Talk2Data InsightGenius
Analyze survey data (.sav, .csv, .xlsx) through Claude — crosstabs with significance testing, ANOVA, correlation, gap analysis, and publication-ready Excel exports. Upload once, analyze unlimited.
README
SPSS InsightGenius API
Professional SPSS processing API + MCP server for market research. Upload .sav files, get crosstabs with significance testing, auto-detected question types, and publication-ready Excel exports. Includes AI-powered zero-config analysis.
Live: spss.insightgenius.io | API Docs: spss.insightgenius.io/docs | MCP: spss.insightgenius.io/mcp/sse
Quick Start
Option 1: Web UI (no code needed)
- Open spss.insightgenius.io
- Drag & drop your
.savfile - Click Auto-Analyze for instant results, or configure manually:
- Select banner variables (demographics for columns)
- Choose stubs (questions for rows)
- Enable Top 2 Box / Means
- Click Generate Excel → download your tabulation
Option 2: Auto-Analyze (zero config)
curl -X POST https://spss.insightgenius.io/v1/auto-analyze \
-H "Authorization: Bearer YOUR_API_KEY" \
-F "file=@survey.sav" \
-o auto_analysis.xlsx
AI auto-detects banners, groups variables into MRS/Grid, applies nets, and generates a complete Excel.
Option 3: Full Control
curl -X POST https://spss.insightgenius.io/v1/tabulate \
-H "Authorization: Bearer YOUR_API_KEY" \
-F "file=@survey.sav" \
-F 'spec={
"banners": ["gender", "region", "age_group"],
"stubs": ["_all_"],
"significance_level": 0.95,
"include_means": true,
"nets": {"sat_overall": {"Top 2 Box": [4,5], "Bottom 2 Box": [1,2]}},
"mrs_groups": {"Brand_Awareness": ["AWARE_A","AWARE_B","AWARE_C"]}
}' -o tabulation.xlsx
Option 4: Python
import requests, json
resp = requests.post(
"https://spss.insightgenius.io/v1/tabulate",
headers={"Authorization": "Bearer YOUR_API_KEY"},
files={"file": open("survey.sav", "rb")},
data={"spec": json.dumps({
"banners": ["gender", "region"],
"stubs": ["_all_"],
"include_means": True,
"significance_level": 0.95,
})}
)
with open("tabulation.xlsx", "wb") as f:
f.write(resp.content)
print(f"Done: {resp.headers['X-Stubs-Success']} tables generated")
Option 5: MCP (for AI agents)
Connect to https://spss.insightgenius.io/mcp/sse and use any of the 12 tools. Files are passed as base64.
Endpoints (14)
| Method | Endpoint | Description |
|---|---|---|
| POST | /v1/auto-analyze |
Zero-config — upload .sav, get complete Excel (AI-detected banners, MRS, grids, nets) |
| POST | /v1/tabulate |
Full tabulation → Excel with sig letters, nets, means, MRS, grids, custom groups. Accepts optional .docx Reporting Ticket. |
| POST | /v1/metadata |
Variable metadata + suggested banners + detected groups + preset nets |
| POST | /v1/frequency |
Frequency table (counts, %, mean, std, median) |
| POST | /v1/crosstab |
Single crosstab with sig letters (A/B/C) + chi-square p-value |
| POST | /v1/correlation |
Correlation matrix (Pearson/Spearman/Kendall) with p-values |
| POST | /v1/anova |
One-way ANOVA with Tukey HSD post-hoc comparisons |
| POST | /v1/gap-analysis |
Importance-Performance gap analysis with quadrants |
| POST | /v1/satisfaction-summary |
Compact T2B/B2B/Mean for multiple scale variables |
| POST | /v1/process |
Multi-operation pipeline (auto-detect or manual) |
| POST | /v1/convert |
Convert .sav → xlsx, csv, parquet, dta |
| POST | /v1/parse-ticket |
Parse Reporting Ticket .docx → tab plan (Haiku AI) |
| GET | /v1/health |
Health check + engine status |
| GET | /v1/usage |
Usage stats for your API key |
MCP Tools (12)
| Tool | Description |
|---|---|
get_spss_metadata |
Variable metadata + auto-detect |
get_variable_info |
