Ultima’s Post

Transcript

Hi there and welcome to UIQ TV. My name is Chris Kirk. I'm the Head of Data and AI here, Ultima. And today we're going to talk about data, AI and business process automation in terms of what it really means to you in your business, especially how you're going to drive data and AI through your business. How are you going to adopt AI, how automation fits into that journey and how Ultimate can really help drive that forward for you to get the most out of it. O We've seen the explosion in AI capabilities over the last 18 months, and it really has been amazing revolution, technically speaking. But it really has blurred the lines between what automation brings to the table and what AI is capable of, and also raise some concerns and some skepticisms around data privacy, data security, and what is being used with my business. Data. Well, Jen, AI and ChatGPT are grabbing the headlines. There are other areas of AI that are really driving forward our capabilities for business. We have computer vision AI, we have predictive analysis based AI as well in in the mix. They're equally come on song the same way Chegg, BT and Jenna I have in general. So what it really comes down to for businesses is how can I drive adoption for this technology? How do I select which part of AI is going to give me the most value? And it can be really tricky to understand which way to go with AI. And the important thing with this is it's all about data. 90% of AI projects are based around a good understanding of your data. And that's really where this story starts. O In the realms of AI and automation, the way to think of this is AI and automation are the engine, but data is the fuel. Without fuel, you're not going to be going anywhere. You need to be able to answer fundamental questions around data. Where is my business data? Is it secured? Is it good quality? Am I able to gain a source of truth without any conflicting information in my data? These questions need to be answered upfront before we dive into the actual nuts and bolts of that engine. Now the critical thing here is data and information, strangely, are not the same thing. Data is raw. Data gives me statistics. But while this is important and critical for me to understand my data, it's information that's going to give me the value I need to transform my data through filtering, through processing, through analytics. To get the information that's going to be valuable for my business, that can be used by automation and AI to really drive that capability forwards. Absolutely and. It's so critical to get this right to start off with because 70% of automation and AI projects fail. To be completed because you don't have a good enough understanding of your data and the information that's needed from that data to begin with. Firstly, we need to identify if this is an AI problem or an automation problem. From the aspect of hitting the ground running, automation is generally what's going to get you those quick wins. And by automation we mean processes that are repeatable, that you can identify a relatively easily, that have predictable inputs and already determined outputs. Getting that automated and automating those day-to-day repetitive tasks is what's going to give you productivity gains and efficiency gains over the business to really Dr. prioritizing your staff, keeping your staff happy, making sure that they are doing meaningful work for your business and not focusing on manual mundane work day-to-day. We can get caught U with. Humans making mistakes as bad now when you're dealing with repeatable processes. You quickly kick into muscle memory. We as humans make mistakes. It's in our nature. You cannot help it when you're doing something over and over again. O By automating these day-to-day repeatable processes through automation, you're not only improving accuracy, but you're reducing the likelihood of human error. And you can even design it so when something unusual happens in a particular process, you'd still backing off to a human through the automation to be able to. More actually pick up on those random little problems that you don't quite expect. Through our travels at Ultima, when we're working with customers through business process automation, what we're generally seeing is back office style processes are the ones that get targeted for automation. So approving invoices, for example, reviewing inventory data, integrating systems together to pass data from 1:00 to another when you have multiple different systems through mergers or acquisitions for example, but. One of the biggest ones we see are HR based processes, so especially. Looking at your joiner, your mover, your lever processes, automating those processes not only gives you more freedom to concentrate on more fun work at the end of the day, but it also improves the experience for that new joiner. You want to make a good impression for your new staff. You don't want to deal with permission issues, missing equipment, things not quite working right. If you can automate that completely or even partially, you're going to. Really improved business satisfaction and Dr. employee retention at the same time. And for your levers, you're making sure that you're not leaving yourself exposed as a business. You're making sure they don't have access to things they shouldn't have access to when they leave your business. Things that ultimately do get missed when staff leave your business and you're trying to do this manually. Well, First off, it comes back to data. Now you might not think that business process needs a lot of digital based data, and you probably be right because what we mean here is process data, the knowledge of those processes. If you don't know how a process works, if it's not written down or defined anywhere, then ultimately