In a recent LinkedIn post, Kieran Flanagan delves into the critical distinction between simply adopting Artificial Intelligence (AI) and achieving tangible business results through its implementation. Flanagan argues that many enterprises are currently focused on “tokenmaxxing” – increasing AI usage – rather than “outcomemaxxing,” which prioritizes measurable business impact.
Flanagan highlights the alarming trend of increased AI spending without corresponding gains, stating:
“The average enterprise company is spending 13x more on tokens YoY, even as 95% of AI pilots in those companies fail.”
This observation underscores a significant disconnect in how businesses are approaching AI adoption. According to Flanagan, the ease of tracking AI token usage makes it an attractive, albeit superficial, metric for demonstrating progress. He points out that this focus on usage, often termed “tokenmaxxing,” can be misleading.
The Perils of ‘Tokenmaxxing’
Flanagan contends that an overemphasis on AI usage, without a clear link to business outcomes, can be detrimental. He suggests that employees prioritizing token usage over impact might inadvertently create more problems than solutions.
“The most dangerous employee in any company right now is someone who’s bad at their job and is tokenmaxxing. They’re likely creating more chaos than they are impact. Just because you can create more, doesn’t mean you should. In many cases, you should create better.”
This perspective challenges the notion that simply increasing AI deployment equates to business success. As Flanagan articulates his core formula:
“AI × Outcome = Strategy. Without the outcome, you have just spend.”
He elaborates that while tokenmaxxing can be a byproduct of learning and experimentation, particularly when subsidized by venture capital, it is not a sustainable long-term strategy. Eventually, businesses and individuals will need to demonstrate the concrete value derived from their AI investments.
Measuring AI Outcomes Across Teams
Flanagan acknowledges that measuring AI outcomes can be challenging, especially for teams in marketing or brand development where results are not always binary. However, he stresses the importance of defining and tracking specific, explainable outcomes for every AI use case. For instance, he suggests that a content team’s AI usage could be measured by the time saved in asset creation, while a brand team’s success might be linked to campaign cost, quality, and subsequent engagement increases.
According to Flanagan, most AI use cases can indeed be tied to a measurable outcome. This requires dedicated effort and individuals with a deep understanding of AI’s capabilities to help set appropriate objectives across different departments.
Looking Ahead: The Shift to Outcome-Driven AI
Flanagan advises businesses and professionals to be proactive in this evolving landscape. He believes there will be an inevitable shift towards evaluating AI initiatives based on their demonstrated outcomes rather than just their usage statistics.
As Kieran Flanagan concludes, understanding and prioritizing these outcomes is crucial for navigating the future of AI in business. He suggests that getting ahead of this trend by focusing on strategy and measurable results, rather than just token usage, is a wise move for any forward-thinking organization.
📝 About This Content
This article is based on insights shared by Kieran Flanagan on LinkedIn.
📅 Originally posted on April 23, 2026 | View original post on LinkedIn →