Case Study: From Manual Control to SLA Operational Intelligence

Agromarket is one of the leading agribusiness companies in Serbia and the Western Balkans, with a history dating back to the early 1990s. As Agromarket's support operation grew, SLA compliance became harder to control. Contractual expectations were clearly defined — but the system itself did not reliably enforce or measure them.

When an SLA breach was questioned, teams had to reconstruct timelines manually. Reporting required validation. Escalations triggered friction.
The gap between contractual expectations and system reality was widening. This was no longer a configuration issue. It was becoming an operational risk.

Where the System Was Breaking

The problems surfaced in daily work:

• As the support team expanded, the company wanted better insight into how customer requests were being handled.

• They needed a reliable way to measure response quality and track key service metrics — such as when a ticket arrived, how long it took to provide the first response, and how much time was spent resolving it.

• While Redmine managed tickets effectively, it did not provide built-in SLA tracking that reflected these needs. As a result, managers lacked a clear overview of response performance and service reliability.

Three structural risks emerged:

• Operational inefficiency

• Reporting inaccuracies

• Internal and external conflicts around accountability

The customer did not need more dashboards. They needed control and RedmineX provided.

The Intervention

We approached SLA not as a timer feature, but as an intelligence layer inside the system.

1. A Central SLA Logic

We introduced a global SLA matrix defining response and resolution targets for each priority (P1–P4). This became the single source of truth across the entire environment.

Instead of scattered rules per project, SLA behavior became predictable and consistent.

2. Real-Time Visibility Inside Every Ticket

Each issue now contains two live SLA indicators:

• Response SLA

• Resolution SLA

These are not static timestamps. They are countdown timers that:

• Start immediately at ticket creation

• Count down toward the deadline

• Continue into negative time once breached

This shifted SLA from retrospective reporting to proactive risk visibility. Teams no longer ask, “Did we miss it?”
They see risk forming in real time.

3. Priority-Aware Automation

When a ticket’s priority changes, SLA recalculates instantly.
Redmine behavior now mirrors real operational urgency. Escalation automatically adjusts timing expectations without manual intervention.

The logic adapts as reality changes.

4. Vendor and External Dependency Protection

One of the most important improvements addressed fairness.
When tickets move into defined waiting states (vendor response, external information), SLA timers pause automatically based on status mapping.

This prevents:

• False breach reporting

• Misaligned performance evaluation

• Internal escalation conflicts

The system now clearly separates internal performance from external delays.

5. Visual Risk Indicators

To make SLA understandable at a glance, we introduced optional visual signals:

• Green → Safe

• Yellow / Orange → Approaching risk

• Red → Breached

This made SLA status transparent not only to support teams, but also to management and non-technical stakeholders.

6. Structured Reporting

For the first time, SLA performance could be filtered within the Redmine issue list and analyzed directly in the environment:

• By priority (P1–P4)

• By outcome (met vs. breached)

• By responsibility (internal vs. vendor-related)

SLA moved from operational noise to management insight.

7. Custom Notifications

In addition to the SLA functionality, we implemented a set of custom notification rules tailored to Agromarket’s workflow. While not directly tied to SLA logic, these improvements strengthened overall communication inside the system and reduced the need for manual follow-ups.

The new notification framework allows the client to:

• Define custom notification triggers aligned with their internal processes

• Send notifications in two languages — Serbian and English

• Trigger a single notification across multiple workflow statuses

These improvements significantly enhanced the overall notification experience, ensuring that relevant updates reach the right teams at the right time while keeping configuration simple and manageable.

The Impact

• The change was immediate and structural.

• Manual SLA tracking disappeared.

• Escalation discussions shifted from opinion to data.

• Reporting no longer required reconstruction.

• More importantly, accountability became clear.

• Operational teams use SLA indicators daily to prevent breaches before they happen.

• Team leads identify bottlenecks early.

• Management now has real compliance visibility and performance trends they can rely on.

• This project was not about adding timers.

• It was about turning SLA from a contractual obligation into a real-time control system.

• The most valuable outcome was not technical.

• It was trust — that the data is accurate, performance is measurable, and decisions can be made based on facts, not assumptions.