It often happens that the citations to articles are recorded with some deviations from the original name... this will bring some incertitude in the calculation. But in fact it is the same with WOK and SCOPES which recode many "secondary documents" with writing variations.
The impact factor from GS data will be very different from that of the World of Science, because of the quantity of data measured. The Publish or Perish database does the calculation for you, or at least it gives you the data in a way that you could calculate the impact factor yourself. In principle, you should divide the number of citations in 2017 to the journal you measure, by the number of publications of this journal in the last two years, 2015 and 2016.