What Agentic Testing Means for SAP Cloud Migration
Meet the Authors
Key Takeaways
The shift from AI as a copilot to agentic AI signifies a move towards systems that understand context, make decisions, and actively work alongside SAP teams, addressing the challenges of rapid cloud innovation.
Tricentis emphasizes AI governance that mirrors HR management, allowing teams to onboard, monitor, and manage AI agents to ensure compliance with corporate standards and prevent rogue scenarios.
The focus of SAP testing is transitioning from maximizing test coverage to maximizing risk coverage, allowing teams to prioritize business impact while efficiently managing increased code velocity.
Artificial Intelligence (AI) in software testing has traditionally focused on assistance in the form of a copilot. However, this model is now deemed insufficient, as SAP professionals manage cloud migrations, where the traditional model cannot keep pace with the velocity of quarterly or weekly updates.
According to experts in a recent Tricentis webinar, the industry is moving from AI that assists to AI that acts.
The Shift to Agentic Intelligence
Markku Riihonen from Tricentis’ Product Marketing Team described this evolution as moving AI from a copilot to a teammate. “Unlike a passive tool, an agentic system is something that can understand context, make decisions, and take meaningful actions for us,” he noted.
Explore related questions
This distinction is critical for SAP environments. In the legacy on-premise world, upgrades were infrequent and massive projects. In contrast, innovation in the cloud is continuous. As Riihonen explained, “Code velocity is exploding, and development teams are shipping up to 10 times more code than they were doing before.” Therefore, manual testing, or even basic automated scripts, cannot keep pace without creating a bottleneck.
A Digital Workforce, Not Just a Tool
During the webinar, the speakers highlighted three components of Tricentis’ AI strategy, each of which will help SAP users journey toward this future:
- Bring Your Own AI: Model Context Protocols (MCPs) let developers use their preferred coding assistants while staying connected to quality standards.
- Prioritize Agentic AI: Specialized agents—specifically Tricentis’ future quality and performance agents—will work alongside QA professionals as quality coworkers.
- Centralized Agentic Management: A centralized AI workspace will enable customers to strategically manage agents.
The most compelling insight from the webinar was the human element of managing this technology. Brad Purcell, Principal AI Strategist at Tricentis, framed the AI workspace as an “HR system for your AI agent.”
“It gives you the ability to onboard a new agent like onboarding a new coworker. You can define corporate standards and rules around it, ground them in the SDLC process, monitor performance.” This ensures that agents don’t go rogue and that they adhere to enterprise compliance, a non-negotiable for SAP users.
From Test Coverage to Risk Coverage
According to Purcell, this technology changes the objective for SAP test engineers, shifting their goal from maximizing test coverage to maximizing risk coverage. This is because agentic AI allows teams to stop burning cycles on the mechanics of writing scripts and focus on business impact.
When Riihonen and Purcell demonstrated Tricentis’ agentic technology, the system analyzed an SAP purchase order requirement, generated a test case, and even assessed that a test case didn’t meet specific performance requirements and updated it accordingly.
Finally, the integration of Tricentis into the SAP Integrated Toolchain suggests that quality assurance is becoming a strategic layer of business process. As Purcell concluded, “Agents aren’t just making suggestions anymore. They’re actually doing the work.”
What This Means for SAPinsiders
Testing is shifting away from a maximum test coverage approach. In traditional on-premise SAP environments, the gold standard meant validating every process for infrequent, massive upgrades. In the cloud, where release cycles happen quarterly or even weekly, this approach is impossible to sustain without bottlenecks. Agentic AI helps here. Instead of mindlessly running every script, AI agents analyze code changes and historical failure patterns to determine which specific scenarios are most likely to break. This allows teams to focus on what truly matters, ensuring that a ten-times increase in developer code velocity doesn’t result in an equivalent increase in testing backlog.
AI governance can resemble HR management. One of the biggest anxieties for SAP professionals is the rogue AI scenario, in which AI tools generate noncompliant data or hallucinate test steps. However, Tricentis has introduced the concept of an AI Workspace, which is effectively an HR system where teams can onboard, train, and even fire AI agents. For SAPinsiders, this agentic mesh ensures that every digital agent acts within defined corporate standards and MCPs. If an agent isn’t performing or violating a standard, the team has the control to remove it, ensuring the human in the loop remains the ultimate decision-maker.
Clean core requires clean testing. SAP’s Clean Core strategy relies on standardizing processes to make upgrades easier. Agentic testing keeps that core clean during operation. The Tricentis webinar highlighted that AI-based testing acts as a quality shield for data. By validating and cleansing data before it hits production, SAPinsiders can prevent the garbage-in, garbage-out cycle. Moreover, connecting tools like LeanIX, Signavio, and Tricentis turns testing from a reactive bug-hunt into a proactive strategic layer of the business process.
