A validated new business idea still needs somewhere to go inside the organization once it's proven — a receiving business unit, or in the most extreme case an entirely new P&L, the way Amazon Web Services was given its own. Without a designated "landing pad," a good idea that has cleared the Evidence Spectrum for Validating Business Ideas bar goes organizationally homeless: no existing unit wants to absorb the risk or the resourcing, so the idea either gets stripped for parts and folded piecemeal into existing products, or is killed outright despite the evidence supporting it.
This is a distinct failure point in Explore/Exploit Dual Portfolio Management — evidence discipline in the Explore Portfolio can be done well and the idea can still die at the exploit/explore handoff if no receiving structure exists.
The landing pad works in tandem with a growth board: the board decides an idea has cleared the evidence bar to scale, and the landing pad is where it's absorbed — an existing business unit, or a new P&L stood up for it, as with AWS inside Amazon. Without a landing pad already designated, a validated idea can clear every evidence gate and still die homeless because no one owns integrating it into the core business.
Osterwalder's core scaling-failure claim: companies fail to scale validated ideas not just by picking the wrong ideas too early, but equally often by being structurally unable to give a validated idea a home — "homeless ideas remain homeless." His example: Amazon Web Services required an entirely new P&L rather than being fit into an existing business line, because no existing line could have been its landing pad.
Related failure mode: formally handing a validated idea from a "testing team" to a separate "executing team" doesn't work well in his experience — see Testing-to-Execution Handoff Failure (No Clean Handover). A landing pad has to be a real organizational home (new P&L, new business unit) rather than a scheduled handover between teams.