Market sizing

A market sizing in Aqmen is a value driver tree (the same framework top consulting firms use to structure markets), not a spreadsheet. That difference is what makes the model interrogable: every number has a place in the structure, and the structure can change without breaking the math.

The building blocks

• Dimensions split the market into segments: geography, customer type, product category, channel. A tree can combine several dimensions, say, geography × customer type.
• Segments are the leaves those dimensions create: DACH/Enterprise, France/Mid-market, and so on.
• Variables are the drivers that value each segment: number of customers, penetration, average price, consumption per user. Market value is computed from the drivers, never typed in.
• Values are time series. Historical years are asserted with sources; forecast years are driven by hypotheses.

Building the tree

Start from the market definition the deal team cares about, then add dimensions one at a time. The tree restructures as you go: split a segment into finer ones, swap a dimension for a more useful cut, or merge segments back together. Because the math is computed from the structure, a re-cut is a restructure, not a rebuild.

Populating values

Historical values carry assertions: the source the number came from and the judgement applied on the way in. When a value is challenged in a partner review, the answer is attached to the cell.

Forecasting

Forecasts are not typed-in growth rates. They're driven by quantitative hypotheses: named, per-period growth impacts that attach to drivers and accumulate. See Hypotheses & scenarios.

The agent way

Connected over MCP, an agent can build the tree from a brief ("size the EU industrial-software market bottom-up, geography × end-market"), research and populate historicals with citations, and propose hypotheses for the forecast. All of it lands in the analysis for your review.