Do you wonder about what AI will mean for your business, and whether you should be using it?
Before you spend anything on AI, we believe you should ask these 7 questions:
- What is a problem worth solving?
- Do I have data that I can access?
- What are the biggest benefits of AI (for my company)?
- What are the biggest risks of AI?
- How much is AI going to cost?
- How is ROI (Return on Investments) of AI Calculated?
- What are the Next Steps?
In this article we’ll give you some suggestion that will help you answer these questions for your company. But before we get into that, first things first: why should you care?
Our guess is that if you are reading this article, you’re an optimist when it comes to technology. You know that you can use new technologies like AI to become a leader in your industry. And you also know that if you don’t make the time to get this technology working for your company, you could fall behind the competition.
At the same time, deciding what to do about AI is tough.
On the one hand, AI will transform every industry. The impact it will have is going to be far greater than the change from steam power to electricity. The CEO of Google says AI is going to more important to humanity than fire or the internet. Let that sink in: more important than the internet.
Imagine that you could travel back in time to the beginning of the internet, and what you would have done differently (other than registering the domains for IBM.com, McDonalds.com, and Toyota.com). No matter what business you are in, the internet changed it.
Well, we are at one of those moments again, except the opportunity is even greater with AI.
On the other hand, there is a big problem for businesses. For all of its potential, most AI projects fail to deliver value because businesses get matched to the wrong solutions. In a whirlwind of excitement, time and money gets dedicated to ideas that were never going to work.
These risks leave many leaders doubtful about their ability to proceed, and feeling that they deserve independent advice before making any big decisions.
That is where we come in. We only care about your success, and provide solution-agnostic advice on tapping into the power and potential of AI.
Because of that, before you spend anything on AI, we believe you should ask these 7 questions.
- What is a problem worth solving?
From the beginning it makes sense to frame AI in a strategic lens, rather than a technological one.
Focusing on strategy plays into the strength of most executives. This is where they can contribute, and ensure that anything that is invested will align with the strategic objectives of the company. Without even realizing it this focus on strategy will limit the possibility of wasting time and money on a “proof of concept” or “pet project” divorced from the strategy.
Then we suggest that you look for micro decisions, or steps, in your processes. Find an important process, and look at it from the point of view of your customer. How could their experience be better? What decision could you automate that the customer would really care about?
Similarly, you can think about how to embed AI in your products, to make them smarter. Again, the value of automation should be appreciated by your customer.
Following this approach ensures you work on something important (aligned with your strategy and important to your customer) yet manageable. You’re only starting with a small decision point. You’re not trying to change the entire company all at once.
Instead, get this right, learn from it and then move on to the next related micro decision.
2. Do I have data that I can access?
Data is a must have for AI. No data, no AI.
The good news is that you almost certainly do have data. Remember data might be in the form of text, audio or video files. It might be in your emails, or in your CRM.
At a minimum, you can access publicly available data. For example, if you wanted to know your customers better you could access what they’re posting on social media or third-party review sites. While this information can be incredibly useful, it is harder to build a competitive advantage based upon it, because all of your competitors have access to it too.
Because of this,the best kind of data could be data that is specific to you, or your partner organizations. Think about what you and your partners know about your customers, your market, your product and so on. If you can make use of this kind of data you can start to build a moat around your business.
As you start making up longer and longer lists of data sources you’ll get excited. You might even recall the phrase “data is the new oil” and believe you’re sitting on an untapped resource.
This is when you should take a breath.
Notice that this question is “do I have data that I can access?” Too many companies get excited about the first part of the question and ignore the second. That’s when they get into trouble.
Here is a metaphor we often use when explaining the ability to access data. Imagine that vendor offers to sell you a lightbulb. If you’ve only ever had candles, the lightbulb is so much better, and would certainly increase the productivity of your business.
Now imagine that you bought the lightbulb and went to plug it in, but there was no electrical wiring. Without the electricity the lightbulb won’t work.
Just like a lightbulb needs the electrical wiring to be in place for the electricity to flow, AI projects need certain infrastructure in place to make sure that the data can flow.
Making sure that before you make significant investments in AI that this problem is on your workplan.
3. What are the Biggest Benefits of AI (for My Company)?
The biggest advantage of AI is that it can learn. You can teach it something, and then it keeps getting better and better over time. In a real sense, the miracle of machine learning is already behind us.
