Previous research made at RISE Bioeconomy (former Innventia) has shown significant improvements to both retention and formation when retention AIDS, cationic polyacrylamide (C-PAM) and microparticulate silica, were added very close to the headbox given that the mixing was adequate. The pilot trial that showed these results used a somewhat idealised system since the furnish used consisted solely of fibre, filler and retention aid. In addition to these components, it is very common to add some cationic starch to the thick stock to increase the paper strength. When cationic starch (0.5%) had been added to the thick stock there was no longer any obvious positive effect on the retention-formation relationship with the late dosage of the retention system. This spurred a further investigation and trials in which also the cationic starch was added just prior to the headbox, such that the contact times of all three components (C-PAM, microparticulate silica and cationic starch) were less than 3 seconds prior to forming. Also in these cases, the general positive effect on the formation-retention relationship was not seen, but the C-PAM dosage needed to obtain a certain retention level was much less as compared to when the normal dosage positions were used. This clearly demonstrates possibilities to drastically decrease the C-PAM consumption. Only marginal negative effect on the mechanical properties of the produced paper was seen when the starch was added just prior to the headbox as compared to in the thick stock. It is common wisdom that starch should have a long contact time to give optimal strength gain. The presented results show that this needs not to be the case, and the key is probably adequate mixing.