A market-research synthesis combines two sources: a 2024 industry report citing 12% YoY growth, and a 2025 academic study citing 8% YoY growth. The current synthesis says: "Growth is between 8% and 12%."
What's missing?
Why did you pick that answer? Two or three sentences. The act of articulating it is what builds the judgment — not the click that follows.
A range without provenance is no better than picking one value arbitrarily — it just hides which source said what. The 2024 report and the 2025 study likely describe different time periods (last year ≠ last year), used different market definitions, or applied different methodologies. A defensible synthesis preserves attribution: "A 2024 industry report estimated 12% YoY growth for FY2023, while a 2025 academic study estimated 8% YoY growth for FY2024 using a narrower definition of 'the market.' The values are not directly comparable." The reader can now evaluate the discrepancy themselves.
A bare range collapses methodologically distinct values into a single statistic, which is exactly what good synthesis is supposed to avoid. The reader can't tell whether 8%–12% reflects true uncertainty about a single quantity or two different quantities being conflated.
A confidence percentage on a range you constructed by collapsing two methodologically different sources is a false precision. The honest move is to surface the sources and let the reader weigh them — not to invent a confidence number on a synthetic range.
Weighting the more recent source higher assumes recency is the dominant quality signal — but the 2025 study's methodology might be narrower or less reliable than the 2024 report's. Weighting based on recency alone hides the methodological question rather than answering it.