We spent years treating dividends like a religion. Every holding had to throw off cash. The unofficial commandment was two thousand dollars a year from each name. It sounded disciplined. It was a way to feel busy.

It also meant we kept sitting in companies because they paid us, not because the business was worth more next year. The portfolio grew coupons. The pile itself did not grow enough.

So we broke the commandment. Roughly seventy percent still lives in dividend payers. Thirty percent is allowed to be growth — names that reinvest instead of mailing us a cheque. Amazon and Tesla were the first obvious ones. Others will come and go.

The mix is not a tattoo. If a dividend stock becomes a bad business we sell it. If a growth name reaches a price that makes us look like tourists, we take the gift. Seventy–thirty is a bias, not a prison.

What changed in practice is simple. We stopped needing every position to fund the next grocery run. Cash flow still matters. It just stopped being the only scoreboard. The year-by-year value chart is ugly in the middle and less ugly after the shift. That is the only chart that counts on this page.

I am not handing you a model portfolio. I am describing the one we actually run, with the usual warning: this is not advice.

If you want the messy version as it happened:

Our Portfolio Strategy Continues to Evolve

April Results: Better Than Expected, Worse Than My Pride Can Handle

How we buy US names at all sits on a different page: How a Swede Buys the US Market

This page changes when the mix changes. The dated posts stay dated. That is the point.