For 1 gram sample, I would use 50 to 100 g silica gel (1-2% loading). I would also use a "solid load" technique.

I'd run a gradient (step gradient in your case) from hexane to ethyl acetate, then a gradient from ethyl acetate to methanol, and a final gradient from methanol to 50% water. I term this "wide polarity range" chromatography.

See:

http://www.teledyneisco.com/en-us/liquidChromatography/Chromatography%20Documents/Application%20Notes/Wide%20Polarity%20Range%20Chromatography%20App%20Note.pdf#search=wide%20polarity%20range

http://www.teledyneisco.com/en-us/liquidChromatography/Chromatography%20Documents/Posters/Wide%20Polarity%20Range%20Flash%20Purifications%20Poster.pdf#search=wide%20polarity%20range

http://www.teledyneisco.com/en-uk/liquidChromatography/Chromatography%20Documents/Posters/New%20Techniques%20in%20Extract%20Column%20Screening%20Poster.pdf#search=column%20screening

http://www.teledyneisco.com/en-uk/liquidChromatography/Chromatography%20Documents/Posters/New%20Techniques%20in%20Extract%20Column%20Screening%20II%20Poster.pdf#search=column%20screening

Please see attached some procedures that were used specifically isolating compounds from Acacia bark extracts:

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3209846/

Firstly, fractionation can be done from hexane to water soluble fraction instead of running column. It will help you to separate different solvent soluble fraction accordingly. Then, according to the desired compounds or based on bioassay guided fraction you can further evaluate your work. If you are looking for polar fractions like from the butanol fraction, using diaion as stationary phase and using water to 100 % methanol will ease your work.. Good luck with your isolation.