# Reducing Alert Volume for a Logistics Platform

> TechSupport Team supported on-call operations for Atlas Freight’s Kubernetes and Prometheus logistics platform. Over five months, client PagerDuty records show weekly alerts falling from 2,100 to 500. The actionable share rose from 31% to 88% in Prometheus.

- Publisher: TechSupport Team
- Canonical page: https://techsupport-team.com/case-studies/logistics-platform-alert-fatigue-sre/
- Updated: 2026-11-16

Client: Atlas Freight · DevOps, SRE & Cloud · US logistics software (170 staff)

## Client and engagement facts

- **Client:** Atlas Freight
- **Industry:** DevOps, SRE & Cloud
- **Business context:** US logistics software (170 staff)
- **Delivery model:** SRE support
- **Application:** Logistics platform with Kubernetes + Prometheus
- **Supplied geography:** US
- **Company size in source:** 170 staff

## Application stack and tools

### Application stack

- Kubernetes
- AWS + Azure

### Tools used

- PagerDuty
- Prometheus
- Datadog
- OpenTelemetry

Related services: [DevOps support](https://techsupport-team.com/services/devops-support/)

## Why Support Was Needed

Many alerts had no defined response. The engineers reviewed which alerts led to action and documented a response for each rule the team kept.

## Technical Work

- Removed 700 noisy alerts in the first four weeks.
- Added 40 alert rules with defined actions.
- Documented who should respond and what to check for each retained alert.

## Representative Incident

The team removed one rule it later needed. The missed alert contributed to a 12-minute outage in month two. The engineers restored the rule with tighter thresholds and clarified what on-call should do.

## Results and how they were measured

Table: Reported results and their measurement scope

| Metric | Baseline | Result | Units | Measurement window | Source and scope |
| --- | --- | --- | --- | --- | --- |
| Alerts/week | 2,100 | 500 |  | 5 months | Client PagerDuty |
| Actionable ratio | 31% | 88% | % | 5 months | Client Prometheus |
| MTTR | 44 min | 13 min | min | 5 months | Client Datadog |

## Evidence and Limitations

The five-month reporting window includes the documented alert-removal incident. Alert counts, actionability and recovery time come from different client tools.

## Alert Removal Review

The team documented the twelve-minute outage and restored the rule with tighter conditions and a clear response action.

## Scope a Similar Engagement

Share alert volume, examples that lead to no action and the rules the on-call team can safely change. [Request a support plan](https://techsupport-team.com/contact/) or explore [DevOps support](https://techsupport-team.com/services/devops-support/).

## Related pages

- [DevOps support](https://techsupport-team.com/services/devops-support/): The related service for this engagement.

## Contact

Request engineers from TechSupport Team or ask a question: https://techsupport-team.com/contact/
