> For the complete documentation index, see [llms.txt](https://docs.pharos.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.pharos.ai/platform-overview/pharos-platform-architecture/issue-tracker/issue-details/occurrence-info.md).

# Occurrence Info

View occurrence charts and occurrence time statistics presented on the issues details page.

Occurrence information tells you **when** and **how often** an Issue occurs. This data is critical for prioritization—a daily error demands more attention than a one-time anomaly.

## Time-Based Metrics

Pharos summarizes an Issue's timing with a handful of metrics that together tell you how urgent it is.

{% hint style="info" %}
**Key to Prioritization**

Occurrence metrics answer the fundamental question: "How urgent is this?" Understanding these time-based patterns helps you distinguish between critical daily failures and rare edge cases.
{% endhint %}

The Issue record displays several key time and frequency metrics:

### First Occurred

The timestamp when this Issue was first created—when Pharos detected this unique error pattern for the first time.

**What it tells you:**

* How long this problem has existed
* Whether it's a new issue or legacy technical debt
* Correlation with deployments or configuration changes

**Example scenarios:**

| First Occurred       | Interpretation                                        |
| -------------------- | ----------------------------------------------------- |
| 2 hours ago          | New problem, likely from recent deployment            |
| 6 months ago         | Long-standing issue, possibly accepted technical debt |
| Yesterday at 3:00 AM | Might correlate with scheduled job or batch process   |

### Last Occurred

The most recent timestamp when a log was created for this Issue.

**What it tells you:**

* Is the problem still active?
* Did a recent fix work?
* How fresh is this Issue?

**Active vs. Dormant:**

* **Last Occurred: 5 minutes ago** → Active, currently happening
* **Last Occurred: 30 days ago** → Dormant, may have resolved itself or users found workaround

### Log Count

The total number of log records linked to this Issue—how many times this error has occurred.

**What it tells you:**

* Scale of the problem
* Whether it's persistent or sporadic
* Historical severity

**Interpreting Log Count:**

| Log Count | Frequency + Duration         | Meaning                      |
| --------- | ---------------------------- | ---------------------------- |
| 3         | First Occurred: Yesterday    | New, low occurrence          |
| 500       | First Occurred: Last week    | High frequency or many users |
| 10,000    | First Occurred: 6 months ago | Chronic, long-standing issue |

{% hint style="warning" %}
**Log count alone can be misleading.** An Issue with 10,000 logs over 2 years (14 per day) is less urgent than one with 100 logs in the last hour.
{% endhint %}

### Average Frequency

The average gap in days between the days this error occurred, measured through today—how often this error typically happens.

**Formula:** Average gap in days between occurrence days, measured through today (rounded to two decimals)

Because gaps are measured between distinct occurrence **days**, this value is never below 1. Multiple-per-day intensity is Daily Volume's job.

**What it tells you:**

* Consistency of error pattern
* Whether it's daily, weekly, or sporadic

**Reading the metric:**

| Average Frequency | Interpretation      | Priority Implication          |
| ----------------- | ------------------- | ----------------------------- |
| 1.0 days          | Daily occurrence    | High (persistent)             |
| 7.0 days          | Weekly pattern      | Medium (may be scheduled job) |
| 30 days           | Monthly or sporadic | Low (unless high impact)      |

**Example:**

```
First Occurred: 30 days ago
Occurrence pattern: One occurrence every 3 days, through today
Average Frequency: 3.0 days

Interpretation: This error happens approximately every 3 days.
```

### Daily Volume

The average number of times this error occurs **per day when it does happen**.

**Formula:** `Total logs ÷ number of days with at least one occurrence`

This metric is different from Average Frequency:

* **Average Frequency** = How often does it happen? (days between)
* **Daily Volume** = When it happens, how many times? (occurrences per day)

**What it tells you:**

* Burst pattern vs. steady state
* User impact multiplier
* System stress

**Reading the metric:**

| Daily Volume | Pattern                   | Example Scenario                        |
| ------------ | ------------------------- | --------------------------------------- |
| 1.0          | Single occurrence per day | Scheduled job failing once daily        |
| 5.0          | Multiple times per day    | User workflow failing occasionally      |
| 50+          | High volume               | Common validation error or bulk process |

**Example:**

```
Average Frequency: 3 days (happens every 3 days)
Daily Volume: 12
Log Count: 120

Interpretation: Every 3 days, this error occurs about 12 times 
(not spread throughout the day, but clustered when it does happen).
```

{% hint style="info" %}
**How These Metrics Are Calculated**

Average Frequency and Daily Volume are recomputed periodically by a scheduled job, and both include logs from Duplicate issues rolled into the parent Issue.
{% endhint %}

## Occurrence Chart Visualization

The **Occurrence Chart** on the Issue page provides a visual timeline of when errors occurred.

### Chart Features

The chart opens on a default timeframe, color-codes days by volume, and offers several zoom levels.

