Here is the shortcut, and it looks so reasonable that sophisticated people take it in front of investors: three assets, three IRRs, weight each by its equity, average them, call it the portfolio return. The arithmetic is clean. It is also wrong, not approximately, not conservatively, just wrong, and the reason it is wrong is worth genuinely understanding, because the same misunderstanding quietly corrupts other corners of underwriting too.
IRR is not a quantity. It is a rate implied by a specific sequence of cash flows on specific dates, and rates from different sequences do not combine by averaging, because the average throws away the one thing IRR is made of: timing. A twenty percent IRR earned over two years and a twenty percent IRR earned over seven years are the same number describing utterly different amounts and durations of money at work, and weighting them by equity does nothing to repair that, since equity measures how much went in, not when anything came back. The only correct portfolio IRR is the one computed from the merged schedule: every capital call, every distribution, every refinance and disposition, from every asset, on its actual date, combined into one timeline, and solved as a single stream.
The gap between the shortcut and the truth is not academic. It widens exactly when portfolios get interesting: staggered closings, so the assets' clocks start on different dates; an early sale returning capital in year two while the rest runs seven; a refinance pulling money out of one asset mid-hold. In those structures the averaged number can miss the real one by several points in either direction, and each direction has a victim. Overstate and the investor deck contains a number the distributions will never reconcile to, which is a credibility event with a delay on it. Understate and the deal loses capital it deserved to a spreadsheet error in its own favor.
The widget below stages the failure so you can watch it. Enter a few assets, equity, profit, hold length, and it computes each asset's IRR, the equity-weighted average, and the true IRR of the combined cash flows, side by side. Stagger the holds and watch the two portfolio numbers separate. The demonstration takes thirty seconds and permanently changes how you read the returns page of other people's decks, which is worth the thirty seconds by itself.
The same principle, cash flows combine, metrics do not, is why Portfolio Deal Underwriting in PropCalc™ computes blended returns from the full merged monthly schedule across every asset in the deal, distributions firing as each asset pays, capital called as deployed. The equity multiple survives averaging, incidentally, since dollars are dollars whenever they arrive, and presenting the pair honestly computed is the standard the room should expect. When a deck shows a portfolio IRR, the first diligence question is now yours: computed from the combined schedule, or averaged?