Artificial intelligence has been in the headlines for years, but recently there's been a major change in focus. The old angle of "AI is coming" has been replaced by a new reality: "AI is here."
Virtual assistants, machine learning, computer vision, natural language processing and other key techs are more capable than ever before. And what's more important is they're no longer the preserve of academics and huge corporations. New tools and services are bringing their benefits within reach of any business who would like to give them a try.
This won't necessarily be easy, because AI is a hugely powerful technology which can transform a business in very unexpected ways. The potential benefits are huge and it's an area you must investigate, but to get the best results you must prepare carefully.
Discover everything AI can offer
Preparing any business for AI starts by ensuring as many key staff as possible understand everything its various technologies has to offer.
Try to avoid any preconceptions. Don't immediately start evaluating technologies you think might be useful, like chatbots, and seek out those in particular. What's more important is to discover new possibilities, new ideas and implementations that you might not have come across or fully understood before.
For example, look at what your direct competition and similar companies are doing. Check out web apps and general applications which you could be using right now, and see if they're adding new smart or AI-powered features. Run more general services for topics that might have interested you before - virtual support assistants, automating CRMs, making smarter decisions within advertising campaigns - but take as wide a view of the market as possible. What you're looking to do is find inspiration and new ideas, not just try to confirm or deny what you believed before.
Identify your business needs
It's easy to be dazzled by shiny new AI technologies, and tempting to make them the focus of your efforts. "Virtual helpdesk bots look great! I wonder if one will work for us? Let's try it and see."
Your starting point should be to understand the issues your business has, and the opportunities that might be available. If your helpdesk is overwhelmed, customers complain about poor service, costs are higher than you would like, and there's a large knowledge base for a bot to work with, a virtual assistant might be worth considering. But that decision should be made because it's one of the most important issues for your business, not because you want to implement an AI project and this is the idea that first came to mind.
Decide where to start
Think about how the kind of technologies AI now offers might relate to your business goals.
Consult as widely as possible. AI tools have all kinds of benefits that might not be immediately obvious, and brainstorming sessions with staff might help generate new ways of looking at old problems.
Depending on your business, there may also be advantages in giving everyone an understanding of the changes that might be coming, how they could affect existing jobs, and what the overall aims of the project are.
Be realistic about what AI can achieve. Customer service bots may only be useful for busy sites that are currently overwhelmed with queries. And data mining and analysis applications work best when there's a lot of information to examine. If you have little data, there's not much any AI-powered analysis tool is going to be able to tell you.
But at the same time, don't feel that your first AI project needs to be something huge. Start small, pick a goal you're sure you can achieve, and work towards that. You'll learn from the process and can get more ambitious next time.
Get your data flowing
AI projects often involve linking data or activities from multiple areas across a business, frequently in ways that haven't been considered before. (That may be what makes the project so appealing in the first place.)
This can sometimes be a real challenge, especially if these different areas use their own technologies and won't communicate easily.
Take the time to iron out these incompatibilities before you start. Invest in relevant IT improvements if it's going to help. This isn't just about one small AI-related project: it'll give you a solid basis for anything that comes next, as well as helping your business to run more smoothly in other ways.
Train staff as necessary
Implementing an AI project can present employees with many new challenges, and the expertise and training they have will often make a huge difference to the results.
Suppose you've decided to create a customer service chatbot, after all. You don't have an in-house developer, so you're using a simple third-party service which enables creating a basic bot without any coding.
This isn't always necessary, and if your needs are very simple, a prebuilt third-party AI tool will be enough. But if you're likely to need more, make sure your employees are equipped to deliver it.
Keep your finger on the pulse
Keep in mind that none of these steps are one-offs; you're going to need to revisit all of them on a regular basis.
Stay up-to-date with AI developments, for instance, especially in any areas that match your current projects and business needs.
Get constant feedback from employees and users on how your new systems are working.
And where it's appropriate, try to think incrementally. Implementing massive new projects on a regular basis may overwhelm and distract your business. Don't be afraid to take smaller steps towards each of your individual business goals. AI is here, but it's still in its early days as a practical technology, and every project will need some time for tweaks and refinements before it's completely matching your needs.