> 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/impact-statistics.md).

# Impact Statistics

Analyze user and data impact charts with overall metrics for prioritization.

Impact statistics answer the critical question: **Who and what is affected by this error?** While occurrence data tells you how often an Issue happens, impact metrics tell you the scope and severity of each occurrence.

## Why Impact Matters

Frequency alone can't tell you how much an error actually matters to the business.

{% hint style="danger" %}
**Frequency Isn't Everything**

An error that happens once per day but affects 100 sales reps processing opportunities is far more critical than an error that happens 50 times per day but only impacts a single admin's test workflow.
{% endhint %}

Two Issues can have identical occurrence patterns but vastly different business impact:

| Metric               | Issue A              | Issue B                 |
| -------------------- | -------------------- | ----------------------- |
| Average Frequency    | 1 day                | 1 day                   |
| Daily Volume         | 10                   | 10                      |
| **Users Impacted**   | **1 (system admin)** | **50 (sales reps)**     |
| **Records Impacted** | **10 (test data)**   | **500 (opportunities)** |
| **Priority**         | **Low**              | **Highest**             |

Impact metrics enable data-driven prioritization based on business consequences, not just technical frequency.

## User Impact Metrics

### Unique Users Impacted

The **deduplicated count** of users who have triggered this error.

**What it tells you:**

* Is this affecting one person or the whole team?
* Is it user-specific (permissions, data access) or systemic?
* What's the blast radius?

**Interpreting the metric:**

| Users Impacted | Interpretation                                | Priority Implication |
| -------------- | --------------------------------------------- | -------------------- |
| 1              | Single user (may be permissions or edge case) | Low-Medium           |
| 5-10           | Small team or specific role                   | Medium               |
| 20-50          | Department-wide issue                         | High                 |
| 100+           | Organization-wide problem                     | Highest              |

**User Impact Context:**

The metric considers:

* **Running User ID** from each log (deduplicated)
* Integration users (e.g., "API User") often indicate backend processes
* System admin users may indicate scheduled jobs or automation

### User Profile Breakdown

A visual distribution showing **which user profiles or roles** are affected.

**Displayed as:**

* **Donut chart:** Each slice represents a profile (color-coded)
* **List view:** Profile names with counts
* **Hover details:** Exact number of users per profile

**What it tells you:**

* Is this role-specific? (e.g., only Sales Reps)
* Permission-related? (e.g., only external Community users)
* Affects internal vs. external users?

**Example scenario:**

```
User Impact by Profile:
- Sales Rep: 35 users (70%)
- Sales Manager: 10 users (20%)
- System Administrator: 5 users (10%)

Interpretation: Primarily affects sales team, likely a workflow 
or object they commonly interact with (Opportunity, Lead, Account).
```

**Use cases:**

* Click on a profile slice to filter related logs to only that profile
* Download CSV of affected users for notification or remediation
* Identify if problem is permission-set-related

### User Impact Timeline

A **bar chart** showing user impact over time (default: last 30 days).

**What it tells you:**

* Is user impact growing or shrinking?
* Did a recent deployment affect more users?
* Are new users hitting the error, or same users repeatedly?

**Pattern examples:**

| Pattern       | Appearance        | Interpretation                     |
| ------------- | ----------------- | ---------------------------------- |
| **Flat**      | Consistent height | Same user base affected            |
| **Growing**   | Bars increasing   | Spreading to more users            |
| **Spike**     | One tall bar      | One-time event, many users         |
| **Declining** | Bars decreasing   | Workaround in place or partial fix |

## Data Impact Metrics

### Unique Records Impacted

The **deduplicated count** of records (Salesforce object records) involved in this error.

**What it tells you:**

* How much data is affected?
* Are specific records consistently problematic?
* Is this a data quality issue?

**Interpreting the metric:**

| Records Impacted | Interpretation                       | Concern Level                   |
| ---------------- | ------------------------------------ | ------------------------------- |
| 1-5              | Specific records (may be data issue) | Low (fix data)                  |
| 10-100           | Moderate dataset                     | Medium (check data patterns)    |
| 500+             | Large dataset or bulk operation      | High (systemic issue)           |
| 10,000+          | Mass process or integration          | Highest (business process down) |

**Record Impact Context:**

The metric tracks:

* **Record IDs** referenced in logs (from log's Related Objects)
* Deduplicated across all logs under this Issue
* Can include multiple object types

### Object Breakdown

A visual distribution showing **which Salesforce objects** are involved.

**Displayed as:**

* **Donut chart:** Each slice is an object (Account, Opportunity, Case, etc.)
* **List view:** Object API names with counts
* **Hover details:** Exact record count per object

**What it tells you:**

* Which data is at risk?
* Business process affected (Accounts = customer data, Opportunities = revenue)
* Object-specific logic issue (validation rule, trigger, workflow)

**Example scenario:**

```
Data Impact by Object:
- Opportunity: 450 records (75%)
- Account: 100 records (17%)
- OpportunityLineItem: 50 records (8%)

Interpretation: Opportunity-centric error, likely affects 
sales pipeline. High business impact if revenue data.
```

**Use cases:**

* Click on an object slice to filter logs to only records of that type
* Download CSV of impacted record IDs for manual remediation
* Identify if error is object-specific or cross-object

### Data Impact Timeline

A **bar chart** showing record impact over time (default: last 30 days).

**What it tells you:**

* Is the volume of affected records growing?
* Did a data migration trigger this?
* Are the same records repeatedly failing?

## Downloading Impact Data

Both impact tabs let you export the underlying data as CSV for offline analysis and remediation.