Single variable detail |
analyze_frequencies |
Frequency table |
analyze_crosstabs |
Crosstab with sig letters |
analyze_correlation |
Correlation matrix |
analyze_anova |
ANOVA + Tukey HSD |
analyze_gap |
Gap analysis with quadrants |
summarize_satisfaction |
Satisfaction summary |
create_tabulation |
Full tabulation → Excel (base64) |
auto_analyze |
Zero-config → Excel (base64) |
export_data |
Format conversion |
list_files |
Available tools |
Tabulate Spec
| Field | Type | Default | Description |
|---|---|---|---|
banners |
string[] | required | Demographics for columns (e.g., ["gender", "region"]) |
stubs |
string[] | ["_all_"] |
Questions for rows (_all_ = auto-select all) |
significance_level |
float | 0.95 |
0.90, 0.95, or 0.99 |
weight |
string | null |
Weight variable name |
include_means |
bool | false |
Add Mean row with T-test sig letters |
include_total_column |
bool | true |
Total as first column |
output_mode |
string | "multi_sheet" |
"multi_sheet" or "single_sheet" |
nets |
object | null |
Per-variable net definitions |
mrs_groups |
object | null |
MRS groups: {"name": ["var1", "var2"]} |
grid_groups |
object | null |
Grid groups: {"name": {"variables": [...], "show": ["t2b","mean"]}} |
custom_groups |
array | null |
Custom breaks with AND conditions |
title |
string | "" |
Report title |
Excel Output
- Summary sheet: file info, column legend (A=London, B=South East...), stub index
- One sheet per stub: headers → letters → base (N) → data with
pct% SIG_LETTERS→ nets → means - MRS sheets: one per group, percentages can exceed 100%
- Grid sheets: compact T2B/B2B/Mean summary
- Significance letters in red, nets in green rows
- Freeze panes for scrolling
Significance Testing
Column proportion z-test with letter notation (A/B/C):
- Each banner category gets a letter (e.g., Male=A, Female=B, London=C, North=D)
- Each cell tested vs every other column
- Significantly higher → other column's letter appears (e.g.,
68.6% Emeans sig higher than column E) - Supports weighted (Kish effective-n) and unweighted
- Means tested with independent T-test
- Confidence levels: 90%, 95%, 99%
Authentication
All endpoints require: Authorization: Bearer sk_live_... or sk_test_...
Rate Limits
| Plan | Requests/min | Max file | Price |
|---|---|---|---|
| Free | 10 | 5 MB | $0 |
| Growth | 60 | 50 MB | $29/mo |
| Business | 200 | 200 MB | $99/mo |
| Enterprise | Unlimited | 500 MB | Custom |
Error Codes
| Code | HTTP | Meaning |
|---|---|---|
UNAUTHORIZED |
401 | Missing/invalid API key |
FORBIDDEN |
403 | Valid key, wrong scope |
RATE_LIMIT_EXCEEDED |
429 | Too many requests |
INVALID_FILE_FORMAT |
400 | Not a .sav file |
FILE_TOO_LARGE |
413 | Exceeds plan limit |
VARIABLE_NOT_FOUND |
400 | Variable doesn't exist |
PROCESSING_FAILED |
500 | Engine error |
PROCESSING_TIMEOUT |
504 | Exceeded time limit |
Local Development
git clone https://github.com/quack2025/spss-insightgenius-api.git
cd spss-insightgenius-api
pip install -r requirements.txt
cp .env.example .env # Edit with your API key hash
python main.py # → http://localhost:8000
python -m pytest tests/ -v # 68 tests
Stack
| Layer | Technology |
|---|---|
| API | FastAPI + Pydantic v2 |
| Engine | QuantipyMRX (crosstab, sig testing, auto-detect, MRS, NPS) |
| AI | Claude Haiku (ticket parsing, smart labels, executive summary) |
| Auth | API keys (SHA256, no DB) |
| Rate Limiting | Redis (fallback: in-memory) |
| MCP | FastMCP with SSE transport |
| Deploy | Railway (Docker, Gunicorn, 4 replicas, auto-deploy) |
Built by Genius Labs.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.