Consider 1 kWh = 1000 W * 3600 seconds = 3,600,000 J.
1 kg lifted 1 m = ~10J.
Let's use 1 metric tonne. 1000Kg. So we get 10,000 J per meter of elevation.
3,600,000 / 10,000 = 360 meters. Around a thousand feet.
Ok. Getting a 15 cm diameter well dug and cased is on the order of \$20/foot. So \$20,000 buys you a 1000 foot deep hole. Use whatever length of 10 cm bar stock gets you a ton. (In actual fact, using drill collar would be a prefab way to do this) Now all you need is some long cable, and a motor/generator.
But really: Add a few more chunks of drill collar increase your weight up to, say 50T.
Now you have a battery that will store 50 kWh -- about 2 days use by a modern home -- and it cost you only somewhere between 20 and 40 grand. Ball park a thousand bucks per kWh.
50 kWh is worth about \$4 at residential prices, about \$1.50 at wholesale prices.
Now we sort of do this with pumped hydro storage. Water is a lot easier to work with than long slugs of steel pipe suspended by cables. It also scales better. Raise the rim of the reservoirs a few feet, and you can store a lot of extra tonnes of water. (A 1 km2 lake 1 meter deep is a million tonnes of water. The cycle efficiency is only about 70% or so, depending on how much you spend on big plumbing. (You put 10 kWh in and get 7 back next Tuesday)
Pumped hydro is very workable. If you have a mountain nearby.
Big LiON batteries are now getting down between 200 and 300 bucks per kWh. (https://www.nrel.gov/docs/fy19osti/71714.pdf 2018 figures) There are other technologies in the wings that should cut this by another factor of 4 for stationary batteries.
Even those mongo batteries such as Tesla made for Australia pay off only when used several cycles per day. They have addition payoff because they can prevent firing up gas generators to fill the gap as often -- takes about half an hour to bring a gas turbine online.
TL:DR; Gravity batteries are too big and bulky, and are too complicated to scale up easily.