The use of artificial intelligence (AI) is becoming mainstream. Many vendors embed AI technology in their products faster than many organizations can scope and deploy their own in-house AI initiatives. Although generative AI created hype and headlines since it exploded onto the scene in 1Q23, in many cases, it often does not yet deliver on its potential. Although interest in AI, and generative AI in particular, remains high, our data show that plans to adopt generative AI have weakened—down to 72-82 percent of organizations, compared to nearly 90 percent reported in 4Q23 (see the Research Insight “Generative AI Smoke Is Clearing; Data Leaders Need A Game Plan”).
Lower planned adoption of generative AI is driven in part by rising pressure to deliver clear business returns on this spending, and less inclination to simply invest unchecked in generative AI in the hope it will somehow and eventually prove beneficial to the organization. User attitudes in enterprise performance management (EPM) reflect this sentiment, with generative AI remaining interesting but perceived as a lower priority, compared to AI use cases that would have a demonstrable beneficial impact—such as predictive forecasting and automating EPM processes.
Data leaders therefore should resist the temptation to over-invest in generative AI. Instead, they should look to leverage AI-enabled EPM applications that deliver functional, incremental enhancements to their EPM environment based on prioritized use cases. EPM is a mature technology, and embedded AI now represents the only area of significant differentiation among vendors. Embedded AI therefore offers early adopters an opportunity to build on their existing EPM investments to create new sources of competitive advantage. Also, users consistently show a preference for AI capabilities in EPM to be embedded in vendors’ standard EPM software offerings, rather than developed in-house. This inclination also represents a lower-risk approach to delivering positive returns on AI investments.
Data leaders therefore should explore how embedded EPM AI capabilities can address high-priority use cases. They will need to exercise a degree of caution, because support by EPM vendors for the use cases we have identified remains mixed. For example, only 31 percent use AI to automate routine accounting activities in EPM, the second-highest-priority use case. However, many vendors plan to add more AI capabilities in their EPM solutions during the next 12 to 24 months, which will make it challenging to distinguish between available product capabilities and potential future deliverables.
Although early adopters will have an opportunity to leverage AI capabilities in EPM to create potential sources of competitive differentiation, data leaders will need to balance marketing promises against realistic capabilities delivered by vendors. Prioritizing potential AI capabilities by use case will identify areas in which potential value gained outweighs the risks commonly associated with early adoption. For organizations evaluating EPM solutions, embedded AI capabilities also represent a means by which to differentiate among vendors—because embedded AI capabilities represent the least mature area in an otherwise highly mature EPM software market.
An opportunity also exists to clarify further the business benefits of generative AI. Half of EPM vendors offer generative AI embedded in EPM. This level of support represents an opportunity for data leaders to explore generative AI’s use in a specific business context, which may help build business cases within EPM and more broadly throughout organizations. Data leaders should ensure they include key representatives from their finance teams in any cross-functional AI-related initiatives and familiarize themselves with current and planned embedded generative AI capabilities from their current EPM vendor (or vendors).You do not have permission to access this document. Make sure you are logged in and/or please contact Danielle with further questions.