Rehearse decisions before the market sees them
The simulation and decision engine for commercial decisions.
Jonbar keeps a simulated market of how shoppers react to prices, campaigns and the world around them. Your sales history calibrates it to your products. Compare versions of a decision side by side, or set a goal with your limits and let it find the plan.
- Calibrated with your sales
- No customer data needed
- 2-week pilot
Run · Wireless headphones, Q4
Sample scenario · sample data| World | Value | 80% range |
|---|---|---|
| A | 2,087,395 | baseline |
| B (Winner) | 2,200,918 | 2,133,428 – 2,256,772 |
| C | 2,071,071 | 2,019,176 – 2,125,434 |
How it works
Three steps. No real customer sees a test.
Calibrate
Share 12+ weeks of sales, prices and campaigns. Jonbar fine-tunes its market model to your products and price levels.
Rehearse
Play versions of a decision side by side, or set a goal and limits and let the engine search the plans for you.
Decide
See which plan leaves the most gross profit, why it wins, and the probable range around it.
Why rehearse
Most promotions are decided on instinct. The numbers say that is expensive.
Outside research on promotions, from field experiments and large industry datasets.
- ~3/4of promotions don't break even.Nielsen, 39M promotional events, via FoodNavigator
- −100% → 700%+return on promotions, from the worst event to the best.Strategy& (PwC)
- ~1/3of a promotion's extra units are won from competing brands. The rest is shoppers buying earlier or more of the category.van Heerde, Gupta & Wittink, JMR
- −20.3%future revenue from existing customers after deep discounts, in a field experiment.Anderson & Simester, Marketing Science
Use cases
What are you deciding?
One engine, several kinds of decisions. Pick yours and see what Jonbar can do for it today.
I want to
Goal to decision
“Grow gross profit next month without going under ₺499 or above 20% off.”
What you get
- The three best plans that respect your limits
- The trade-off each one makes
- Estimates corrected for the winner's curse
Where this is going
A market that keeps living between your decisions.
We are building a living market: a population of simulated shoppers with their own budgets, habits and pay days, a behaviour model trained on real purchase sequences, and world signals such as inflation, the campaign calendar and competitor shelves. Every forecast is locked before the outcome, so each pilot makes the market sharper for the next one.
- Today: engine v0, classic demand models calibrated with your sales
- Next: behaviour model, live world signals, shopper population
Real data, not guesses
We don't ask an AI what your customers think.
Language models are a weak stand-in for real shoppers. Jonbar learns from what people actually bought.
How the market model is built- 40.8/100best language models' score at simulating human behaviour.SimBench, ICLR 2026
- r = 0.20average match between people and their AI copies.Peng et al., 2025
Start
One decision. Two weeks. One clear answer.
In the pilot we first test the engine blind on one of your past campaigns, then rehearse your next decision. Fixed price: $999.
- Blind test on a past campaign
- Rehearsal of your next decision
- A decision memo you keep
Join the waitlist
What is Jonbar, in one sentence?
A simulation and decision engine for commercial decisions. Jonbar keeps a simulated market of how shoppers react to prices, campaigns and the world around them, calibrates it to your products with your sales history, and shows which plan leaves the most gross profit before you run it.
Where does the market come from? Do you make up data?
No invented customers, no invented sales. Jonbar keeps its own market model: how shoppers in a category respond to price, discounts, timing and competitors, built from published research and the data we accumulate. Your sales history fine-tunes that model to your products and your price levels. Where your history is too thin to pin an effect down, the run falls back to published ranges and says so.
Isn't this just asking an AI what shoppers think?
No. Language models are a weak stand-in for real shoppers: the best ones score 40.8 out of 100 at simulating human behaviour (SimBench). Jonbar's engine learns from what people actually bought. A language model may help explain a result in plain words later; it never makes the decision.
What data do I need?
Product × day (or week) sales for the last 12 weeks or more: units, net price, list price, campaign flag and unit cost. A CSV export is enough. We never need customer names, addresses or order numbers.
What does the pilot cost?
The pilot has a fixed price; we share it when we invite you from the waitlist.