listeners will be I am co-founding an advisory and implementation company for data and ai. So if you are struggling with what to do with data, what to do with ai, it is not delivering you value or the outcome I'm building a team of specialists. We will be happy to help you and uncover where we can add value. Thank you so much.

ben parker:

T thank you. And yeah, please do reach out to very knowledgeable. Thank you. Thank you, byebye.