Find the decisions that grow revenue and profit.

Kapnova uses specialized agents to find opportunities across pricing, promotions, marketing, inventory and demand, then quantitative models determine the best decision.

AI finds the opportunities. Math computes the answer.

Built on published MIT research Design partners: consumer brands from $50M to $200M+

Kapnova doesn’t wait for you to ask the right question.

It is built around one decision universe: revenue, gross profit and contribution margin. Kapnova continuously looks for decisions that can materially move those outcomes and ignores the ones that don't.

Your campaign generated $1.4M.
One line caused none of it.

Paid media, creators, email and a promotion all ran at the same time and revenue increased. Traditional reporting credits each one. Kapnova separates what each lever actually caused from what would have happened anyway.

CAMPAIGN DECOMPOSITION / two weeks / daily SKU panel / illustrative
identification: quasi-experimental
What the campaign returned reported $1.4M
Actually caused
$1.4M
across three levers, not four
Not incremental
$0
display, funded all quarter
If reallocated
~3x
per dollar, at the measured rate
Next step
Holdout
two geos, three weeks, no extra spend
What each lever caused / net of the counterfactual
Promo and discount
+$520K
Creator seeding, mediated
+$470K
Paid, direct
+$410K
Display
$0
Reported lift
$1.4M
quasi-experimental identification4 sourcescreator path measured as mediated, not directfigures illustrative

Based on reported attribution, you would keep funding display. Kapnova shows the opposite: creators generated $470K through organic conversation that last-touch never credited, while display looked stronger than it really was. See the other decisions we run

Agents find the opportunities. Models determine the value. Outcomes make the system smarter.

Agents find

Specialized agents scan your business and outside market data to surface the decisions with the most revenue or profit at stake.

Models answer

Causal inference, econometrics, forecasting, simulation and optimization determine what each decision is actually worth.

Outcomes compound

Kapnova recommends experiments, measures what happened and uses the results to improve the next decision.

We recommend. You decide. Nothing is executed on your behalf, and your raw data never leaves your tenant.

Why our number is different.

Most reporting optimizes toward what correlated hardest. Kapnova separates what actually caused the outcome from what simply moved alongside it, revealing where the real revenue and profit opportunity is.

PricePromoCreatorSentimentRevenue
What the reporting shows

Everything moved together, so the budget went to whatever correlated hardest.

PricePromoCreatorSentimentRevenue
What actually moved revenue

Price moves revenue through sentiment. Promo caused none of the lift, so it leaves the model.

Move the lever. Watch the answer move with it.

Causal simulation / price change 10,000 Monte Carlo paths
Price change +8.0%
$0 0% +15% +$2M −$3M

This is the simulator your team works in. Set the change you are considering and the model returns the effect, the interval around it, and the recommendation. On a decision this size, the gap between the two reads can be worth millions.

On your data, we test this first against a decision you’ve already made, so you can validate the result yourself.

Figures illustrative. See this run on your own decisions

What the trend line says
+$1.2M
Raise price. The curve says go.
What Kapnova projects
−$0.4M
80% CI  [−$0.9M, +$0.1M]
The increase sours sentiment, feeding back to demand through a second path the curve never models.
Hold

What your team actually touches.

Four surfaces. Two for when you have a decision in front of you, two for the weeks in between.

Every decision and every outcome makes the next one sharper.

When you have a decision
Decision Diagnostic
The expected outcome, confidence range, drivers and risks.
Scenario Simulator
Change the price, budget or constraint. See what happens.
In between decisions
Continuous Intelligence
What changed, what drove it and what needs your attention.
Decision Ledger
What we predicted, what you decided and what actually happened.
What powers it
Causal Data Foundation
Your data, structured to understand what actually drives outcomes.
Outside-In Signals
Sentiment, reviews, competitors and market signals, continuously brought in.

The science behind consequential decisions, applied to business.

Kapnova’s causal engine grew out of MIT research studying whether treatments actually change health outcomes. We brought that same rigor to business to understand what truly drives an outcome, then model what happens when you change it.

MIT / doctoral thesis
Causal Inference with Survival Outcomes via Orthogonal Statistical Learning
Examined at MIT by a committee including a leading econometrician and epidemiology faculty.
Read the research

Start with your URL. Go deeper with your data.

Start with your URL and Kapnova will surface what it can find from public data alone. Then connect your own data to measure what past decisions actually caused and where the next revenue or profit opportunity may be.