Forecast new product launches, even with no sales history

Yes. That is a core use case. Simporter Forecasting is designed for new launches where no direct sales history exists. It combines product master data, historical sales from related products, consumer demand signals, comparable launches, and machine learning models to estimate demand for new products before launch.

Simporter Forecasting combines multiple data sources:

• Product master data provided by the client
• Historical sales data, including sell-in, sell-out, category sales, seasonality, and holidays
• Consumer demand signals from Simporter Insights
• Concept and attribute inputs from Simporter Concepts
• Comparable product and launch patterns identified by Simporter models

All of this data is processed and enriched by Simporter to support forecasting.

Sales data is essential for training the forecasting models and understanding how launches perform in the real market. Even when the new product has no history, historical sales from existing products help the models learn launch patterns, seasonality, distribution effects, pricing behavior, and category dynamics.

Sell-out data adds market visibility beyond your own internal shipments. It helps capture category dynamics and competitor product performance, which is especially valuable when evaluating similar launches, understanding demand patterns, and strengthening scenario assumptions.

No. Users should not have to figure out the “correct” proxy on their own. Colton does the heavy lifting: it analyzes the available data, identifies relevant comparables, and creates forecasting scenarios based on the user’s business requirements. Users can still review and adjust assumptions, but they do not need to manually perform the underlying analysis.

Colton can work across the full Simporter workflow. It has access to:

• data from Simporter Insights
• data from Simporter Concepts
• product master data
• historical sales data
• scenario context and user inputs

This allows Colton to create, compare, and explain scenarios using both internal business data and external consumer demand signals.

Fully transparent. Every scenario shows the assumptions behind the forecast and where each value came from. Users can see which inputs were provided directly, which were inferred from data, and how different factors influence the final output. This makes forecasts easier to trust, explain, and defend internally.

Each scenario represents a different launch assumption set. For example, scenarios may vary by distribution, price, pack size, marketing support, timing, or cannibalization. This allows teams to compare possible launch outcomes side by side and understand what drives the differences.

Forecasting is part of a connected workflow.

• Insights provides consumer demand signals such as attribute momentum, relative importance, sentiment, and opportunity size.
• Concepts can provide new product ideas or structured concept inputs.
• Forecasting turns those inputs into launch scenarios.

This means teams can move from consumer insight to concept to forecast in one connected system.

Simporter Forecasting helps teams:

• improve ad hoc forecasting accuracy, typically by at least 10% on average, and often more depending on the client’s current baseline
• speed up forecasting by hundreds of times
• understand scenario assumptions much more clearly
• create scenarios for products with new attributes in the category
• compare hundreds of scenarios efficiently
• make faster and better launch decisions