Platform

Everything you need for causal analytics.

From raw CSV to executive insights — automated, validated, explained. No data science PhD required.

Start here

Your dashboard tells you what happened. It cannot tell you what caused it.

Reporting tools are very good at showing which numbers moved together. They cannot separate the thing you did from everything else going on at the same time — the season, the brand campaign, the fact that your best customers were always going to renew. That gap is where budget quietly disappears.

Marketing

Your dashboard says

Paid search converted 12,000 users last month.

CausoAI says

Paid search caused 3,400 of those conversions. The other 8,600 would have converted anyway — they were already looking for you.

Sales

Your dashboard says

Deals with a discount close 18% more often.

CausoAI says

Discounting above 15% does not improve close rate at all. Below that it adds 6 points. Everything above 15% is margin given away.

Customer Success

Your dashboard says

Customers who use Feature X churn 40% less.

CausoAI says

Feature X does not cause retention — engaged customers were always going to stay. The onboarding call does: it cuts 90-day churn by 12 points.

What using it actually looks like

You will never build a statistical model, pick a method or write a formula. There is one step that genuinely needs you: checking the causal diagram CausoAI proposes before it measures anything.

No algorithm can work out every cause-and-effect direction from data alone — the data will happily tell you that ice cream sales cause sunshine. Deciding which way an arrow points is a judgement about how your business works, not a statistical one, which is exactly why you are the right person to make it. In practice it is a short review of something already drawn for you.

01

Bring your data

A CSV export, or a direct connection to your database or Google Sheet. Customer records, campaign spend, deal history — whatever you already track.

02

Check the diagram

CausoAI proposes how everything connects. You review it and fix anything that is wrong — this is the one step that genuinely needs a person, and it is the step where your knowledge of the business does the work.

Your input needed

03

Ask a business question

Something you would actually argue about in a meeting. Does discounting close deals? Does onboarding prevent churn? Is this channel worth the budget?

04

Get an answer you can defend

A number, an honest confidence range, who it applies to, and a plain-English explanation of what it means and how far to trust it.

Feature 01

It drafts the diagram of what drives what. You just check it.

Before you can measure the effect of anything, you need to know how the pieces of your business connect: what feeds what, and what is quietly influencing both. Getting that wrong is why most analyses give the wrong answer.

CausoAI reads your data and drafts that map for you. You review it and correct anything that does not match how your business actually works — no algorithm can settle every cause-and-effect direction on its own, and that call is about your business rather than about statistics. No modelling and no formulas: just a check of something already drawn for you.

  • Drafted automatically from the data you already have
  • Drag-and-drop editor — your team’s knowledge of the business wins
  • Spots the hidden factors that make two things look related when they are not
  • Every version is saved, so you can compare and roll back

What a causal diagram looks like

  1. Email frequency

    something you control

  2. Site visits

    what it moves first

  3. Trial signups

    the step in between

  4. Revenue

    what you actually care about

And the part people miss

Seasonality pushes up email engagement and revenue at the same time. Miss it, and you credit the campaign for money the season was always going to bring in. CausoAI holds factors like this constant before it measures anything.

Feature 02

5-Layer Causal Readiness Score

Not all causal analyses are equally trustworthy. CRS quantifies exactly how confident you should be — and tells you what gaps need to be fixed.

L1

Data Coverage

25%

Sample size, missingness, column completeness — do you have enough high-quality data?

L2

Identifiability

25%

Temporal ordering, backdoor criterion, collider detection — is the causal effect computable?

L3

Estimation Feasibility

20%

Does your data meet what the method needs to give a trustworthy answer?

L4

Statistical Power

20%

Propensity score overlap, effective sample sizes — is the sample large enough to detect effects?

L5

Robustness

10%

Sensitivity to unmeasured confounders — how stable are results across methods?

Score Interpretation

80–100

High confidence

Suitable for decision-making

60–79

Directionally reliable

Inform hypotheses

40–59

Use with caution

Significant assumptions

< 40

Do not act

Insufficient for action

Feature 03

A straight answer to “did it actually work?”

You never choose a statistical method. CausoAI looks at what you are measuring and how much data you have, picks the right approach on its own, and gives you the answer in four parts.

The size of the effect

+8.3pp conversion rate

A number you can put in a plan, not a direction. Moving email frequency from 2x to 4x a week caused conversion to rise 8.3 percentage points.

How sure we are

95% confident: 6.1 – 10.5pp

Every estimate comes with an honest range. If the range crosses zero, we say so plainly instead of dressing up a result that is not there.