This is actionable information for leaders. Peter Diamandis, entrepreneur and founder of Singularity University, is fond of saying that “by 2030 there will be two kinds of businesses: those that fully adopt AI and those that are bankrupt.”
The change is coming faster than people think. Doing nothing with AI exposes your business to the risk that a competitor will appear that provides the same level of service at a fraction of the price.
Rather than being disrupted, getting motivated to adopt AI can help transform your business into an industry leader.
As well, if done properly, AI will take over the parts of a job that most people do not like, freeing them up to carry out higher value activities. As a result, in making this transition you may come to see another benefit: your team will love it.
AI and humans have very different strengths. AI can find insights in data sets so large that no human could even hope to grasp, and intelligently automate tasks. But AI also has limitations. AI is typically not suited for situations that can change rapidly or problems that cannot be clearly and concretely defined. And AI usually doesn’t have much of an understanding of cause and effect. Humans are good at these things.
The best results come not from AI working in isolation, but in combining the power of AI and humans each doing what they do best.
4. What are the Biggest Risks of AI?
The biggest risk with an AI project is that it will fail. Approximately 80% of all AI projects fail because businesses get matched to the wrong solution.
Failure means wasted time, and effort. These costs are easy to track, and can be in the millions of dollars at large organizations.
Failure means wasted opportunities. These costs are harder to track. What is the cost of not improving your product or service? What is the cost of not moving when your competitors do?
Failure is also demotivating. Teams become more jaded about the potential of AI, and become less willing to support other projects going forward.
Even under the best of circumstances, launching an AI project is a bit of an experiment. It is not the same as purchasing a new piece of software and expecting it to work out of the box.
Leaders should be aware that the field is new. They should know that, because of the complexity of the field, what worked in one organization, or one part of the company will not necessarily work in another part. Organizations that are thriving with AI proceeded through an iterative process, and become more mature over time.
Therefore, part of the value of an AI project is learning how to repeat the process, and become better over time.
5. How much is AI going to cost?
As a leader, this will require and investment of time by you. Getting an AI project right requires direction, otherwise it will drift into irrelevance.
You will also need to have talent. At a minimum, from your organization you’ll need the team running the business function, and your IT resources. You will also need the AI talent, which you can hire, or contract out for.
For a larger project, it will probably be cheaper to hire your own staff. The challenge is in knowing what skills sets to hire for (do you need a data scientist, data engineer or something else?) and then finding that person. The job market for AI staff is extremely competitive.
Last, there is a need for a financial investment in technology. The least expensive option is to use “off the shelf” AI tools for specific functions (see our catalogue). Many are available for a few hundred to a few thousand dollars per year. Or you can hire a consultant to build something that is customizable to your business. A small project can be done for a little as $10k – $25k. Only once the project proves its worth, and you know it is scalable, should you invest anything further. While projects can easily grow to being in the hundreds of thousands of dollars, you should only be paying for that once you’ve established that the returns will exceed that investment.
6. How is ROI (Return on Investment) Calculated with AI?
We believe that you should have two ideas for ROI in mind.
First, we think there are hard measures of ROI, where the benefits of the project need to exceed the costs. Benefits can be tracked in several different ways:
- Increased revenue that could come from better product recommendations, or smarter email campaigns.
- Increased efficiency, or money saved. This could come from reduced salaries in one area, or speeding up the time it takes to complete a task.
Second, you should think about harder to measure benefits too. The most common ones could include:
- Improvements to non-financial measures, such as customer satisfaction, that are important to a business.
- Reduced risk. This could come from minimizing the risk of human error in a transaction, or limiting the chance that you didn’t adequately search all your documents for a key clause in a legal dispute.
Even though it is hard to calculate, you should remember that part of the value of completing a project is the learning process. Doing one project well sets you up to do even better the next time. Each project is an opportunity to improve your AI literacy and capability. Over time you will want to get better and more comfortable with this new way of work. Practice makes perfect.
7. What Are the Next Steps?
Do not try to solve all of this on your own.
You can start by calling your team together, and asking them questions from our blog. That will get you started down the right track.
We also suggest that you get some independent, vendor-agnostic advice. You will want someone that has no financial interest in whatever decision you make. And you’ll want someone that is equally comfortable speaking with business leaders as technical teams.
That is what we do.
If you’d like to schedule a free 30-Minute Breakthrough Consultation. We guarantee that you’ll come away with insights on how AI can transform you into an industry leader.