Pre-analytical variables interact with standard coagulation parameters. How these variables affect the platelet function analysis is not completely known. How collection site and puncture method affect multiple electrode aggregometry (MEA) and platelet function analyzer (PFA-100®) was compared regarding contact activation. First, volunteers scheduled for elective cardiac surgery had blood collected from four lines: venous, arterial, central venous and by venipuncture. MEA and PFA-100® were analysed blinded for site origin. Second, two samples (citrate, Corn Trypsin Inhibitor, CTI) were collected in syringe or vacuum tubes. Thrombin generation (TG) was determined. MEA was triggered by adenosine diphosphate (ADP, 6.4 µM), arachidonic acid (ASPI, 0.5 mM), collagen (Col, 3.2 µg/ml), ristocetin (Risto, 0.2 mg/ml) and thrombin receptor-activating peptide (TRAP, 32 µM). PFA-100® was triggered by collagen/epinephrine and collagen/ADP. TG was assessed in platelet-poor plasma with 1 pM tissue factor and 4 µM phospholipids and without trigger. Data were analysed using a two-way mixed-effects model for the intraclass correlation (ICC) and by the Mann-Whitney U-test. MEA and PFA-100® revealed good correlation (ICC) between the sites. CTI inhibited TG significantly more effective than citrate. Contact activation was independent of the collection method. Only the MEA ASPI test revealed significant differences between the two collection methods. Blood sampling from all lines for MEA and PFA-100® assays is justified. Contact activation is always present. Apparently this does not influence platelet function test results. Collection methods do not seem relevant, although, one should always consider using a standardized method.