Forecasting & optimization

Forecast outcomes, compare plans, and choose with confidence

Project revenue, demand, or signups with the model that backtests best on your history. Compare scenarios, see 65% and 95% intervals, and choose the plan that hits your target without ignoring risk.

No code and no model wrangling. The platform fits, backtests, selects, and stores the assumptions behind every forecast.

Backtested model choice

Compete candidate models on held-out history and use the best performer.

65% and 95% intervals

Plan with a likely range and a conservative range, not just one line.

Scenario optimization

Compare plans and choose the one that hits the target within acceptable risk.

Assumptions stored

Version forecasts, model choices, and planning assumptions for future review.

Revenue forecast: 65% and 95% intervals

Actuals Forecast 65% CI 95% CI
Today95%65%HistoryProjection

Every forecast comes with 65% and 95% confidence intervals — so you see the likely range and exactly how sure the model is.

What you get

Forecasting built for decisions, not just charts

The best model for your data, honest uncertainty on every number, and scenario optimization to turn a forecast into a plan.

The best model for your data

We fit a competition of time-series models, backtest them on your own history, and automatically pick the one that forecasts your business most accurately.

65% and 95% confidence

Every forecast ships with both a 65% and a 95% confidence interval, so you see the most-likely range and exactly how sure the model is.

Scenario optimization

Forecast under different plans for spend, pricing, or headcount and find the scenario that best hits your target, not just one fixed projection.

Backtested accuracy

See how each model performed on held-out history before you trust it, and track forecast error over time as real actuals come in.

Marketing-aware

Layer in incrementality tests and MMM outputs so forecasts reflect causal evidence, not just last year's trend repeated forward.

One central library

Every forecast, model choice, and assumption is stored and versioned, ready to compare against what actually happened.

How it works

From history to a plan you can defend

Four steps, no spreadsheets, no model-wrangling. You stay in control of the assumptions and the call.

1

Bring your history

Add historical outcomes such as revenue, demand, or signups at daily or weekly grain. Connect a file or a source, no code required.

2

Auto-fit & pick the best model

We fit a competition of time-series models, backtest each on your data, and select the most accurate one for your business.

3

Forecast with 65/95 CIs

Get a forward forecast with both a 65% and a 95% confidence interval, with the likely range clearly shown instead of a single guess.

4

Optimize the plan

Compare scenarios, see the forecasted outcome of each, and choose the plan that best hits your target.

Optimization

Forecast every plan. Pick the one that wins.

Don't forecast a single future. Compare scenarios side by side, each with its own confidence range, and choose the plan that best hits your target.

Projected outcome by plan

Plan A
Plan B
Plan C

Each plan is forecast with its own 65% and 95% interval, so you optimize for the target without ignoring the risk.

The best model, chosen for you

No single model is best for every business

So we don't force one. We fit several, backtest them on your own history, and let the most accurate model win, then show you why.

Human in the loop

Built for everyone on the team

You don't need to be a statistician. The hard math runs under the hood; you point, click, and make the call with the full picture in front of you.

MarketerAnalystCMOData ScientistVP of MarketingDirectorGrowth ManagerBrand ManagerAgencyFounder
The measurement triangle

Forecasts get sharper when they learn from your tests

Business forecasting sits on top of the measurement triangle: incrementality testing and your MMM tell the forecast what marketing really drives, so the plan you project forward is grounded in causal evidence, not just last year's curve.

Inform. Attribution gives the MMM a fast, directional read on what's working between tests.

Calibrate. Incrementality tests anchor the MMM to causal ground truth, not just correlation.

Validate. Holdout tests check that attribution and the model agree with what really happened.

See the full framework
FAQ

Business forecasting & optimization FAQ

Experiment OS

Measurement Ops Make the Product Repeatable

The models and tests matter, but the workflow around them matters too. Shako Stats is designed to become the operating system around experiment planning, metadata, documentation, and cross-test learning.

Data Management

Centralize datasets, mappings, and historical records so experiments and models always start from the same source of truth.

Tags and Parameters

Organize tests by audience, creative strategy, bidding logic, or business objective so learnings remain searchable and reusable.

Gantt Timeline

See what tests are planned, in-flight, or completed so overlapping interventions and measurement conflicts are easier to manage.

Documentation

Turn methodology, definitions, and experiment design guidance into an internal operating system instead of leaving them scattered across decks.

New: AI Marketing Scientist

Or just ask your AI.

Connect any AI agent through MCP/API to read-only forecasting context: scenarios, assumptions, selected models, intervals, backtests, and actuals. Shako Stats runs the statistical forecast and stores the assumptions; the AI helps your team understand which plan is most defensible.

Meet your AI Marketing Scientist

One-click OAuth for Claude & ChatGPT, read-only, anyone on the team can ask

Bring rigorous measurement to your team

Design tests, understand results, and connect your team to one statistics-backed measurement workspace.

Paid B2B plans built for brands & agencies.