Why the AI adoption challenge is little to do with technology 

Businesses have poured millions into AI tools over the past two years, yet most are still waiting to see a  return on their investment. Sebastien Marchon, CEO of expense management platform Rydoo, explores  how this gap has less to do with the tech itself and more about how confident employees feel using it. 

Nearly every company now claims to be investing in AI and research is consistently showing that C suite executives remain bullish on AI, allocating future budget towards investing further in AI  capabilities across all areas of their business. However, despite the increased spend, companies have  not yet seen a return on their investment. 

In fact, a BCG study of more than 1,250 firms found that only 5% of companies are generating  meaningful bottom line value from AI at scale, while 60% report minimal or no material gains despite  substantial investment. 

This can be due to a number of reasons. Firstly, the failure to create clear frameworks and KPIs when  introducing AI makes it notoriously difficult to track the efficiency gains. Secondly, the lack of a  company-wide strategy on AI can lead to some departments potentially overspending on tools while  others lag, which again makes measuring ROI challenging. And finally, failure to ensure employees are  actually using AI properly can mean that companies aren’t seeing the ROI they expect.  

These three are interlinked, but ultimately, frustration around a lack of progress on AI can result in  companies buying more and more AI tools in order to close the gap between investment and results.  But without clear guidance on how to use tools, employee buy in remains low and overall returns  diminish. 

In short, AI usage needs to be baked into the culture of an organisation in order to see real results.  Leadership must set the direction  

Research shows employees are ahead of their employers when it comes to AI use. A Gallup  study found that while 44% of employees say their organisation has begun integrating AI, only 22% say  their employer has communicated a clear strategy for doing so, and just 30% report any guidelines or  formal policy at all. Separate research from Resume Now found 41% of workers say their employer has  provided no tools, training or guidance whatsoever, pushing more than three in four employees to  find and use their own AI tools instead. 

That mismatch of high appetite paired with low direction is where the risk sits. Employees are going to  experiment with AI regardless, and it’s often with tools that may involve data sharing that the  executive team hasn’t approved. Leadership needs to clearly communicate to employees which AI  tools they are permitted to use, data that is safe to share, and what responsible AI use really looks like. 

During our own company-wide survey, we found that 77% of our employees were already  experimenting with AI, and more than half were excited about its potential. But enthusiasm is only  half the battle, you need a clear plan to turn that into meaningful capability.  

The first step for us was to create an internal grading system to take stock of where we were in order  to form a plan. We created an engaging martial arts belts model, scoring employees on their AI skills  from white through to black. The current distribution, with the majority sitting around the 

intermediate stage, shows that most employees sit somewhere in the middle of the learning curve  rather than at extreme. From there, we could build a strategy that works for every employee. 

Communication matters more than the tech itself  

At this stage in the AI revolution, the question that leaders should be asking themselves is how do we  build an organisation where every employee can effectively leverage AI?  

In order to be effective, the use of AI shouldn’t be a side project but a whole company transformation  initiative. This is the only way to create an AI-first culture, which is where the real productivity gains  lie.  

However, this kind of bold approach needs structure, and it needs to be well communicated. CEOs mustn’t take a solely top-down approach that involves simply telling their competent teams how to do  their jobs. Rather, two-way communication will determine how teams right across the business can best leverage the benefits available to them with AI.  

A recent Thomson Reuters report showed that more than 1,500 legal, tax, accounting and compliance  professionals recommended that organisations curate approved toolsets and build communities of  practice rather than rely on formal training days. 

At Rydoo, that meant setting up an AI Guild, a rotating group of twelve employees from across the  business who meet weekly to flag blockers and share what’s working. It also meant running a  dedicated weekly session open to the whole company, covering everything from live demos to  workflow-building exercises. The shift to an AI-first culture was less about the rules themselves and  more about giving employees a voice in shaping them. 

The real measure of success  

As budgets tighten, the bar for AI investment is rising. In PwC’s latest Global CEO Survey, 56% of chief  executives said that AI has produced neither higher revenue nor lower costs over the past year, and  just 12% have achieved both. BCG’s research points to a similar split, with the top performing “future  built” companies generating 1.7x the revenue growth and 1.6x the EBIT margins of the rest. 

The number of licences bought or training sessions delivered says little about any of that. The more  useful measure is whether AI use has become a repeatable part of how people work rather than an  occasional experiment. As access to the technology becomes more widespread, the gap between  organisations where employees feel confident and supported and those where they are left to work it  out alone looks set to be what separates the companies pulling ahead from those standing still.