it's not really a process. That's your starting point for a lot of businesses to understand how this process. Works end to end before you can go ahead and automate it. When you then get down to the automation, what we generally see are customers having challenges with integrating systems together. You will have an old legacy system that it's too expensive to get rid of. It's come about because our merger or acquisition and you've got a new shiny CRM style system that you want to integrate it with. It's integrating old applications that don't seem to work. An automation perspective and integrates them with the new world is what we have had a lot of experience in Ultima and what really drives automation engagements here with our customers. AI in effect is able to give scale and speed and analysis a bit like a human, but at much faster rates, at much greater accuracy based on the data that you're feeding that particular API service. And as mentioned previously, there's generally 3 sub versions of AI that we we focus on with customers. So you've got generative AI that's used for transcription or summarization of large amounts of data to give a really detailed summary without having to wade through lots of documents. We've got predictive AI or machine learning where we're going through a large amount of data and we're looking for anomalies in that data and we're looking for trends in that data to try and predict what's coming down the road for our business. And lastly, we've got computer vision as well. Where we're able to analyze images or video for what is actually in that image or video? What are we looking for in that content? And it sounds slightly strange to say, but a computer to look at the picture of a strawberry, a random example for you. And to tell you it's a strawberry is a massive technological feat that's 20 years ago was practically impossible. We completely take it for granted as a human being that you know through knowledge growing up as a child what things are. And for computer to do the same is incredibly complicated, but very impressive in terms of what we can do in this day and age. That's a really good question and one that I want to preface with the way I started my IT career about 18 years ago. And back then, virtualization was the revolution. But not to trivialize it, but virtualization was aimed to do one thing and one thing very well from an AI perspective, being the revolution that it is, it's very nuanced. You can do lots of different things. It's not aimed at one particular use case. And therefore it's through us and our capabilities are Ultima and with our delivery partners that we can really help customers focus on delivering what they need for the use. Cases that they have when they're driving that AI adoption. Absolutely. And it's been really interesting seeing different use cases out there and when the Marxist question, there's always 2 that spring to mind. The first one is doing some work a few years ago with a distributor around fish. Believe it or not, they would have to supply, cook, deliver the fish to supermarkets as a distributor. And effectively what they were doing as part of that process is they were manually inspecting cooked fish to make sure it looked right for delivery to the supermarket as a visual inspection. And that was a very manual process for them, so through AI. Through computer vision AI, we were able to add in cameras to take images and video and a design a threshold of what good quality fish needed to look like to pass a visual inspection. And through AI that meant they could really scale and speed up that process. They could reduce the amount of manual effort needed, but still have manual effort in play for where something fell outside of that threshold and they could have manual inspection for those. One or two odd use cases where they needed to have a bit more of a look of what was going on, but it's a great example of AI really helping their business. O another is an indoor farm that we have done some work with as well around AI use cases. This was effectively looking at how they could reduce waste in their business by planting fruit and vegetables in a more optimal way based on weather information, weather patterns. Humidity sensors, temperature sensors, growing the right crops at the right time of year and it just meant that it really optimized what they were doing as a business. Reducing that waste effectively reduced the overheads for their business. And in many of their customers were more happy because they got fresher fruit and vegetables. There was less fruit and veg being produced that wasn't quite good enough quality and it means they could really succeed as a business. Without a lot of manual work and a lot of manual intervention, O as you've seen from this video, this is quite a complex subject in general, but there's just a few takeaways I just want you to consider taking away from this video. It's all about understanding your data. It's all about looking where automation can take your business to the next level. And it's all about thinking around those use cases with AI where automation isn't going to do it, but maybe using some of the AI. Examples that we've given in this video can be used to really help your business take off O it's through Ultima and our experience and our expertise around automation, data and AI means that we're in a great position to help you in your business. So please go to our website, ultimate. com to get more information or have a chat with your account manager and let's see what we can do to try and help you in this particular area. Finally, I just like to say thank you so much for joining us for this. Video You can find more video content via our website thatultimate.com or via our YouTube channel. And just leads me to say see you next time.