Who it worked on

Strongest: inactive 15–30 days

The average hides the opportunity. CausoAI breaks the effect down by segment so you know which customers to spend the next euro on.

What it is worth

≈ €340K revenue impact

Effects are translated back into the units you report on — revenue, margin, retained accounts — so the analysis lands in a business conversation.

For the analysts in the room: the method is chosen automatically from the treatment type and sample size, and CausoAI always names the method it used and the assumptions it relied on — so your data team can check the work rather than take it on trust.

Feature 04

What-If Counterfactual Simulator

Once the causal model is built, you can run forward simulations: change the value of a treatment variable and see the predicted outcome — with confidence intervals.

  • Continuous slider for treatment intensity (e.g., ad spend ±50%)
  • 95% confidence intervals on all predictions
  • Dose-response curves for continuous treatments
  • Segment-level counterfactuals for targeted decisions

Counterfactual Simulation

+20%

Predicted Outcome

+$340Krevenue lift
95% CI: $280K — $410K
$280K$340K$410K
AI Insightsclaude-sonnet-4-6
"Increasing email campaign frequency from 2x to 4x per week causally increased 30-day conversion rate by 8.3 percentage points (95% CI: 6.1–10.5 pp). The effect is driven primarily by the re-engagement segment — users inactive for 15–30 days show the strongest response (+12.7 pp)."

Feature 05

AI-Powered Insights via Claude

CausoAI integrates with Claude (claude-sonnet-4-6) to translate statistical outputs into plain-language executive summaries, assumption explanations, and actionable recommendations — cached for 7 days to control API costs.

  • Executive summaries explaining causal effects in plain English
  • Assumption & limitation explanations for non-technical stakeholders
  • 3+ concrete, prioritized action recommendations
  • Natural language Q&A — ask follow-up questions about your analysis

Feature 07

Connect your data warehouse directly.

Skip the CSV export. Connect CausoAI to PostgreSQL, MySQL, or Google Sheets, write a query, preview the results, and import — all without leaving the platform.

  • Preview 100 rows before committing to an import
  • Read-only query validation — no accidental writes
  • Imported data is auto-profiled and ready for analysis
  • Credentials stored encrypted at rest

PostgreSQL

Database

Connect to any PostgreSQL instance. Write a SELECT query, preview 100 rows, and import directly — no CSV export needed.

MySQL

Database

Connect to MySQL or MariaDB. CausoAI validates your query is read-only before execution.

Google Sheets

Spreadsheet

Authenticate via service account and pull a sheet directly into your analysis workspace.

Feature 08

The questions that normally need a specialist.

These are the analyses that usually mean hiring someone or waiting a quarter for the data team to get to your ticket. CausoAI runs them for you, checks them against your data, and explains the answer in plain English.

Why it worked, not just that it worked

The route an effect takes

Your campaign lifted revenue. Was that because it brought more people to the site, or because it reached better-quality buyers who spent more? CausoAI splits the result into the routes it travelled, so you can tell which part of the machine actually did the work — and put money into that part rather than the whole thing.

Measuring something you rolled out

Before and after

You launched a change in one region, for one segment, in one month. CausoAI compares what happened against a comparable group that did not get it, so a strong quarter or a seasonal bump does not get counted as your result. It also checks the two groups were behaving alike beforehand, and warns you when they were not.

Who you should actually target

Segments that respond

Almost nothing works equally well on everyone. CausoAI finds the groups where the effect is strongest and where it is nil, then turns that into a plain targeting rule you can hand to whoever runs the campaign — along with an honest view of how much you gain by targeting instead of blanketing.

Why the number moved

Root cause

Churn jumped last month and nobody can say why. Instead of a list of everything that moved at the same time, CausoAI traces the change back through your causal diagram and ranks what actually drove it — so the post-mortem starts from causes rather than coincidences.

Why you can trust the answer

Rigorous underneath. Readable on top.

Causal analysis is easy to get wrong, and a wrong answer is worse than no answer because it looks just as convincing. CausoAI is built so you can act on a result without having to audit the statistics yourself.

Established science, not our own invention

The methods underneath CausoAI are the peer-reviewed standards used by research teams and large tech companies to measure real-world impact. We automated them — we did not make them up.

It shows its working

Every result names the approach it used and the assumptions it depended on. Nothing arrives as a number you are simply asked to believe.

It tells you when not to act

A readiness score runs on every analysis. When your data cannot support a confident answer, CausoAI says so and explains what is missing, rather than producing a confident-looking chart anyway.

Your data team can audit it

Estimates, confidence ranges, assumptions and the causal diagram itself all export. If someone technical wants to check the result before you act on it, they can.