Forecasts and dashboards
Numbers computed by code, with an honest range instead of a single figure, and dashboards your team actually opens.
§01 How it works
- A range, not a point
- Monte Carlo simulation drawn from real historical spread, not fixed percentages, so the answer shows how wide the uncertainty is.
- Per-person data
- Model each performer separately and sum to the team, instead of working from team aggregates. Incomplete periods are flagged and imputed with labelled assumptions.
- Reproducible
- A fixed seed and a data fingerprint per run; back-tests on history truncated to a past date, so you can check the method against a known outcome.
- Dashboards
- P&L, trainer and marketing reports built from the systems that already hold the data, so the people who need them open them.
§02 Related work
- Team forecasting systemDeterministic Monte Carlo, no LLM on the numbers
- CRM for a network of martial-arts clubsWhatsApp funnels, reports, trainer pay
- LLM integration into working systemsWire a model into the POS, booking, CRM, messengers
- Process and quality auditA written list of fixes ordered by cost
§03 Format
Remote, from Sofia (UTC+2/+3), as a B2B contractor. $85/hour for short engagements up to 20 hours; longer engagements by arrangement. The scope is agreed after a short call. I work with clients in the US, Canada and Europe.
My working hours overlap with both Europe and the US.
§04 Short answers
- Why a range instead of one number?
- A Monte Carlo simulation drawn from real historical spread shows how wide the uncertainty is, so a decision does not rest on a false single figure.
- Can I check the method?
- Yes. Every run uses a fixed seed and a data fingerprint, and can be back-tested on history truncated to a past date.
- How do we start?
- A short call to agree the scope. $85/hour for short engagements up to 20 hours; longer engagements by arrangement.
Have a task like this? Write to me.
Send me a message
It goes straight to my Telegram. I reply to the address you leave.