From what I'm reading its purpose is as a power sink for areas where day and night rates differ and/or where you get different rates on power you produce vs. consume (for a variety of reasons). You buffer power from the cheaper hours and use it in the more expensive. I don't really see a big upside there in most places and this is probably more a "see, we can do it" stunt. The price is cheap for that kind of battery, and that is a good indicator it is really some kind of test flight.
$250/KWh is a good price on deep discharge AGM lead-acid cells... and you can't fully discharge them without impacting their service life. This is $3500 for 10KWh ($350/KWh) you can discharge all the way down and it's all packed and wired (you can probably guess what the cabling on a bank 10KWh of AGM cells runs). I don't see the inverter included so I don't know where they come up with the 92% efficiency. Nor do I see the charge controller. So I'm taking this as a decent, below market price on a brick of high-end batteries. No more. It's a neat idea and I am guessing Tesla isn't making any near-term money on this. If it happened that I needed a big pile of batteries for some kind of home power system, I'd be looking at these. By comparison, long-life NiFe batteries run about $800/KWh (and more).