**Default View:** 30 days\
Shows a bar chart with one bar per day, color-coded by occurrence volume.

**Color Coding:**

* **Red bars:** High occurrence days (above 75th percentile)
* **Orange bars:** Moderate occurrence (50-75th percentile)
* **Yellow bars:** Low occurrence (25-50th percentile)
* **Gray bars:** No occurrences

**Zoom Levels:**

* 7 days (last week)
* 30 days (last month)
* 60 days (last two months)
* 6 months (long-term trend)

### Pattern Recognition

The chart reveals important patterns at a glance:

#### Daily Consistent Pattern

```
█ █ █ █ █ █ █ █ █ █ █ █ █ █ 
Mon Tue Wed Thu Fri Sat Sun Mon Tue Wed Thu Fri Sat Sun
```

**What it means:** Persistent issue affecting normal operations\
**Likely cause:** Common user workflow, scheduled job, or trigger on frequent operation\
**Priority:** High (consistent user/system impact)

#### Weekday-Only Pattern

```
█ █ █ █ █ _ _ █ █ █ █ █ _ _
Mon Tue Wed Thu Fri Sat Sun Mon Tue Wed Thu Fri Sat Sun
```

**What it means:** User-driven during business hours\
**Likely cause:** Manual process, screen flow, or business-hours scheduled job\
**Priority:** Medium-High (affects active users)

#### Single Spike

```
_ _ _ █ _ _ _ _ _ _ _ _ _ _
```

**What it means:** One-time event\
**Likely cause:** Data load, deployment issue, or edge case\
**Priority:** Low (unless spike is very large)

#### Increasing Trend

```
▁ ▂ ▃ ▄ ▅ ▆ ▇ █ █ █
```

**What it means:** Problem getting worse over time\
**Likely cause:** Data volume growth, worsening performance, spreading bug\
**Priority:** Highest (trend indicates worsening)

#### Decreasing Trend

```
█ █ █ ▇ ▆ ▅ ▄ ▃ ▂ ▁
```

**What it means:** Problem resolving or workaround in place\
**Likely cause:** Partial fix deployed, users avoiding the action, or issue self-correcting\
**Priority:** Low-Medium (may resolve naturally)

#### Sporadic Pattern

```
█ _ _ _ █ _ █ _ _ _ _ █ _ _
```

**What it means:** Intermittent, hard to predict\
**Likely cause:** Edge case, environmental condition, specific data pattern\
**Priority:** Medium (may need tracing to diagnose)

## Using Occurrence Data for Prioritization

Combine these metrics to make informed priority decisions:

### High Priority Indicators

Escalate an Issue when several of these signals line up:

* **Average Frequency = 1 day** → Happening every day (pair with Daily Volume for intensity)
* **Daily Volume > 10** → High occurrence when it happens
* **Last Occurred within last hour** → Currently active
* **Increasing trend in chart** → Getting worse
* **Consistent pattern (no gaps)** → Persistent, not intermittent

### Low Priority Indicators

Deprioritize when the signals point the other way:

* **Average Frequency > 30 days** → Rare occurrence
* **Daily Volume = 1** → Single occurrence per day
* **Last Occurred > 30 days ago** → May have resolved
* **Decreasing trend in chart** → Problem fading
* **Single spike pattern** → One-time event

### Example Triage Decision

Here is how those indicators combine into a single priority call:

```
Issue: Flow Failed - Account_Creation_Flow

Occurrence Metrics:
- First Occurred: 45 days ago
- Last Occurred: 10 minutes ago
- Log Count: 1,270
- Average Frequency: 1.41 days
- Daily Volume: 39.7

Chart shows: Consistent daily pattern with slight increase last week

Decision: HIGH PRIORITY
Reasoning: Daily occurrence (1.41 day frequency), currently active, 
persistent over time, trending worse. Likely impacts user workflow.
```

## Correlation with Deployments

Occurrence patterns are most powerful when you line them up against your deployment history.

{% hint style="info" %}
**Connect Errors to Changes**

Always correlate occurrence patterns with your deployment history. Issues that start immediately after a release are usually easier to diagnose and fix than long-standing problems.
{% endhint %}

When reviewing occurrence data, consider recent changes:

**Questions to ask:**

* Does **First Occurred** align with a deployment?
* Did **Daily Volume** spike after a release?
* Does the **chart** show a clear before/after pattern?

**Tip:** Keep a deployment log or reference your CI/CD history when investigating new Issues.

{% hint style="success" %}
**Next Steps:**

* [Impact Statistics](https://docs.pharos.ai/platform-overview/pharos-platform-architecture/issue-tracker/issue-details/impact-statistics.md) – Understand WHO and WHAT is affected
* [Issue Details Overview](https://docs.pharos.ai/platform-overview/pharos-platform-architecture/issue-tracker/issue-details.md) – See how occurrence data fits into the full Issue view
  {% endhint %}
