> Markdown version of [/videos/348-sre-methods-in-an-agency-environment?t=2127](https://www.wearedevelopers.com/videos/348-sre-methods-in-an-agency-environment?t=2127). Every page supports `.md` or `Accept: text/markdown`. Links point to the HTML versions so they work for humans too. Agent guide: [/agents.md](https://www.wearedevelopers.com/agents.md).

---

# SRE Methods In an Agency Environment

Exhausting your error budget every month isn't bad luck—it's a flawed engineering process. Discover how to successfully adapt SRE principles for fast-paced agency environments and seamless client handovers.

- **Speakers:** Martin Beránek
- **Event:** WeAreDevelopers LIVE
- **Published:** January 26, 2022
- **Duration:** 44:31
- **URL:** https://www.wearedevelopers.com/videos/348-sre-methods-in-an-agency-environment

## Summary

Implementing Site Reliability Engineering (SRE) within an agency model introduces unique challenges, as the typical software lifecycle bridges the agency, the direct customer, and the end-user clients. A foundational step is standardizing the strict definitions of Service Level Indicators (SLIs), Service Level Objectives (SLOs), and Service Level Agreements (SLAs). While SRE is sometimes dismissed in fast-paced agency sprints as non-agile, establishing these metrics pre-production is critical for setting realistic baseline expectations. Initial threshold estimation often requires synthetic clients and load benchmarking, but these objectives must be periodically evaluated after launch. A critical realization is that exhausting a project's error budget every single month indicates a fundamentally flawed engineering process that requires system-level recalibration, rather than just continuous operational fire-fighting.

As agencies transfer applications out of development and into the hands of customers, successful adoption hinges on preventing knowledge silos via a "holy trinity" of documentation: the SLO document, support playbooks, and postmortems. If a customer assumes maintenance, the SLO acts as a functional baseline for their internal SRE teams; if the agency retains operations, it serves as a strict commitment value. Postmortems must remain completely blameless, compiled within a day or two of the event while memories are fresh, and must generate concrete action items that prevent recurrence. Effective incident management during active downtimes relies heavily on strict role division—specifically an Incident Commander who maintains absolute operational authority and delegates tasks without being derailed by status-update calls from project managers.

Beyond basic operational workflows, agencies must proactively account for unexpected environmental overhead, such as abrupt cloud ecosystem changes or massive egress costs on platforms like AWS. Assessing security vulnerabilities, such as those surfaced during NPM audits, should be directly mapped to the project's error budget policy to scale development urgency accordingly. Finally, securing the project handover requires meticulous Identity and Access Management (IAM) audits to strip unused permissions. Relying on fully managed, native solutions like AWS or GCP Secret Manager—rather than maintenance-heavy alternatives like HashiCorp Vault or insecure CI/CD variables—ensures a seamless, secure administrative transition to the customer.

**Keywords:** site reliability engineering, service level objectives, software agency handover, error budget policies, incident management roles, blameless postmortems, support playbooks, synthetic client benchmarking, cloud iam auditing, aws secret manager, gcp stackdriver monitoring, npm vulnerability severity, cloud environment events, infrastructure egress costs, agile metric standardization

