Going deeper
6 min readExpanding use cases and measuring ROI.
Agentic use cases are expanding beyond coding
Trend summary
Organizations expect AI agents to expand well beyond engineering and IT functions over the next 12 months. Research and reporting leads adoption plans at 56%—particularly among mid-market and enterprise organizations—followed by supply chain optimization, product development, and financial planning. The breadth of planned use cases signals a shift toward treating AI agents as enterprise-wide infrastructure rather than department-specific tools.
| Response | Startups & SMBs | Mid-Market | Enterprises | Total |
|---|---|---|---|---|
| Research and reporting | 44% | 57% | 58% | 56% |
| Supply chain optimization | 48% | 47% | 53% | 49% |
| Product development | 43% | 50% | 46% | 48% |
| Financial planning and analysis | 43% | 47% | 51% | 47% |
| Regulatory compliance | 37% | 40% | 45% | 41% |
| Recruitment | 36% | 36% | 44% | 39% |
| Vendor procurement | 31% | 31% | 29% | 31% |
Why this matters
Research and reporting work spans every function and level of an organization, making it a high-leverage starting point that builds institutional comfort with AI agents before deploying them in more sensitive or complex workflows. Organizations that successfully implement agents for research and analysis can establish governance frameworks, build internal expertise, and demonstrate ROI in ways that accelerate adoption for higher-stakes use cases like financial planning or supply chain decisions. The cross-functional nature of these early deployments means capabilities compound across the business, not just within isolated teams.
In addition to coding, data analysis and process automation are the enterprise's most impactful agentic use cases
Trend summary
Beyond coding, the highest-impact AI agent use cases are data analysis and report generation (60% say this is one of the most impactful tasks) and internal process automation (48%). Enterprises are particularly bullish on data analysis and reporting, with 65% citing these as high-impact applications
| Response | Startups & SMBs | Mid-Market | Enterprises | Total |
|---|---|---|---|---|
| Data analysis and report generation | 60% | 56% | 65% | 60% |
| Internal process automation | 45% | 46% | 52% | 48% |
| Managing and summarizing internal knowledge bases | 43% | 39% | 43% | 41% |
| Complex customer query resolution | 33% | 34% | 37% | 35% |
| Competitor and market monitoring | 32% | 30% | 30% | 30% |
| Personalized content creation | 32% | 25% | 23% | 25% |
Why this matters
Data analysis and reporting work touches every part of an organization—finance needs monthly reports, sales needs pipeline analysis, operations needs supply chain visibility. The enterprise enthusiasm is telling, as larger organizations typically have more data, more complex reporting requirements, and more people spending time on analysis work that agents can accelerate or automate entirely. Internal process automation delivers a different kind of value, reducing friction in repetitive workflows that slow teams down but don't require deep expertise. Organizations should prioritize use cases where agents can either amplify expert judgment (data analysis) or eliminate low-value work (process automation), rather than simply digitizing existing manual processes.
Efficiency gains are the primary unlock from AI agents
Trend summary
Organizations expect AI agents to deliver efficiency gains over the next 12 months, with 44% anticipating faster task completion. Enterprises also anticipate an additional benefit beyond velocity: measurable cost savings from their agent deployments.
| Response | Startups & SMBs | Mid-Market | Enterprises | Total |
|---|---|---|---|---|
| Increased efficiency / faster task completion | 41% | 46% | 42% | 44% |
| Higher quality / accuracy of outputs | 33% | 34% | 43% | 37% |
| Improved customer satisfaction / experience | 40% | 37% | 29% | 35% |
| Improved employee productivity or capacity | 32% | 31% | 39% | 34% |
| Cost savings / reduced operating expenses | 31% | 27% | 40% | 32% |
| Revenue growth or new revenue streams | 29% | 29% | 30% | 29% |
| Faster time-to-market for products or services | 23% | 27% | 28% | 27% |
| Stronger compliance or risk management | 20% | 27% | 18% | 23% |
| Creation of new products, services, or business models | 16% | 21% | 17% | 19% |
Why this matters
The split between efficiency gains and cost savings reveals two distinct paths for AI agents today, and both create room for what comes next. Speed improvements help organizations do more with existing resources while cost savings, which enterprises are particularly positioned to capture at scale, come from reducing manual effort and avoiding expensive errors. As organizations mature their agent deployments, these gains unlock entirely new categories of work: comprehensive competitive analysis, continuous documentation, proactive customer outreach—efforts that weren't economically viable before. The organizations capturing the most value in 2026 will be pursuing opportunities that only exist because of compounding efficiency gains.
Leaders expect AI agents to drive ROI across the company
Trend summary
In 2026, software development (57%) and customer service (55%) are expected to see the greatest near-term impact from AI agents, with marketing and sales (46%) and supply chain, logistics, and operations (44%) close behind.
| Response | Startups & SMBs | Mid-Market | Enterprises | Total |
|---|---|---|---|---|
| Software development | 41% | 60% | 61% | 57% |
| Customer service | 52% | 56% | 56% | 55% |
| Marketing and sales | 49% | 45% | 47% | 46% |
| Finance and accounting | 37% | 45% | 36% | 41% |
| Supply chain, logistics, and operations | 41% | 44% | 45% | 44% |
| HR and recruiting | 40% | 37% | 43% | 39% |
| Education and training | 31% | 33% | 42% | 36% |
| Legal and compliance | 20% | 18% | 24% | 20% |
Why this matters
These four functions share key characteristics that make them ideal proving grounds for AI agents: they involve high-volume repetitive work, require fast iteration cycles, and have clear performance metrics that make ROI measurable. The proximity in expected impact across these functions—ranging from 44% to 57%—suggests we're seeing multiple viable entry points rather than one dominant use case.
80% of leaders say AI agents are delivering financial value today
Trend summary
The majority of organizations (80%) report that their AI agent investments are already delivering measurable economic impact today, and confidence is even higher looking forward—88% expect continued or increased returns. This isn't speculative ROI; most organizations are seeing concrete business value from their deployments.
| Response | Startups & SMBs | Mid-Market | Enterprises | Total |
|---|---|---|---|---|
| Agree | 72% | 83% | 79% | 80% |
| Neutral | 19% | 11% | 16% | 14% |
| Disagree | 9% | 6% | 5% | 6% |
| Response | Startups & SMBs | Mid-Market | Enterprises | Total |
|---|---|---|---|---|
| Agree | 79% | 87% | 93% | 88% |
| Neutral | 17% | 9% | 7% | 10% |
| Disagree | 4% | 3% | 1% | 3% |
Why this matters
Organizations have moved past the proof-of-concept phase and into measurable returns, shifting the conversation from "should we invest?" to "how do we scale what's working?" These findings suggest that returns compound as organizations deploy agents across more use cases, refine their implementations, and build institutional knowledge. Early movers are building the expertise and infrastructure that will let them capture disproportionate value as the technology continues to mature.