But the question is how does it work?

We do know that people do get better with homeopathy,

What one would have expected is that large studies of high quality would have shown an effect, which wasn't the case for homeopathy. This means that clinical effects of homeopathy are placebo effects.

[The authors of the new Lancet study were Matthias Egger and colleagues from the University of Berne in Switzerland. They looked at 110 trials using homeopathic remedies and 110 using conventional medicine in matching conditions. They looked for an effect in both the smaller, low quality trials and in larger, higher quality trials. Homeopathic remedies were more likely to have had a positive effect in the small, low quality trials. In the better trials, they say, homeopathy was no better than placebo.] Our study powerfully illustrates the interplay and cumulative effect of different sources of bias, ... We acknowledge that to prove a negative is impossible, but we have shown that the effects seen in placebo-controlled trials of homeopathy are compatible with the placebo-hypothesis.

What we saw was [that] when you looked at these good, large studies, you did no longer see an effect for homeopathy, whereas you still saw an effect for conventional medicine,

Conventional medicine interventions did better than placebo, whereas the homeopathy interventions basically did the same as placebo.