AI adoption has continued at pace throughout 2024, but the vast majority of organisations have yet to embed AI enabled innovation within core operational processes. One third are engaging in limited implementation, while 45% are still in the exploratory phase. While there is no denying the power of GenAI, the majority of businesses have struggled both to identify tangible use cases for AI and determine how best to safely and effectively use the technology in customer and/or employee facing activities.
A number of trends are set to change those figures significantly during 2025. Firstly, technology partners are leveraging AI technologies to deliver packaged solutions based on proven use cases to ease adoption. Secondly, AI is transforming companies’ ability to use predictive analytics across multiple internal and external data sources to achieve the next level in real-time business management, including dynamic pricing. Finally, of course, the deployment of GenAI tools such as SAP’s Joule within public cloud solutions is adding a further incentive to organisations’ digital transformation strategies. Why remain on premise when competitors can routinely explore, innovate and gain benefits from embedded AI in the cloud?
Don Valentine, Commercial Director, Absoft expects Generative AI to evolve from a solution in search of a problem to become embedded in Business-as-Usual activity during 2025 to deliver tangible operational benefits.
Solving Specific Problems
Companies are on different AI adoption curves but, while conceptually exciting, many have yet to determine just how and where AI could be deployed to deliver tangible, repeatable value. This to set to change during 2025, not only as business use cases become more obvious but also as IT vendors and consultants come to market with packaged bites of AI solutions. Simple tasks such as using AI to match electronic bank statements will enable a finance team to move from handling 50% exceptions to perhaps just 5% – and can be quickly deployed.
This packaged approach is helping organisations to identify pertinent business use cases. SAP, for example, is embedding its Joule GenAI tool within its public cloud offerings, including the Success Factors HR and Payroll solution. This native deployment of AI will take the Employee Self-Service facility to the next level, allowing employees to not just view their payslip statements and history, but also ask questions about everything from salary sacrifice contributions to the reasons for tax deductions.
Taking this a step further, an employee will be able to quiz the system to gain a personal view of HR policies, for example to understand the specifics of parental leave, including payment value and leave duration options. Beyond the employee facing solutions that both reduce pressure on the HR team and improve employee engagement, AI can improve business insight. A line manager quickly interrogating the data to understand why head count dropped the previous month, will be able to take a quicker and more targeted response to boost retention.
Dynamic Pricing
Indeed, AI’s data analytics power is even more compelling for many businesses, not least the ability to run predictive analytics across multiple data sources – both internal and external. For example, one Seafood Company has leveraged GenAI to achieve highly effective dynamic pricing models.
Understanding both the likely amount of in-bound stock and also the forecast weather – which affects customers’ buying habits as well as catch volumes – has allowed the company to determine appropriate pricing for the next week or two weeks. Furthermore, with an in-built feedback loop, the business is constantly learning from its pricing model and continuously improving the process to drive additional profit.
The ability to extend the use of AI beyond internal data by folding in other, public data sources is hugely exciting, especially for any business operating in a volatile marketplace. Oil companies, for example, can combine internal data on production volumes with inflation forecasts, projected windfall tax costs, even country specific tariffs, to quickly model likely cash position. This use of historic, current and trusted external data provides a powerful new predictive aspect to business modelling that will also accelerate AI adoption during 2025.
Building Confidence
For the majority of organisation still wrestling with how and where to deploy AI, this ‘packaged’ approach to AI adoption will presage an enormous step forward in both confidence and targeted usage. It will also influence cloud adoption strategies, with AI tools embedded within public cloud solutions reinforcing and likely accelerating system migration arguments.
This productization of AI will not, however, remove the need for careful planning and testing. Indeed, the fact that so many people have already embraced free GenAI tools outside work to summarise documents and fast track research will make it even more critical to ensure everyone recognises the need for robust and rigorous implementation models.
The benefits of allowing employees to ask questions about payslips and HR policies are clear, not least in releasing HR staff to focus on added value activities. But if there are any errors in the AI’s interpretation, the repercussions will be significant. Companies require confidence in their data, the toolset/ solution and the business case and this can only be achieved through rigorous trialling, benchmarking and testing prior to deployment. These tools are enormously powerful – and with power comes responsibility.
Conclusion
GenAI’s accessibility has been key to its rapid growth but, until now, the sheer breadth of deployment opportunities has been overwhelming. Throughout 2025, as IT vendors release targeted AI solutions that address specific business needs, companies will have the chance to fine tune their perceptions of AI and identify the most compelling business cases.
Whether that is within the area of predictive analytics or specific transactional process improvement, external support, such as an SAP partner, will play an important role in allowing companies to exploit these new native AI solutions. Working closely with the business experts, a third party can help to define and refine the boundaries of AI deployment and ensure the company is comfortable with the way it is using AI.
Some may prefer to start with allowing managers to use AI to interrogate data simply to gain a better understanding of business trends, rather than going straight to employee or customer facing usage. Others will be confident in the latter use case and look to improve employee and customer engagement. Either way, a close collaboration with experienced experts will be an important aspect of building up AI adoption throughout 2025, even in an increasingly packaged environment.
