Is AI Bringing You Closer to Your Customer — or Driving Them Away? Here Are 5 Steps to Bridge the Gap The first principle of business hasn't changed: customers pay you to solve their problems, not give them another AI promise. AI can either help or hinder this basic reality.

I'm not going to lie, the latest generation of AI, especially large language models and agentic AI, is nothing short of impressive. At Human Cloud, we used tools like Claude and Windsurf to accomplish in 5 minutes what had previously taken us 5 years.

On the surface, it's a story of overnight magic. But dig deeper and you'll find that the real magic wasn't the AI itself; it was the five years of groundwork that came before. We spent that time using spreadsheets, Canva graphics, CRM automations and hacky off-the-shelf tools to create the right sales and delivery motion, and validate our customers' needs.

Only then did the AI become a true accelerator, as we used Claude, Windsurf and AWS to create the Human Cloud Platform in less than 5 minutes.

This brings up a crucial point. AI can easily be a distraction, prioritizing hype and buzz over real revenue and profitability. Why? Because the fundamental principle of business remains unchanged: every breakthrough starts with a deep understanding of what your customers need.

Before you invest another dollar in AI, ask yourself one question: Is this technology making us closer to our customers, or pulling us further away?

Here are five steps to ensure AI helps you get closer.

1. Manually implement before automating

"Do things that don't scale" is a famous startup moniker brought up by Paul Graham, co-founder of Y Combinator, in his essay in 2013. As a 4x founder myself, this ethos has always run true.

In the case of AI, in every scenario, ask yourself if there is a manual alternative. If there is, try that first, then automate based on customer demand.

Related: LinkedIn's Reid Hoffman: To Scale, Do Things That Don't Scale

2: Capture enough manual feedback

Step 1 is only half the story. The other half is ensuring you have enough of the right type of feedback to automate what really works. My strongest recommendation is to capture feedback that's closest to customers actually paying, engaging and sharing.

I learned this the hard way in a former startup. We spent 3 months listening and iterating on prototypes based on feedback. We were maniacal in the level of detail we captured, from the user experience to the design. Then we launched, and less than 5% of these users actually paid. Instead, we shouldn't have listened to what they said, but instead prioritized what they did.

If you want a book to help you capture the right type of feedback, check out The Mom Test.

Related: How the 'Mom Test' Can Help You Cut Through B.S. and Find Important Answers

3: Make AI accessible for everyone, not just AI experts

Rather than investing in an AI team or hiring AI experts, give everyone an opportunity to apply AI across their team and their work.

Preston Mossman, Senior Director of AI Consulting for Galaxy Square, told me, "learning to use AI is a muscle you have to build. A lot of people self-select out because they can't use AI today to help them, but the first step is to accelerate their comfort and understanding in a way that feels valuable to them."

When asking Preston about ways companies have helped their leaders get comfortable with it, he brought up investing in AI-related tools for interested individuals.

In his words, "if your mechanic told you about a $50 wrench that could get your job done just as well for half the cost, you would buy it for them or find a new mechanic (with the $50 wrench)."

Leaders not using AI in 5 years will be like leaders not using a computer today.

Related: Why Your AI Strategy Will Fail Without the Right Talent in Place

4: Hire independent experts first

Telling someone to use AI with no support is like telling someone to jump out of a plane without a parachute.

Obviously, hiring AI experts as full-time employees would be expensive and out of reach for most of us. Likewise, AI trainings take time, might be expensive, and rarely has direct applicability from training to application.

But a shortcut is hiring individuals who already use AI, as 65% of independent experts were already using AI as far back as 2024, and 95% of independent experts stated that AI makes them more competitive.

This brings up step 4: to hire flexible talent first, with flexible talent defined as independent, freelance, and fractional experts.

The data is clear that flexible talent upskills faster than full-time employees and is ahead of the curve in AI adoption and effectiveness. It's not just AI, Deloitte research shows that the independent workforce upskills faster than their full-time peers.

There are also four massive benefits of flexible talent compared to full-time. You can control cost. You have a quicker time to effectiveness. You learn by seeing their expertise. And the most important benefit is that this is the future workforce.

To get started, look for a flexible talent platform that is specialized in your region, industry, and the application you need AI for. There are over 800 of these specialized solutions.

Related: Solopreneurship and Freelancing Is Here to Stay — Are You Ready?

5: Scale like the cloud

We take for granted how transformational cloud computing has been for us entrepreneurs. Without getting too geeky, what it really did was enable us to scale in line with customer demand rather than taking big bets because of large fixed costs.

Apply this same mindset to AI.

Do you think your AI idea is the next big breakthrough that will transform your company, your industry, and the world? That's great. Now go through steps 1-4 before you bet the farm.