Diversification and position sizing for infrastructure

The most important investing decision isn't what to buy — it's how much. Position sizing is what lets you be wrong without being ruined.

The core-satellite picture

A sane portfolio has a boring core and interesting satellites. The core — most of your money — sits in broad, liquid, diversified assets. The satellites are concentrated positions with higher potential and higher risk: individual stocks, crypto, and yes, infrastructure participations like data centers.

Infrastructure belongs in the satellite ring for one dominant reason: illiquidity. Your capital is committed for the project term. It can't be your emergency fund, your house deposit, or money with a date on it.

The 10% rule of thumb

EU crowdfunding regulation suggests retail investors keep crowd-investments below 10% of their net worth — and it's a genuinely good heuristic, not just red tape. At 10%, even a total loss of one position is a bad year, not a changed life. We'd rather you invest €50 comfortably than €5,000 anxiously; anxious money makes bad decisions.

Diversify inside the satellite too

If you do allocate to infrastructure: spread across facilities rather than concentrating in one, prefer platforms that hold each project in its own structure (so one project's failure doesn't drag down the rest), and stagger your entries over time instead of committing everything the day you discover the asset class.

The uncomfortable honest bit

Diversification protects you from single failures — it does not protect you from an entire sector cooling off. If AI compute demand fell broadly, well-diversified data-center investors would still feel it. That residual, undiversifiable risk is precisely why the asset class pays a return at all.

Educational content — not investment advice. Capital at risk.