Observability & Metrics

Overview

Observability is the practice of understanding why a brewing process behaves the way it does.

Traditional brewing logs often focus on recording events and measurements. While these records are valuable, they typically answer only one question:

What happened?

Observability seeks to answer a deeper question:

Why did it happen?

BevOps defines Observability as the ability to understand the state and behavior of a brewing process through measurements, observations, and historical data.

The goal is not simply to collect information.

The goal is to generate actionable insights.

⸻

Monitoring vs Observability

These terms are often used interchangeably, but they serve different purposes.

Monitoring

Monitoring answers:

What happened?

Examples:

  • Original Gravity was 1.052
  • Final Gravity was 1.011
  • Fermentation lasted 8 days
  • Temperature averaged 67°F

Monitoring focuses on known metrics and known conditions.

⸻

Observability

Observability answers:

Why did it happen?

Examples:

  • Why did attenuation increase?
  • Why did the beer finish drier than expected?
  • Why was clarity improved compared to previous batches?
  • Why did hop aroma fade more quickly than normal?

Observability focuses on understanding relationships between measurements, processes, and outcomes.

⸻

The Three Pillars of Brewing Observability

BevOps adapts a concept commonly used in software operations known as the Three Pillars of Observability.

Metrics

Quantitative measurements collected throughout the brewing process.

Examples:

  • Original Gravity
  • Final Gravity
  • Attenuation
  • Temperature
  • Carbonation
  • Yield

Metrics provide objective visibility into brewing performance.

⸻

Events

Significant occurrences during the brewing process.

Examples:

  • Yeast pitched
  • Dry hops added
  • Cold crash started
  • Fermentation completed
  • Beer packaged

Events provide context for metrics.

Without events, measurements often lack meaning.

⸻

Logs

Human-recorded observations and notes.

Examples:

  • Strong krausen observed on Day 2
  • Citrus aroma increased after dry hopping
  • Fermentation activity appeared sluggish
  • Packaging losses higher than expected

Logs capture information that cannot always be measured automatically.

⸻

Observability Data Sources

Brewers may collect data from a variety of sources.

Examples include:

Manual Measurements

  • Hydrometer readings
  • Refractometer readings
  • Volume measurements
  • Packaging counts

Environmental Data

  • Fermentation temperature
  • Ambient temperature
  • Humidity

Telemetry Systems

  • Tilt Hydrometer
  • Tilt Pico
  • Fermentation controllers
  • Temperature probes

Human Feedback

  • Tasting notes
  • Batch reviews
  • Beer Satisfaction Scores
  • Community feedback

Every source contributes to overall observability.

⸻

Brewing Telemetry

Telemetry is the automated collection of brewing data.

Examples include:

Tilt Hydrometer ↓ Tilt Pico ↓ Data Pipeline ↓ Time-Series Database ↓ Dashboard

Telemetry allows brewers to observe trends that may otherwise go unnoticed.

Examples:

  • Fermentation curves
  • Temperature stability
  • Attenuation progression
  • Fermentation duration

The objective is not automation for its own sake.

The objective is visibility.

⸻

Observability Workflows

Observability becomes most valuable when data is used to answer meaningful questions.

Example:

Monitoring

OG = 1.052 FG = 1.011

Useful, but incomplete.

⸻

Observability

Mash Temperature = 150°F Attenuation = 79% Compared to: Mash Temperature = 152°F Attenuation = 74%

Observation: Lower mash temperatures increased attenuation.

Insight: Mash temperature influences perceived dryness.

Action: Adjust future mash schedules accordingly.

This is where data becomes knowledge.

Dashboards

Dashboards provide visual access to brewing metrics and trends. Potential dashboard metrics include:

  • Gravity
  • Temperature
  • Attenuation
  • Fermentation Progress
  • Yield
  • Batch History Examples:
  • Grafana
  • Google Sheets
  • Brewfather
  • Custom Dashboards A dashboard should provide clarity, not complexity.

Alerting

Observability should not only help brewers understand the past. It should help identify issues in the present.

Examples:

  • Fermentation temperature outside target range
  • Unexpected gravity readings
  • Telemetry failures
  • Data collection interruptions The goal is early detection and faster recovery.

Observability Maturity

BevOps recognizes that brewers operate at different levels of sophistication.

Level 1 — Manual

  • Brew logs
  • Handwritten notes
  • Hydrometer readings

Level 2 — Digital

  • Spreadsheets
  • Digital logs
  • Basic charts

Level 3 — Telemetry

  • Automated gravity collection
  • Temperature monitoring
  • Continuous data collection

Level 4 — Observable Brewery

  • Time-series databases
  • Dashboards
  • Alerting
  • Historical analysis

Level 5 — Data-Driven Brewing

  • Continuous improvement informed by operational metrics
  • Repeatable experimentation
  • Integrated observability practices

Every level provides value. The goal is progress, not perfection

The Goal of Observability

The purpose of observability is not to create more dashboards. The purpose of observability is understanding.

Metrics become information.

Information becomes knowledge.

Knowledge becomes improvement.

A brewer who understands why a beer behaves the way it does is better equipped to improve future batches.

In BevOps, observability transforms brewing from a collection of isolated events into a continuously improving system.

Measure → Observe → Learn → Improve