The reason I chose small shops over high risk ones is because I don't want to have to deal with getting frozen and starting over. I also hate the trading post with a passion so selling items through my shop is the perfect solution. I don't know if this is more profitable or not compared to Magic/BM/DM/Merifoods/Stamps but it's a hell of a lot safer. To each his own I guess.
Shops AB'd at: 1,3,4,5,7,13,14,15,16,30,37,41,58,68,86,98,108 (LOL FUCK MAGIC)
Items bought: 4647
Of those, 64 were Unbuyables. Out of 64 UB, only 1 sold for more than 300k.
Average Items per day: 38.7
Buyable:Unbuyable ratio - 71.6:1
Most items in 1 day: 97 (Shattered this record a few weeks after the study with 138)
Best advice I can give is: Go all out on Half Price Day
Minimum profit: 5000, then 2500.
Profits: Each day I recorded how much I made onto an excel file.
Profit from shops: 88,956,000 NP
NP Generator: 3,931,000 NP
Trades/Auctions: 10,100,000 NP
Total: 102,980,000 NP
Average Profit/Day: 741,300 NP
This was done with a 1 month account. I didn't hit 3 months until halfway through. **You do not need a 3 month account to make $money$**
Restocked Banned: 37 times or 30.8% of the time.
I restocked 8h a day the first few weeks, with Abrosia automatically pricing items for me which took up a chunk of the time. Later I stopped letting it price for me. Somewhere in the middle I was RS banned like crazy, hitting streaks like 6 days in a row if you look at the chart you'll see. So I slowed down to 4h a day, then later making my way up to 6h a day again.
I started out with 5.5 - 11.0 refresh rates/.4-.7 prehaggle/1.1-1.4 OCR and eventually got bolder. That and more and more competition forced me to lower them again and again to stay competitive. By the end I was using 4.5 - 9.0/.4-.7/.6-.9.
Here is the graph covering only shop data, not NP Gen. The peaks and troughs are very jagged because some days were better than others, as well as restock bans putting a dent in the day's profits. The green line is the 1 Million NP/day threshold. The red line is the average for that range of data.
From this graph we can infer that factors such as increased used of ABing has made it increasingly harder for me to make NP each day. Efforts I made to correct this like lowering my refresh rates and OCR/Prehaggle times didn't do much.
Here is a screenshot of the Excel file. (It covers a little more than 4 months)
Red = <500k/day. Yellow = >500k, <1mil, Light Green = >1 mil, <1.2mil, dark green = > 1.2mil
3rd row shows how I often I get RS banned.
I've tried to compile as much data as I can and to see this from as many angles as possible. Let me know if there are any more stats you want to know about.