{% hint style="info" %}
**Export for Action**

Use CSV downloads to take action outside of Pharos—notify affected users, remediate failed records with Data Loader, or create detailed reports for stakeholders.
{% endhint %}

Both User and Data impact tabs provide **CSV download** capabilities for offline analysis and remediation.

### Downloading User Data

From the **User Impact** tab:

1. Navigate to **All Time** subtab (or select date range)
2. Click **Download CSV**
3. File contains:
   * User ID
   * User Name
   * Profile
   * Number of times user hit this error
   * First and Last occurrence for that user

**Use cases:**

* Send notification email to affected users
* Investigate permission differences
* Identify power users who may have workarounds

### Downloading Record Data

From the **Data Impact** tab:

1. Navigate to **All Time** subtab (or select date range)
2. Click **Download CSV**
3. File contains:
   * Record ID
   * Object Type
   * Number of times record was involved
   * First and Last occurrence for that record

**Use cases:**

* Manually remediate failed records (Data Loader, manual edits)
* Identify data quality patterns (e.g., all records missing Industry field)
* Export for reporting or stakeholder communication

## Visual Representations

### Donut Charts (Profile/Object Distribution)

Donut charts show how profiles or objects are distributed across the Issue.

**Features:**

* **Hover:** See exact counts and percentages
* **Click:** Filter related logs to selected segment
* **Color-coded:** Each profile/object has unique color
* **Legend:** Shows all segments with counts

**Best for:** Understanding distribution at a glance

### Bar Charts (Timeline)

Bar charts plot user or record counts over time.

**Features:**

* **X-axis:** Time (days, weeks, or months depending on zoom)
* **Y-axis:** Count of users or records
* **Color-coded:** By profile or object type (stacked bars)
* **Hover:** See exact date and count

**Best for:** Identifying trends and change over time

### Table Views

Table views support detailed, sortable analysis and specific lookups.

**Features:**

* **Sortable columns:** Click to sort by count, name, etc.
* **Search/filter:** Find specific user or object
* **Pagination:** Handle large datasets
* **Drill-down links:** Click user/record to see their specific logs

**Best for:** Detailed analysis and specific lookups

## Combining Occurrence and Impact Data

The most effective prioritization uses **both** occurrence and impact metrics together.

### Prioritization Matrix

The matrix below pairs occurrence and impact levels to suggest a priority.

| Occurrence         | Impact          | Priority        | Example                                |
| ------------------ | --------------- | --------------- | -------------------------------------- |
| **High frequency** | **High impact** | **Highest**     | Daily error, 50 users, 500 records     |
| **High frequency** | **Low impact**  | **Medium-High** | Daily error, 1 user, 10 records        |
| **Low frequency**  | **High impact** | **Medium-High** | Monthly error, 100 users, 1000 records |
| **Low frequency**  | **Low impact**  | **Low**         | Monthly error, 1 user, 5 records       |

### Real-World Example

Here is how the two dimensions combine on a real Issue:

```
Issue: DML Exception - Account Trigger

Occurrence Metrics:
- Average Frequency: 2 days
- Daily Volume: 8
- Last Occurred: Today

Impact Metrics:
- Users Impacted: 23 (Sales Reps)
- Records Impacted: 120 (Accounts)
- Object Breakdown: 100% Account

Priority Decision: HIGH
Reasoning:
- Frequent (every 2 days, 8x when it happens)
- Affects core sales team (23 reps)
- Impacts customer master data (Accounts)
- Currently active (occurred today)
```

## Impact Metric Gotchas

A few situations can skew impact metrics, so investigate the context behind the numbers before deciding priority.

{% hint style="warning" %}
**Common Misinterpretations**

Be mindful of these scenarios that can skew impact metrics. Always investigate the context behind the numbers before making priority decisions.
{% endhint %}

### Integration Users Inflate Counts

If "API Integration User" shows as highly impacted:

* This often represents backend processes, not humans
* Check if it's a scheduled job or external system
* Impact may be lower than it appears (no human users blocked)

### Same Users Hit Error Repeatedly

High user count doesn't always mean wide impact:

* Check if same 5 users hit error 100 times each vs. 100 users once
* Use **User Impact Timeline** to see if it's growing user base
* Differentiate between persistent problem for few vs. occasional for many

### Test/Sandbox Records

If record IDs look like test data:

* Verify environment (production vs. sandbox)
* Check if users are system admins testing
* May be lower priority if not affecting real business data

## Using Impact Data for Triage

During your triage workflow, impact metrics guide priority assignment:

**Questions to ask:**

1. How many users are blocked by this?
2. What's their role? (Customer-facing? Revenue-generating?)
3. Which data is affected? (Mission-critical objects?)
4. Is the impact growing or stable?
5. Can users work around this, or are they fully blocked?

**Triage decision tree:**

```mermaid
graph TD
    A[Impacted Users Count > 20?]
    A -->|YES| B[Priority: High or Highest]
    A -->|NO| C[Records Impacted > 100?]
    C -->|YES| D[Priority: Medium or High]
    C -->|NO| E[Affects customer-facing users?]
    E -->|YES| F[Priority: Medium]
    E -->|NO| G[Priority: Low]
```

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

* [Occurrence Info](https://docs.pharos.ai/platform-overview/pharos-platform-architecture/issue-tracker/issue-details/occurrence-info.md) – Understand WHEN and HOW OFTEN errors occur
* [Debug Views](https://docs.pharos.ai/platform-overview/pharos-platform-architecture/issue-tracker/issue-details/debug-views.md) – Investigate WHY errors are happening
* [Issue Details Overview](https://docs.pharos.ai/platform-overview/pharos-platform-architecture/issue-tracker/issue-details.md) – See full Issue details page structure
  {% endhint %}
