Simpli Insights

Insights helps support teams understand what customers are really asking about. Cluster cases into themes, detect emerging topics, suggest category taxonomies, and analyse distribution patterns — all powered by LLM analysis.

Simpli Insights provides theme discovery, trend detection, and category analysis for support case data.

Configuration

Variable	Default	Description
`APP_PORT`	`8012`	Server port
`LITELLM_MODEL`	`openai/gpt-5-mini`	LLM model for analysis
`MAX_CLUSTERS`	`20`	Maximum number of theme clusters
`MIN_CLUSTER_SIZE`	`3`	Minimum cases per cluster
`CORS_ORIGINS`	`*`	Allowed CORS origins (comma-separated)

Start the server

simpli-insights serve

API endpoints

All endpoints are under the /api/v1 prefix.

POST `/api/v1/themes`

Discover themes from a batch of support cases.

Request:

{
  "cases": [
    {"id": "c-1", "subject": "Can't log in", "content": "Password reset not working"},
    {"id": "c-2", "subject": "Login broken", "content": "Getting 403 on sign-in page"},
    {"id": "c-3", "subject": "Account locked", "content": "Too many failed attempts"}
  ]
}

Response:

{
  "audit_id": "i-abc123",
  "total_cases": 3,
  "themes": [
    {
      "theme_id": "t-1",
      "name": "Authentication Issues",
      "description": "Login, password, and account access problems",
      "case_count": 3,
      "case_ids": ["c-1", "c-2", "c-3"],
      "sample_subjects": ["Can't log in", "Login broken"]
    }
  ],
  "uncategorized_case_ids": []
}

POST `/api/v1/emerging`

Detect emerging topics by comparing recent cases against a baseline.

{
  "recent_cases": [
    {"id": "c-10", "subject": "SSO not working", "content": "SAML error after update"}
  ],
  "baseline_cases": [
    {"id": "c-1", "subject": "Password reset", "content": "Standard password issue"}
  ]
}

POST `/api/v1/categories`

Suggest a category taxonomy from case data, or validate existing categories.

POST `/api/v1/distribution`

Analyse how cases distribute across existing categories.

GET `/health`

Health check.

Integration example

import httpx

client = httpx.Client(base_url="http://localhost:8012")

# Discover themes
themes = client.post("/api/v1/themes", json={
    "cases": [
        {"id": "c-1", "subject": "Billing question", "content": "Charged twice"},
        {"id": "c-2", "subject": "Double charge", "content": "Two charges on card"},
        {"id": "c-3", "subject": "Wrong amount", "content": "Billed $99 instead of $49"},
    ],
}).json()

for theme in themes["themes"]:
    print(f"{theme['name']}: {theme['case_count']} cases")

# Detect emerging topics
emerging = client.post("/api/v1/emerging", json={
    "recent_cases": recent_cases,
    "baseline_cases": last_month_cases,
}).json()

for topic in emerging["topics"]:
    print(f"NEW: {topic['topic']} ({topic['case_count']} cases, +{topic['growth_rate']:.0%})")

Next steps

Executive Reporting — Feed Insights data into executive dashboards
Integration Overview — Connect Insights to your ticketing system
Model Selection — Choose the right model for clustering accuracy

On this page