I seem to recall making a suggestion to do a thorough mineralogical investigation on a size-by-size basis to identify the opportunities to reject Fe and Si. Where it went in the www/ether is an open question but it may also be that it did not get posted due to getting a new computer and troubles familiarizing myself with the keyboard and ultra-sensitive mousepad. Unless the mineralogical characterization is done, you cannot have the confidence that a suggested beneficiation process is going to work; A good mineralogical/material characterization study will give you that confidence and point you in the right direction.

I agree, knowledge first. First step would be to "know your ground", full majors elemental analysis (Na if you can, K, Mg, P, Ca, Ti, Cr, Zr, Ni). Is your reported Fe FeO and Si SiO2?

Then XRD, but you may have poorly crystalline or amorphous material included, but the knowledge would be useful anyway.

Then possibly a mineral to element map, either through an automated mineral analyser (ala MLA/QEMScan/INCA etc.) or just find a friendly SEM and run a quick scan or two. This will give you more knowledge on deportment and association, preferably in intact samples. If your presented datain your message is in oxide form you still have 28% other elements(+oxygen). Then you will get an idea of which metallurgical approach to take. The SiO2 deportment will be essential as it is a prerequisite target.

Or, you can do it the limited pre-knowledge way, and do a large batch of incremental metallurgical tests. Both approaches work, depends how much time, labour, laboratory facilities and money you have.

As this is posted in geomet, you would also want to understand the variability in your ore. Do you suspect significant variability across your ore zone? Grain size/agglomerate size may then become important.