## Chapters

1. **Defining service level terminology for agencies** (00:23) — The core differences between SLIs, SLOs, and SLAs dictate how availability thresholds and penalties are applied.
1. **Mapping the agency and customer relationship for software delivery** (01:35) — Understanding how customers, agencies, and end-users interact clarifies the context of application delivery and service agreements.
1. **Overcoming agency objections to reliability planning** (04:04) — Project managers and developers often push back against service level agreements when striving for minimum viable products.
1. **Introduction to aki agency and operations context** (04:49) — A fast-paced software agency structure relies heavily on DevOps workflows and standardized application deployments.
1. **Establishing realistic SLO documents before production** (06:50) — Expectations must be managed during pre-production to set rational latency and availability error budgets.
1. **Defining project handover scenarios and knowledge silos** (10:50) — The operational approach shifts depending on whether a customer intends to maintain an application or requires ongoing agency reliability support.
1. **Structuring three core documents for incident support** (14:03) — Service level documents, support playbooks, and postmortems function together to streamline continuous reliability improvements.
1. **Writing blameless and detailed incident postmortems** (15:58) — A constructive framework structures postmortems with root causes, impact assessments, timeline logs, and actionable resolution items.
1. **Assigning strict roles for incident response teams** (19:33) — Clear escalation boundaries isolate the specific duties of an incident commander, communications lead, and operations lead during an active issue.
1. **Accounting for unexpected cloud provider environment events** (23:07) — Proactive communication strategies mitigate the broader stakeholder impact of sudden architecture shifts or pricing changes enforced by cloud vendors.
1. **Aligning automated security issues with error budgets** (26:14) — Predefined error budget policies accurately determine the prioritization response required for automated dependency security scans.
1. **Managing cloud access credentials during project handovers** (28:25) — Routine practices like rotating identities and trimming unused IAM permissions ensure project infrastructure remains secure upon client transfer.
1. **Transferring robust application secrets to external clients** (31:01) — Native cloud secret managers usually provide a safer external handover experience than self-hosted solutions or simple continuous integration variables.
1. **Conducting a transparent support and toolset adoption period** (32:49) — Stripping away agency-specific automation constraints and reviewing past postmortems directly empowers client teams during operational deployment transitions.
1. **Advocating for SRE practices within agency environments** (35:27) — Educating stakeholders early and focusing directly on end-user experience ultimately secures project management buy-in for sustainable reliability work.
1. **Answering inquiries on SLA negotiations and observability tooling** (38:01) — Strategic responses to audience inquiries detail practical approaches for resolving early service agreement disputes and selecting native tracking metrics.

## Related Moments

- [Integrating service level objectives into incident management](https://www.wearedevelopers.com/videos/854-serverless-observability-where-slos-meet-transforms) (from "Serverless Observability: where SLOs meet transforms")
- [Applying software development methodologies to incident response](https://www.wearedevelopers.com/videos/101-applying-agile-principles-to-incident-management) (from "Applying Agile Principles to Incident Management ")
- [Navigating specialized roles and toolsets across engineering teams](https://www.wearedevelopers.com/videos/680-handling-incidents-collaboratively-is-like-solving-a-rubix-cube) (from "Handling incidents collaboratively is like solving a rubix cube")
- [Scaling shift left practices within large engineering organizations](https://www.wearedevelopers.com/videos/233-what-developers-get-wrong-about-application-quality) (from "What Developers Get Wrong About Application Quality")
- [Shifting software delivery bottlenecks to operations and incident response](https://www.wearedevelopers.com/videos/100332-software-that-fixes-itself) (from "Software That Fixes Itself")
- [Balancing feature releases with reliable service targets](https://www.wearedevelopers.com/videos/854-serverless-observability-where-slos-meet-transforms) (from "Serverless Observability: where SLOs meet transforms")

## Related Articles

- [Events like RSAC Get You CISOs. Developers Decide What Actually Gets Deployed.](https://www.wearedevelopers.com/magazine/693-events-like-rsac-get-you-cisos-developers-decide-what-actually-gets-deployed)
- [Now is the time for industrialized software development](https://www.wearedevelopers.com/magazine/601-now-is-the-time-for-industrialized-software-development)
- [Walking Into The Era of Supply Chain Risks](https://www.wearedevelopers.com/magazine/106-walking-into-the-era-of-supply-chain-risks)
- [MLops – Deploying, Maintaining And Evolving Machine Learning Models in Production](https://www.wearedevelopers.com/magazine/115-mlops-deploying-maintaining-and-evolving-machine-learning-models-in-production)

## Related Jobs

- [Senior Software Engineer, Enterprise Products](https://www.wearedevelopers.com/jobs/ext/1841248-senior-software-engineer-enterprise-products) at **GitHub**
- [Staff Software Engineer](https://www.wearedevelopers.com/jobs/ext/1425755-staff-software-engineer) at **GitHub**
- [Staff Software Engineer, Copilot Experiences](https://www.wearedevelopers.com/jobs/ext/164361-staff-software-engineer-copilot-experiences) at **GitHub**
- [Senior Engineer, Infrastructure Platform](https://www.wearedevelopers.com/jobs/ext/328836-senior-engineer-infrastructure-platform) at **Intercom, Inc.**
- [Senior Software Engineer, Client Apps Platform](https://www.wearedevelopers.com/jobs/ext/1773893-senior-software-engineer-client-apps-platform) at **GitHub**
- [Cloud Foundations Team](https://www.wearedevelopers.com/jobs/ext/1483289-cloud-foundations-team) at **GitHub**