A blended IRR across a book of platform deals landing well under the advertised numbers is worth investigating at the platform level
Take a book of 14 equity deals across three crowdfunding platforms, with target IRRs ranging 11 to 16 percent, where the realized and marked blend comes in around 8. Nothing has gone to zero, but distributions get paused, hold periods stretch from three years to five, and capital sits longer than projected. The useful reframe is underwriting the platform layer rather than the deal layer. If every sponsor's projection is assumed to be optimistic by roughly the same amount, the real differentiator among platforms isn't the average projection, it's the dispersion around it. Fee load is visible in the offering documents and easy to compare. What's harder to see upfront is how a platform handles a sponsor who stops reporting, or what real recourse exists when a capital call goes out. The honest answer is that this is difficult to test before committing capital. The closest proxies are a platform's track record of sponsor terminations or replacements, how transparent its reporting cadence stays when a deal underperforms, and whether it discloses realized versus marked returns separately rather than blending them. None of those substitute fully for time, but they narrow the guesswork more than fee comparisons alone.