The Practical Application of AI

In the second article based on our event Innovation Driving Growth, we looked at the profusion of data available to businesses and how that could be used.

However, the truth is that our systems nowadays produce so many data points that making practical use of these can be difficult to do.

So in our third article in the series, we are looking at some practical applications of AI together with the benefits and challenges that this brings.

In this article;

  • Where AI adds value
  • Deciding why and where to use AI
  • Developing a practical strategy for AI
  • Summary: if not now then when?

Where AI Adds Value

When we look at the key attributes of a good AI implementation what we see are systems that can take huge amounts of data and collate, process, and draw conclusions that would take humans months or even years to do.

AI systems can make decisions in a fraction of a second based on empirical data and apply those to your business to deliver outstanding benefits.

Ryan England, Founder of Many had an excellent example of this. His company are working with an e-commerce clothing retailer that uses AI to intelligently assess customer body sizes and this has provided a very real dividend in that their return rate (always a problem in online clothing) is now below 1% of sales.

This type of advance doesn’t just improve the bottom line for the company. It also has the benefit that there is an instant reduction in their carbon footprint and the customer who knows that their clothes are much more likely to fit will buy again and again.

Deciding Why and Where to Use AI

There is no doubt that the use of AI and ML in a business can transform the way it goes about its work.

But that isn’t necessarily just about making more profit.

Ryan explains “This isn’t about making more money for our business but about putting our people in the position where they can do their absolute best work.” He continues “ we have very highly paid people with MBAs and Doctorates and yet they are spending up to five hours a week just moving cards around Trello and doing mundane admin things”.

So using AI and ML can help free people from the mundane and manual and allow them to get on with adding real business value.

Does this mean that AI and Machine learning are necessarily right for every business?

Well as Joe Allnutt from Spyrosoft explained, it is hard to think of a business that couldn’t use AI or that isn’t susceptible to attack from AI-enabled competitors

“Every business has data, has a website or other systems so they are all susceptible to disruptors but they also have the option to make a significant change”

A lot of the answers to questions about AI are in fact, just fundamentally good business practice. If your processes or systems are poor and you don’t have a customer focus then adopting AI won’t necessarily solve that.

So businesses in many cases don’t need AI to make a significant change to their operating methods, instead, adopting Robotic Process Automation (RPA) can achieve most of the gains that they are seeking.

Ryan England makes the point that “if you think that you need AI to solve a business problem then you don’t. It’s very easy to think that this is something that you can throw at a problem and it will solve it.”

AI and Machine Learning are tools that you can use to improve once you have an automation solution in place that will then give you more options. 

Where AI and ML can help with almost any business is handling large volumes of data and presenting it in a way that makes sense of what might initially appear to be chaos.

So the starting point has to be a clear understanding of what problem you are seeking to solve and a deep drill down into the reasons for that problem.

Developing a Practical Strategy for AI

Starting on the journey to develop a practical AI solution for your business is a big step but our panellists all agreed that this isn’t a step to be taken lightly or in haste.

It is important to remember that AI will help you with decision making, analysis and data handling, so confining its use to back-office procedures rather than customer-facing touchpoints is often a smart move.

A good example here would be in a restaurant that uses personal service and a friendly atmosphere to generate a USP. AI could be used to improve marketing or stockholding but when it comes to the actual customer experience, you need friendly and efficient waiters.

As Nuno Almeida from Nourish adds, “optimising the booking process using AI will be good, but if the customer experience actually in the restaurant isn’t good then it will be a waste of time.”

It is also true that if you don’t have decent data sources then AI probably isn’t going to help.

Nuno again makes a key point here “The question is where are you getting your data from and how much do you trust that data? Is that data consistent and of high quality?”

So the first port of call has to be to develop your data sources, ensure that they are robust and consistent which will then allow you to develop credible and trustworthy analysis.

From there you need to identify the points in your business that you think you could optimise using AI and what data sets you need to make the change.

Joe also makes a valuable point here when he says “It is important to have an understanding about the high-level topics and ways that AI can help and then you can start to do the right thing for your business.”

A word of caution here though. AI is a very new and emerging technology, so you probably aren’t going to find exactly the solution for your needs or a company that is doing exactly what you want.

Instead, you need to take a side-step and look for ways that AI and ML are being used in other industries and see if a similar practical application would work for your situation.

Developing an AI strategy then almost becomes somewhat circular. You need to know what parts of your business you are looking to optimise and at the same time, you need to have a high-level understanding of what is available on the market.

Once you have these in place, you can start to ask the all-important ‘what if’ questions and that is where massive value can be added.

Summary: If Not Now Then When?

In the final session of our event, Ross Thornley of AQai gave the telling example of Relativity space. A company that has in turn become a disruptor of space disruptors SpaceX and Blue Horizon.

He told us “we all have to be aware that the pace of change is such that your world will be totally different in ten years, maybe even five”

And the truth is that many of us are already customers of companies that are already using AI in the background to give us great service, we just don’t know it.

So the question for all business owners is if you aren’t going to look at AI and ML now, when will you? 

Because leaving it five years might be five years too late.

Join Today

  • Showcase your business
  • Find and attract new talent
  • Post articles and events
  • Support the local sector
Join Silicon South

Get the latest from
Silicon South

  • Receive directory updates, innovation posts and articles straight to your inbox
  • Find out about upcoming events and webinars
  • Be the first to hear about opportunities and talent
  • Average 2 emails per month
  • Unsubscribe at any time