Peter - various databases use different metric scoring systems. Some are more generous than others. It's not about being 'organised'. Perhaps the most 'conservative' is ISI WoS. The most generous is often seen as Google Scholar. Scopus tends to come in the middle - and RG tends to sit around the Scopus measure. I have close to 4,000 on Google Scholar - but you will see that it is around half that on RG.
From what I understand both RG and Google Scholar count citations they have in their own database. As RG has less than GS - because they only have the publications that the members list themselves - they will tend to report less citations. GS is in that sense perhaps more correct, but there are other problems with them as they refuse to publish how their alegorithms work. But the site mentioned by Anton Vrdoljak seems to be more accurate on this. :)
With ResearchGate, it is not only the publications the members list, but rather publications posted with full-text versions from which ResearchGate can extract citation information.
I think we have answered this question before. It's partly a function of the journals in researchgate's database. Additionally, I think it's researchgate's presence. If you are on researchgate and the author citing your work is not, then researchgate cannot record the count but Google Scholar will if the journal is indexed there.
Researchgate depends on what we upload here, If scholars who cited you did not upload their articles on researchgate, your citations will not be counted. The same applies to google scholar, authors who cite you should have published with journals indexed on google scholar.