AI has the power to drive or diminish revenue depending on the effectiveness of implementation
AI is already boosting revenue in pockets of many organisations, particularly in functions like engineering and sales.
However, many companies are still failing to translate those wins into company-wide financial impact.
That is the warning from Dan Lawyer, chief product officer at Lucid Software, who in conversation with HRD, warned businesses of the issues that arise from an ineffective tech stack.
He said many organisations are running into a new kind of “last mile problem” with AI: the technology exists, the ambition is there, but the operational groundwork to turn pilots into profit at scale is still missing.
“Those organisations that are seeing real ROI (return on investment) from AI are likely succeeding because they’ve built the foundation for it to truly thrive,” Lawyer said.
“While individuals and specific departments… are driving revenue through efficiency, many organisations struggle to scale these gains.”
Lawyer draws a parallel with logistics: companies can optimise warehouses, trucks and routes indefinitely, but if the package never makes it to the customer’s front door, the value evaporates.
“Most companies still struggle to fully integrate, scale, and redesign their operations around AI because they're missing the foundational elements needed for enterprise-wide success,” he explained.
Data from Lucid revealed how shallow those foundations currently are, with just 16% of employees claiming their workflows are “extremely well-documented”, while 61% believe their organisation’s AI strategy is not well-aligned with its operational capabilities.
The result, Lawyer argued, is that AI often ends up accelerating dysfunction rather than fixing it.
“Without proper documentation of processes and knowledge, AI simply accelerates existing inefficiencies rather than solving them. If you automate bad data, it’s still bad data.”
Nearly half of employees reported struggling to find the information they need to do their jobs – a fragmentation problem that is now being passed directly into AI systems. Instead of enabling faster, smarter decisions, poorly structured data and ad hoc processes create new bottlenecks and slow decision-making across the organisation.
Fragmented tech stacks as a hidden tax
At the core of the issue is a messy, overlapping technology landscape inside many companies. Over time, teams have accumulated a patchwork of tools, systems and point solutions that don’t talk to each other cleanly.
“Fragmented tech stacks force teams to manage redundant tools and inconsistent workflows,” Lawyer said.
“Not only is this acting as a hidden tax that drains an employee’s time, but it’s also driving up operational costs and creating inefficiencies that make it difficult for AI systems to integrate or access unified data.”
This disjointed environment erodes the very conditions AI needs to deliver financial returns. When information sits in silos and workflows differ from team to team, it becomes harder to automate at scale or to build reliable AI-driven experiences for customers and employees alike.
Process clarity is emerging as a critical enabler. Further research from Lucid revealed 34% of employees said process documentation is a top tool that would help them adapt to AI.
This proves many workers are not asking for more AI – they are asking for better-defined work so AI can be applied meaningfully.
Governance gaps fuel hesitation and cost
Another barrier sits not in the tech stack but in governance. The research revealed that only 45% of companies have established AI ethics guidelines. Meanwhile, 42% of knowledge workers express concern about misusing AI due to unclear rules.
Lawyer says that uncertainty is materially slowing adoption and inflating costs.
“When teams don't know what's allowed or how to use AI appropriately, they either avoid it entirely or use it incorrectly. Both of those scenarios waste resources and block meaningful adoption,” he added.
In practice, that can translate into AI investments that sit underused, or a sprawl of unsanctioned tools and experiments that require additional oversight and clean-up. Either way, the distance between AI’s promise on paper and its real impact on revenue keeps growing.
A split between the C-suite and the front line
The readiness gap is also visible in who is actually using AI. Lucid’s research revealed that executives are adopting AI at much higher rates than entry-level staff – 54% compared with just 21%.
On top of this, nearly half of junior employees say they feel hardly or not at all knowledgeable about AI-powered features.
For organisations betting on AI to drive top-line growth, that divide is significant. Revenue impact hinges less on isolated executive use and more on how deeply AI is embedded in everyday workflows across sales, service, operations and product teams.
AI literacy, training and change management are quickly emerging as key determinants of whether AI becomes a genuine revenue lever or just another executive talking point.
To move beyond pilot projects and convert AI into measurable fiscal value, Lawyer urged companies to act on three fronts:
- Streamline workflows and documentation
- Close the adoption and training gap
- Align AI strategy with operational capabilities
As organisations enter the next phase of AI adoption, the narrative is shifting from experimentation to execution. Localised gains in efficiency and revenue are no longer enough; boards and executives are increasingly focused on whether AI can reshape performance at scale.
For Lawyer, the answer depends less on the sophistication of the models and more on the discipline of the operating environment they are dropped into. Consolidated tech stacks, clear processes, documented knowledge and robust governance are emerging as the real differentiators.
