One of my colleagues at Oztera is a Project Manager and an exceptional gardener. She also happens to maintain over 1,000 succulents, all thriving and blooming beautifully.
Watching her work has given me the clearest analogy I know for AI-driven performance tuning in Microsoft Dynamics 365 Business Central. Just like her garden, when Business Central is running smoothly, the most important work is happening where you cannot see it.
To the casual observer, her garden looks effortless. Rows of healthy plants. Consistent growth. Color everywhere. But beneath that visible success is constant, intentional root-level care.
- She custom mixes soil so roots can breathe without drying out.
- She plans drainage before planting, knowing that water pooling underground will kill a plant long before the leaves show stress.
- She inspects roots, not leaves.
- She corrects conditions early, not after visible failure.
She never sets it and forgets it. She monitors. Anticipates seasonal change. Prepares long before stress arrives.
That same preventative discipline is how Business Central performance should be managed, with attention below the surface so users rarely notice issues above it.
Over the last few posts in this series, I have written about how AI is reshaping specific functional areas inside Business Central, including AI in Finance, AI in Purchasing, and AI in Accounts Receivable. Those are visible AI wins.
What gets less attention is the invisible foundation that keeps all of those features working reliably: performance tuning.
What is AI-Driven Performance Tuning in Dynamics 365 Business Central?
One of the most common questions people ask AI tools about their ERP system is simple.
Why does Business Central feel slow sometimes?
The answer is almost never random. Performance shifts when operating conditions shift. AI-driven performance tuning detects, interprets, and corrects those shifts before users feel the impact. It does this through Business Central’s built-in performance monitoring services and AI optimization routines, which use telemetry as a primary input signal.
Performance Changes When Conditions Change
Business Central performance is dynamic, not static. Slowdowns typically correlate with measurable environmental changes, such as:
- More users logged in simultaneously
- Increased transaction or posting volume
- Month-end or quarter-end processing spikes
- New automations, extensions, or integrations
- Background jobs running more frequently
- Data growth without indexing or optimization adjustments
AI-driven tuning assumes that performance signals follow patterns, and patterns can be detected.
Monitoring and Telemetry: The Signal and the System Working Together
Behind the scenes, Business Central continuously captures telemetry, detailed operational signals about how the system is behaving, including:
- Page load times
- Posting duration
- Report execution time
- Job queue behavior
- Extension execution impact
- API latency
- Environmental health indicators
However, it is important to be precise about how this works.
Telemetry itself does not “drive” performance tuning.
Instead:
- Telemetry serves as the observability layer, the raw, high-fidelity signal about what is happening in the system.
- Business Central’s performance monitoring services and AI optimization routines analyze that telemetry, along with other workload and system data, to identify issues, detect patterns, and recommend or enable tuning actions.
As Microsoft describes it:
“Monitoring telemetry gives you a look at the activities and general health of your environments/apps, so you can diagnose problems and analyze operations that affect performance.”
Microsoft Learn, Telemetry Overview
In other words, telemetry provides the evidence. AI-powered monitoring and performance algorithms interpret that evidence.
The challenge is that the volume and complexity of telemetry data are far beyond what humans can manually monitor effectively. AI makes that possible.
Like soil imbalance does not immediately show up in a plant, performance degradation often starts long before visible symptoms appear. Telemetry gives visibility. AI-powered monitoring provides insight and guidance.
How does AI help identify and fix performance bottlenecks in Business Central?
Clients sometimes ask, “Did something change in Business Central, or am I imagining it?”
Performance problems rarely appear overnight. They creep in slowly.
- A page takes an extra second.
- Reports feel heavier.
- Posting becomes less predictable.
Humans adapt. AI does not forget.
Business Central’s AI-enabled performance monitoring compares current system behavior to historical norms and flags drift, when something still works, but no longer works as well as it used to.
Is AI Watching Individual Users in Business Central?
This concern deserves a clear answer.
No.
AI-driven tuning analyzes system behavior in aggregate, not individuals. Microsoft is explicit that telemetry is designed to understand system health, not monitor people:
Telemetry data is collected to analyze system behavior and performance, not to evaluate individual users.
Microsoft Learn, Telemetry FAQ
That distinction matters for both governance and trust. Performance optimization focuses on system signals, not employee monitoring.
Microsoft reinforces this approach:
“Telemetry insights detects optimization opportunities and provides actionable guidance based on usage patterns in your Dynamics 365 environments.”
Microsoft Learn, Telemetry Insights
In practice, this shifts performance work from reactive troubleshooting to guided optimization. Teams get direction before disruption, not just diagnostics after it.
