Forty-Eight Hours in Hong Kong, Planned Backwards
Two days is not enough for Hong Kong. It is, however, what a lot of people get, so it is worth planning properly rather than pretending otherwise.
Start from the end
The usual method is to write down everything you want to see, group it roughly by area, and hope the days fall into shape. In a city with a functioning grid that mostly works. In Hong Kong it produces an itinerary that looks efficient on a map and is miserable to walk, because the map is flat and the city is not.
So I did it the other way round. I decided what I wanted the last hour before the airport to be, which was the Star Ferry crossing at dusk with nothing left to carry. Everything else got scheduled backwards from that fixed point, and each decision was made by asking what had to be true an hour earlier for it to happen.
The effect is that the trip has a shape. Rather than a list of boxes, you get a sequence with a destination, and the days build towards something instead of simply running out.
What the method actually changed
Three things dropped out of the plan immediately. A museum I had included because it appears on every list, a viewpoint that would have cost two hours of queuing for a photograph I have seen a thousand times, and a restaurant that would have required crossing the harbour at the wrong end of the day.
Two things moved in. A full morning with no fixed plan in a single district, which is where every good travel day I have ever had came from, and the ridge walk on the eastern end of the island, which I had originally cut for time and which turns out to be the thing most people say they wish they had done.
The itinerary got shorter and better at the same time, which is usually the sign that the method is working rather than that you have given up.
The version I would hand to someone else
Day one is arrival, the transport card, and one neighbourhood on foot until you are too tired to continue. No sights. The point is to stop being a person who has just got off a plane, and to learn how the city moves before asking anything of it.
Day two is the hill in the morning while it is cool, lunch somewhere unremarkable near where you come down, an afternoon that is deliberately unscheduled, and then the crossing at dusk. That is the whole plan. Everything else is optional and can be dropped without the trip collapsing.
It is a thin itinerary by the standards of most two-day guides, and that is the argument. Forty-eight hours will not let you see Hong Kong. It will let you understand its shape, which is the only thing worth attempting in the time.