Some findings are robust. Venture investors stage their funding and write contracts that shift control with performance, as Gompers and Kaplan and Strömberg documented; terms such as liquidation preferences divide outcomes very differently from headline ownership; and venture returns are highly skewed, with a few investments producing most of the money.
The returns to venture capital as an asset class are less settled. Measured performance depends on the period, the benchmark and the data source; strong 1990s vintages lift long-run averages, and many funds have not beaten public markets after fees. Data are reported by funds, and failed funds and companies drop out, so survivorship bias pushes estimates upwards.
Whether venture capital causes success or selects it is the deepest debate. Venture-backed companies grow faster and professionalize sooner, but investors choose promising companies, and studies that separate sorting from influence, such as Sørensen's, find that selection explains much of the difference. Gompers, Gornall, Kaplan and Strebulaev (2020) found that venture capitalists themselves rank the founding team as the most important factor in their decisions, which is consistent with both views.
For founders, the interpretive question is fit. The power law pushes funded companies towards maximum growth, which suits a few and harms many that could have been good, smaller businesses. The Northvolt and Databricks applications show two ends of the spectrum: capital that could not overcome an operational problem, and capital raised from strength to provide liquidity. Founder stories in the press are a biased sample of the successes, so they are poor guides to the typical outcome of taking venture money.