Can AI forecasts and predictive analytics improve ERP performance in Business Central?
From my perspective, AI performance tuning is most valuable when it predicts issues before a user notices them. This is where AI becomes a true invisible ally.
Predictive analytics improves ERP performance not by directly making the system faster, but by forecasting load, risk, and bottlenecks so teams can act early.
Performance issues stem from usage patterns, data growth, processing spikes, and extension behavior. Predictive AI helps surface those patterns early and translate them into operational decisions.
Forrester describes this model as AIOps:
“AIOps leverages advanced analytics and machine learning to process vast amounts of data, helping IT teams proactively identify and resolve issues and make more informed decisions.”
Forrester
Efficiency gains do not only come from smarter transactions. They also come from keeping the platform consistently responsive.
What AI Forecasting Enables Teams to Do
Predicting Load Before It Becomes Slowdown
AI can forecast:
- Transaction spikes
- Posting peaks
- Report and batch surges
- Integration traffic growth
This allows admins to:
- Reschedule heavy jobs
- Stagger batch processing
- Adjust job queues
- Add capacity before impact
Detect Emerging Bottlenecks
Predictive signals include:
- Rising page load times
- Longer posting durations
- Query degrading with data size
- Extensions increasing execution time
Instead of waiting for complaints, teams get early warning indicators and tune sooner.
Guide Extension and Customization Decisions
AI pattern analysis highlights:
- Extensions tied to slow transactions
- Automations causing downstream delay
- Processes degrading fastest with growth
That supports better decisions around:
- Extension design
- Automation scope
- Integration frequency
- Customization tradeoffs
In gardening terms, this is fixing drainage before roots rot, rather than trying to rescue the plant later.
Performance Tuning vs. Service Scalability in Business Central Online
It is also important to distinguish between two related but different concepts.
- AI-driven performance tuning refers to how efficiently Business Central runs within its allocated resources. This is shaped by monitoring services, AI routines, and telemetry-driven insights.
- Service scalability in Business Central Online, how Microsoft’s cloud infrastructure dynamically scales compute, memory, and throughput to meet demand.
Performance tuning focuses on how the system behaves.
Service scalability focuses on how much capacity the platform provides.
They work together, but they are not the same mechanism.
What common pitfalls should you avoid when tuning Business Central with AI?
AI tuning is powerful, but not automatic. Key pitfalls include:
Data and Technical Risks
- Poor Data Quality
- Departmental data silos
- Over-customization
- Weak security controls
Strategic and Operational Pitfalls
- No defined objectives
- Over-reliance on automation
- Weak change management
- Assuming AI is “plug-and-play”
How to Avoid Them
- Clean and structure data first
- Keep human oversight for critical decisions
- Start with focused pilot use cases
- Monitor models and adjust for drift.
- Align tuning efforts with Business Central’s monitoring insights and scalability behavior
What End Users Actually Experience
When AI-driven performance tuning is working well, end users feel the difference every day.
- Faster page loads
- More predictable posting
- Fewer timeouts
- Stable performance during growth
Forbes describes AI’s ERP role this way:
“AI enhances an ERP system’s ability to process and analyze vast amounts of data in real time, facilitating better decision making, system efficiency, and operational productivity.”
Forbes
Performance stability is one of those efficiency gains, even when users never see the mechanism behind it.
Final Thought
AI in Finance, Purchasing, and Accounts Receivable gets most of the attention because those capabilities are visible and tangible. But none of them succeed without a stable, well-tuned foundation underneath.
When Business Central feels fast, stable, and dependable, that isn’t luck.
It is AI-driven performance tuning, powered by Business Central’s monitoring services and AI routines, informed by telemetry, and supported by scalable cloud infrastructure, working quietly in the background, learning from real usage, detecting stress early, and optimizing continuously.
Just like a thriving garden, the most important work happens underground.
May your roots stay strong and growth stay healthy,
Mike Stallmann
Oztera
About Mike Stallmann
Meet Mike Stallmann, Director of Product and Business Development, Co-founder at Oztera, and the original “Chief Geek Juggler.” With decades of ERP innovation under his belt and over 200 successful deployments, Mike’s involvement with business technology is extensive.
From wineries to agriculture and beyond, Mike and Oztera specialize in solving complex, industry-specific challenges. If you’re looking to leverage technology for growth and efficiency, our experience is your secret weapon.
For insights and actionable advice, connect with Mike on LinkedIn and discover what tech-driven business transformation looks like.