“Achieving AI” Isn’t A Milestone
AI is everywhere.
It’s touted as a panacea for every problem for every company in every field. Every software company has recast and reframed itself as “AI-driven.”
As a result, industrial companies feel a lot of pressure to implement AI. Their board members and shareholders don’t want to miss out.
There are good reasons to incorporate AI into businesses but it’s not a solution for every problem. Many of my initial conversations with prospective new clients have a bit of a FOMO direction. Approaching any buying decision this way can result in lost time and dollars.
“Achieving AI” isn’t a milestone. The measurement is the results these tools bring to your organization.
First, gain some perspective on what AI can do for your business by asking the right questions: about your business and then about the AI solutions you are considering.
How not to think about AI for your business
AI is not a general blanket application. But companies are turning to their trusted IT partners (who are also experiencing the same pressure to "get AI now”) and saying: "We need AI."
The response is typically, "Sure, we'll sell you 5 pounds of AI right now."
What happens if we stop a minute and ask the simplest question: "Why?"
AI was previously opaque, esoteric and unapproachable technology. When developers gave AI a chatbot interface in the form of ChatGPT, the technology became more accessible to the masses and the floodgates opened.
There's so much excitement around AI it's created peer pressure, rushing us toward adoption. Before you invest in that shiny new wrapper, ask the important question of all: What is the business problem and is AI the best solution?
AI hype vs. AI reality
At the end of the day, you’re simply trying to find the right tool to make your business run better. Since technology offers us many tools, the question is, which is the most ubiquitous at solving the problems you’re facing?
AI allows us to rethink business processes and work smarter. It allows you to go after problems that you wouldn't have considered previously. With AI, you can cover a lot of ground.
But approaching AI from the perspective of “everyone else is doing it” will only give you buyer’s remorse. Instead, approach AI with a goal in mind. For example, I want this technology to tackle a workflow that takes too much time. Or I want this to cut my labor costs. Or I want this tool to make my customer’s online experience more intuitive.
You now have the luxury of rethinking workarounds you’ve accepted as standard operating procedure (SOP). You can finally start building efficiencies around those challenges that slow operations and sales or increase risk. I'm talking about core business problems you've accepted as "the way we do things here."
AI value out-of-the-box
You can even get value right now from AI.
The most often overlooked out-of-the-box AI opportunity is Microsoft's Copilot. It's easy to integrate it into your everyday tasks, whether it's composing a sales email or writing computer code. I use it daily, making me about 30% more effective. It is there to assist, not replace. It's an accelerant, and it's excellent at that.
How to evaluate AI solutions for your business
What if you spend the money and get the wrong solution? Execute the same level of due diligence toward AI as you would toward any other technology tool.
The reality of any SaaS product is that they’re built for the mass market. They are generalized implementations targeting average problems. You can't tune it precisely for the customer; even the customization features are based on the law of averages.
But AI gives us the opportunity to rethink and rework how we conduct business. No two businesses run the same. So, when an IT vendor talks with you about AI, they should ask you a series of questions about your business — instead of talking about features and benefits. If they’re not asking you these questions, consider it a red flag:
How does your business work?
What matters to your customers?
What are your biggest business challenges?
What workarounds have you created?
How have you tried to solve these problems?
Did your solutions create more challenges?
This evaluative process aims to find the sweet spot where AI could offer real value for you.
Everyone's under pressure to do AI — including your IT team — but very few understand what it means. It is not even reasonable to be an expert. The math behind AI is intense and real. What we understand is evolving so rapidly that I’d be suspicious of anyone who says they’re not continually learning new things in this space.
Ask vendors to explain what their tech does and how AI plays a role in their solution. Ask why you should care.
Don't let hype overshadow your scrutiny of AI-powered products. When a vendor comes to you with an “AI product,” ask:
Have you tried to solve this before? How did it go?
What’s the difference between your solution before, without AI, and now, with AI?
How does it work? (Keep asking this question until you’re satisfied.) Any vendor you’re considering needs to be able to relate their tool to you in terms you can understand. Don't let anybody wave their hands and say, “It’s very complicated.” Because the truth is, if they can’t explain it to you, they don’t understand it well enough to sell it to you. If you don’t understand, keep asking.
Do you have references?
Was your product created from an in-house model? Was it bespoke and created by your team? What is it built on? For example, Microsoft’s AI-enabled solutions has an OpenAI (the maker of ChatGPT) backbone. Microsoft is an investor in OpenAI. But other companies are building their AI features using the same underlying platform. That doesn’t make it bad — but you should know. As with any other technology, you may not be comfortable knowing it was built on something the vendor doesn’t control.
Anyone that does technology consulting right now has tacked on AI support into their service model. Ask them: “How many AI models have you shipped into production?” and “How are those models performing in a live environment?” and “How are they being maintained?”
The field of AI has been around for a long time. What's new is the marketing hype pressuring us to "Get AI! Quick!" The reality is that a good IT consultant with real experience can analyze the problems you want to fix and build the most precise models to solve them.
Remember, AI doesn't have to result in a grand initiative. Scope a test project, develop, implement and measure the impact. For example, focus on applying automation to improve productivity.
At the end of the day, you have a business to run. Can this be a tool in your arsenal? It can be. It’s great to be excited about AI. But remember to execute the same due diligence you’d apply to any other technology tool in evaluating AI. Return to the basics.
If you take the word “AI” out, would it still make sense for your business?
Signal Mine solves unsolvable problems with custom solutions in industrial sectors. We apply our experience in AI methodologies to create highly targeted technologies to solve quantifiable business problems. Start the conversation with our team to find out more.
