When Efficiency Revolution Turns Into Complexity Hell

Companies are investing in artificial intelligence (AI) tools, but in many cases, they're only increasing work complexity instead of improving productivity. Yonatan Bley, Head of Marketing Machine Learning (ML) at Intuit, recently pointed out in an interview with Dynamic Business that "the problem is companies buying technology based on trends rather than needs."

The promises of AI—customer service automation, administrative task simplification, and freeing up time for business growth—are attractive, but reality is different. In many cases, companies end up managing multiple platforms simultaneously, work processes become fragmented across systems, and the expected productivity gains fail to materialize.

Bley explains that this gap stems from purchasing decisions based on trends rather than needs. The solution is to ask one fundamental question before investing.

Problem First, Tool Later

Step 1: Clearly Define Manual Tasks That Need Elimination

Bley emphasizes that specific problems should define solutions, such as invoice reconciliation, customer inquiry management, or employee scheduling. This approach prevents accumulating technology that solves problems you don't actually have.

The diversity of available AI tools is both an opportunity and a risk. Without strategic direction, companies add layers of complexity while pursuing efficiency. The result is frustrated team members and fragmented workflows.

Only One Pilot at a Time

Step 2: Pilot Only One Tool at a Time with Clear Success Criteria

Real-world testing requires focus. Bley recommends piloting one tool at a time with clear success criteria. "Is it saving employee time? Is it reducing errors? Is it improving customer response times?"

Simultaneous pilots blur results and overwhelm employees, reducing the likelihood of adoption. By focusing efforts, companies can gain accurate insights into whether a specific tool delivers its promised benefits or just adds noise to existing systems.

Training Determines Success

Step 3: Provide Continuous Training, Not Just Initial Deployment

Even promising AI solutions fail when employees don't receive proper training to use them effectively. Bley emphasizes that training must be integrated not just at initial launch but as part of ongoing professional development.

"When you embed AI into daily processes, it becomes a natural part of the workflow, not a forgotten icon on the desktop," he says.

Tools are only as effective as the preparation provided to those who use them. Investment in employee capability determines whether technology transforms operations or becomes another unused subscription draining the budget.

Measure or Abandon

Step 4: Measure ROI Within 3-6 Months, Switch if Ineffective

Return on investment (ROI) is a core indicator of tool value. If AI doesn't reduce workload or improve customer experience within 3-6 months, Bley recommends switching to another tool. The Software-as-a-Service (SaaS) model makes switching costs lower than ever.

Companies can measure AI effectiveness by linking tools to clear business objectives such as efficiency, cost reduction, revenue growth, and customer satisfaction. Tracking quantitative KPIs like error reduction and processing speed improvement alongside qualitative feedback from employees and customers enables comprehensive evaluation.

[AI Analysis] Approach as Experimental Cycles

"Companies must treat AI adoption as experimental cycles of test-measure-improve-replace," Bley says. This iterative approach prevents organizations from being locked into tools that don't deliver real value.

As the AI tools market continues to expand, companies will face more choices. However, as options broaden, strategic selection becomes even more critical.

Successful AI adoption doesn't start with buying the latest technology, but with following the 4-step rule: clear problem definition → focused testing → systematic training → thorough measurement. This framework ensures AI investments create actual business value rather than becoming cost centers.

In the future AI tools market, tools optimized for specific industry problems are likely to show higher success rates than general-purpose solutions. Companies must establish a purchasing culture that prioritizes measurable ROI over flashy features.