The number of cards I use was dictated by the win rate I wanted, so I started with that & then worked backwards.
It was important to get this right, as a high win rate would mean small jackpots & a low win rate would mean massive jackpots but people might be put off...as it would hardly ever get won.
I knew roughly how much cash I'd be taking each week & I knew what sort of level I'd like the maximum jackpot to reach (around £400). From that, I worked out that I needed a win rate of about 1 in 14.
Once I knew the win rate I needed, I played out the game many hundreds of times at home with various different numbers of cards - some where you could change the first card & some where you could not. This showed me how many cards I needed to use & whether I could let them change the first card in order to get the required win rate.
It worked out that using 8 cards & giving them the option to change the first card produced the win rate I was looking for...about 1 in 14.
Common sense suggested that 8 cards were best split into 2 rows of 4, so I then went out & got the materials to make up the board.
I always feel that the more planning you can do when trying something new, the less change there is of things not working out how you hope!