How To Take Control Of Your AI Game And Win

“Almost all companies invest in AI, but just 1% believe they are at maturity.” That’s what global research firm McKinsey said in a new report it published back in January and it’s easy to see why that’s the case. While the AI hype cycle remains alive and billions of dollars are still being sunk in the space today, hard questions are getting asked about the ROI that AI really delivers to businesses.

As I wrote before in a previous article, the problem isn’t primarily with the AI systems — it’s about how to cut through the hype and get real value from AI. When companies invest in AI without the right strategy, they are bound to fail. Businesses need to move past seeing AI as a fancy, fortune-costing gizmo that’d magically strengthen their business use cases and shoot them to the top of their industries.

Deciding on the best way to deploy AI in your business can be confusing sometimes, especially since the AI industry evolves at such a fast pace. Just a while ago, generative AI was the major talking point of the industry. Now, it’s all about small language models and agentic AI. Who knows what it’s going to be next?

Instead of chasing never-ending trends and playing catch-up with AI, how can you take control of your AI game and win? Industry experts weigh in on the best ways to do that.

Start With A Clear Purpose

Many businesses start with AI first, then try to force it into their operations. That’s a mistake, according to Peter Ferrari, CEO of Jaca. “Begin with an in-depth analysis of the core needs of the business and its vision, and then determine how AI can be integrated,” he told me. “I’ve observed numerous companies get overly enthusiastic about AI without considering how it benefits their business. That’s how you squander both money and time.”

At Jaca, which offers a unique sugar substitute called Allulose, Ferrari and his team use AI for analyzing customer behaviour. What’s the objective? Anticipate when clients will make their next purchase and which messages will have the greatest impact — a strategy that’s resulted in a 40% decrease in customer acquisition expenses and a 25% rise in customer lifetime value for Jaca, according to Ferrari.

But the defining factor for this result, he noted, was a data-focused approach that helped Jaca to effectively use AI in advertising its product at scale. In other words, the company didn’t just dive right into using AI for its ads. It first used the data from its business to decide the specific places where it needed to use AI.

“We would produce a few ad variations and then wait for weeks to determine which one performed the best. Currently, AI enables us to operate fifty iterations at once, modifying the text, images and call-to-action in real time using performance metrics.”

That’s an example of how starting with a clear purpose could help businesses achieve the right outcomes when using AI.

Define Your North Star

Too often, businesses adopt AI just to keep up with competitors without defining their own North Star — their unique value proposition and differentiators. That’s a mistake, according to Andy Thurai, VP and principal analyst at Constellation Research. As Thurai noted in a previous article co-authored with Joe McKendrick on Forbes, “the secret sauce to AI success is selecting the right business use case — a robust and expansive business use case.”

Some great ways to define your North Star, said Thurai in a LinkedIn post, include “finding your unique differentiators and clearly explaining why customers should care, as well as focusing on how you solve a business problem instead of comparing functions or features with competitors.”

Make Your AI Tools Work Together

While getting your employees to embrace AI is challenging enough, ensuring it operates smoothly across various platforms and teams is a completely different ball game. “AI integration is like managing a music group,” said Ferrari. “You might have four incredible musicians, but if each is playing a different tune, it will sound awful.”

“We have Claude generating content, Midjourney producing visuals, Meta’s AI enhancing ads and Klaviyo managing emails.” However, he continued, “if they aren’t linked, you’re not fully utilizing their potential.”

To close this gap, his company, Jaca uses Zapier, an automation platform that facilitates communication between AI tools instead of keeping them in silos.

Prioritize Ethics And Regulations

Many experts have predicted that ethics and regulations will be the defining issues of the AI era. They are not wrong to think that way, when you consider the harm that AI systems pose to humanity. Take the AI-powered disinformation crisis, for example.

In 2024, a tragic triple murder in Southport, England, led to the rapid spread of misinformation, including AI-generated content, falsely attributing the crime to an immigrant. This disinformation fueled anti-immigrant sentiments, resulting in violent protests and attacks on minority communities, such as the assault on the Southport Mosque.

As Troy McGuire, cofounder of Fintech.TV, said in an interview, when using AI, “it’s important to remember that AI is not infallible and ensure it’s being used responsibly — especially in the media space.” So, even though AI can save lots of research hours, generate content at speed and even help with fact-checking, according to McGuire, it’s important to always have a human in the loop.

“When we use any written material from AI, much of it still doesn’t sound like a human wrote it or how our news anchor would say it. We’ve all seen how many bad articles are being published with AI on our social feeds. This is already an issue. In the near future, we may have a ‘Written by a Human’ stamp on it or something,” he said.

That’s why McGuire believes that if you’re in the media business “you need to evaluate your content and cross-reference AI-generated content when necessary,” adding that journalists need to get the facts correct and be transparent, even when they use AI.

“Not publishing wrong information, whether it’s from AI or a source, will always be our mission. If the audience doesn’t trust what you tell them, they won’t have anyone watching or reading you.”

In Ferrari’s opinion, a privacy-first approach is important for any business that wants to incorporate AI into its business operations. Transparency is crucial as well as regular audits and compliance with regulations, including the GDPR and CCPA, will be important to ensure that customers’ data are properly handled and the most stringent ethical standards are upheld.

“Our customers need to know how their data is being used, and they need to trust that we’re handling it responsibly. That’s why we disclose every AI-driven interaction,” he said.

Measure ROI

The only way to ensure that the value you’re getting from your AI deployment is commensurate with your investment is by measuring the ROI. Sadly, most companies don’t have a standardized way of measuring AI’s impact, so they just deploy AI and hope for the best.

When it comes to getting value from AI, hope is not a strategy. But by defining your specific business use case, aligning AI initiatives with your business strategy, employing cross-functional collaboration a and prioritizing AI ethics, you’re well on your way to getting the right outcomes from your AI deployments.

As Ferrari noted, “balancing AI tools against ROI is crucial as these expenditures can quickly become overwhelming and suffocate the business — not simply financially, but also functionally. The companies that succeed in AI aren’t merely those that implement it, but those who do so more effectively than others.”