The path forward
5 min readDriving AI agent adoption in 2026.
Leaders are optimistic about the business impact of AI agents
Trend summary
Organizations across all segments expect AI agents to deliver meaningful business impact in 2026, with enterprises showing especially strong confidence in the technology's potential.
| Response | Startups & SMBs | Mid-Market | Enterprises | Total |
|---|---|---|---|---|
| Very optimistic | 78% | 38% | 64% | 54% |
| Somewhat optimistic | 22% | 53% | 36% | 42% |
| Neutral | — | 3% | — | 1% |
| Somewhat pessimistic | — | 3% | — | 1% |
| Too early to say | — | 3% | — | 1% |
Why this matters
Enterprise optimism is a significant signal because larger organizations typically move more cautiously—they have longer evaluation cycles, stricter governance requirements, and higher bars for proof of value. When enterprises express strong confidence, it suggests they're seeing results at scale, not just in pilots. Their bullishness also tends to influence the broader market: enterprise adoption drives vendor investment in security, compliance, and integration capabilities that eventually benefit organizations of all sizes.
Data quality and integration are the biggest barriers to AI agent adoption
Trend summary
Integration and data quality challenges top the list of implementation barriers across organizations of all sizes. Nearly half (46%) cite integration with existing systems as a primary obstacle, while 42% point to data access and quality issues and 43% to implementation costs. Small and mid-sized businesses face a distinct challenge: they're significantly more likely to struggle with the human side of adoption, including employee resistance and training needs (51% compared to lower rates among larger enterprises).
| Response | Startups & SMBs | Mid-Market | Enterprises | Total |
|---|---|---|---|---|
| Integration with existing systems | 44% | 46% | 46% | 46% |
| Data access / data quality | 33% | 46% | 40% | 42% |
| Cost of implementation | 48% | 43% | 40% | 43% |
| Employee resistance or training needs | 51% | 38% | 36% | 39% |
| Security or compliance concerns | 36% | 40% | 44% | 40% |
| Lack of internal expertise / skills | 40% | 35% | 33% | 35% |
| Unclear ROI or business case | 17% | 19% | 21% | 20% |
Why this matters
These barriers are predictable and addressable, but they require different strategies depending on your organization's size and maturity. Enterprises need to prioritize technical integration and data infrastructure work upfront—treating AI deployment as a systems challenge, not just a software purchase. Organizations across segments that address both the technical and change management dimensions simultaneously will see faster time-to-value than those focused on technology alone.
AI agents are shifting employee time away from rote tasks and toward strategic work
Trend summary
Agents are shifting how employees spend their time—increasing focus on strategic work (66%), relationship building (60%), and skill development (70%) rather than routine execution.
| Response | Startups & SMBs | Mid-Market | Enterprises | Total |
|---|---|---|---|---|
| Strategic / creative work | 57% | 65% | 70% | 66% |
| Relationship building with clients / colleagues | 52% | 62% | 58% | 60% |
| Learning new skills | 63% | 71% | 74% | 70% |
| Routine / repetitive tasks | 39% | 60% | 56% | 56% |
Why this matters
This addresses one of the core questions about AI adoption: whether it primarily displaces work or elevates it. The data suggests organizations are seeing the latter—agents handling execution while humans focus on judgment, relationships, and learning. That shift has implications beyond productivity metrics: teams that spend more time on strategy and skill-building become more valuable over time, not less. Organizations should design agent deployments with this goal in mind, measuring not just task completion rates but whether people are working on progressively higher-leverage problems. The companies that use agents to develop their people while improving efficiency will build sustainable advantages over those focused solely on cost reduction.
8 in 10 businesses plan to implement more complex agents in 2026
Trend summary
The majority of organizations (81%) plan to move beyond simple task automation toward more complex AI projects in 2026, with enterprises leading this shift at 87% compared to 78% of SMBs. Looking at what "more complex" actually means: 39% expect to develop agents that handle multi-step processes, while 29% plan to deploy agents for cross-functional projects that span multiple teams or departments.
| Response | Startups & SMBs | Mid-Market | Enterprises | Total |
|---|---|---|---|---|
| Building agents for repetitive single-step tasks (e.g., data entry, basic customer queries) | 16% | 20% | 11% | 16% |
| Developing agents for multi-step processes within single departments | 43% | 40% | 35% | 39% |
| Deploying agents for cross-functional projects | 23% | 26% | 38% | 29% |
| Creating autonomous agents for strategic initiatives requiring minimal human oversight | 12% | 12% | 14% | 13% |
| We don't plan to build or implement AI agents | 7% | 2% | 2% | 3% |
| Using AI agents for multi-step tasks (aggregate of the above categories) | 77% | 78% | 87% | 81% |
Why this matters
Organizations are preparing to tackle harder problems with AI—work that requires coordination across systems, functions, and decision points. The companies that identify their highest-leverage complex use cases now can build capabilities and institutional knowledge while others are still focused on basic automation. Think customer intelligence that informs sales strategy, contract lifecycle management that connects legal through procurement, or strategic planning agents that synthesize inputs from finance, operations, and product teams.