143 algorithms · Live

One question.
One API call.
One answer you can defend.

Data science has always required a team, a methodology, and a week of waiting. Not anymore. Send your data and a question in plain English — get back structured analysis and a clear recommendation in seconds.

A stateless HTTP API with 135 marketing science algorithms, callable from any agent framework, script, pipeline, or tool.

Algorithm Library 135 total
A/B Testing & Experimentation7
Marketing Science Core12
Attribution & Multi-Touch6
Audience & Propensity7
Campaign Analytics10
Next Best Action6
Media Mix Modeling6
Multi-Armed Bandit6
Data Cleaning21
Data Wrangling15
Feature Engineering22
Exploratory Data Analysis9
Monte Carlo & Simulation6
135 algorithms live
Seconds not days
JSON + NL output
Zero data stored
One API call
How it works

Four steps from question to answer

01
Discovery
Semantic search across 135 algorithms finds the best candidates for your query — no routing configuration required.
02
Selection
AI selects the statistically correct algorithm and extracts the right parameters from your natural-language input.
03
Execution
The algorithm runs server-side against validated inputs. Pure computation — no LLM involved in the math.
04
Interpretation
Results come back as structured JSON with metrics, the algorithm used, and a plain-language verdict your system can act on immediately.
Algorithm library

Every method your team reaches for

Testing
A/B Testing & Experimentation
7 algorithms
Core
Marketing Science Core
12 algorithms
Attribution
Attribution & Multi-Touch
6 algorithms
Audience
Audience & Propensity Modeling
7 algorithms
Campaign
Campaign Analytics & Optimization
10 algorithms
Decision
Next Best Action & Offer
6 algorithms
Media
Media Mix Modeling
6 algorithms
Bandit
Multi-Armed Bandit & Online Learning
6 algorithms
+67 data infrastructure algorithms
Data Cleaning · Data Wrangling · Feature Engineering · EDA · Monte Carlo
Browse all 135 →
Why teams choose Your Data Stories

Built for decisions, not dashboards

📋
Paste any data format
Numbers, spreadsheet paste, or CSV snippets — the API understands context and extracts what it needs. No rigid schema, no data wrangling required on your end.
🧭
Automatic algorithm routing
Describe your question in plain English. The gateway selects the statistically correct method and runs it every time — Bayesian, frequentist, or non-parametric.
💬
Answers your team acts on
Every result includes a plain-language recommendation alongside the statistics. Designed for analysts and marketers, not only data scientists.
Works in any stack
Call it from a Python script, a Claude or GPT agent, a no-code workflow, or a cron job. The engine does not care — it returns the same structured JSON to every caller.
🔒
Stateless by design
Every request is fully independent. No sessions, no data persistence, no compliance surface. Your data is used to run the algorithm and immediately discarded.
🔬
Real math, not confidence
The algorithms run as validated scientific code — not LLM inference. Confidence intervals, p-values, and posteriors computed exactly, every time.
Built for builders

Stop explaining.
Start shipping.

Hiring a data scientist to answer statistical questions is a solution from another era. Your Data Stories gives any product, agent, or workflow the ability to reason correctly with numbers — without the overhead, the methodology debates, or the waiting.

When your orchestration layer hits a decision that needs proof, not just confidence, one API call returns the analysis, the algorithm used, and a verdict your system can act on immediately.

Returns structured JSON your agent can act on
Works as an MCP tool in any Claude or GPT agent
Natural language verdict alongside raw statistics
No sessions, no state, no infrastructure to manage
Get API access →
JavaScript
const res = await fetch('https://api.yourdatastories.com/analyze', {
  method: 'POST',
  headers: {
    'Authorization': `Bearer ${apiKey}`,
    'Content-Type': 'application/json',
  },
  body: JSON.stringify({
    message: 'Control 1200/8400, Variant 1380/8400.\nRun Bayesian AB test.'
  })
})

const { algorithm_used, metrics, verdict } = await res.json()
// → { algorithm_used: "bayesian_ab_testing",
//     metrics: [...], verdict: "Ship the variant." }
Pricing

Simple, transparent pricing

Starter
$49
per month

500 API calls / mo
Full algorithm library
JSON + NL output
Email support
Get started
Scale
$299
per month

Unlimited API calls
Full algorithm library
Custom integrations
Dedicated support
Contact us

See it in action

Try the live demo — paste your data, ask your question, get a verdict in seconds.

Try the demo → Get API access