Apple cluster control is a technique for centrally managing multiple iPhones from one computer, without jailbreaking, without plugins and without dev boards.
This one answers: How Do You Run Amazon Stores With Apple Cluster Control.
Liang sells home goods across five US-marketplace stores.
His day goes like this. Up at seven, first thing is walking through all five dashboards for orders. Again before lunch. Once more before bed. Five stores, fifteen passes a day.
Fourteen of those passes show nothing. You still cannot skip them, because the one that matters is the one you missed.
The three things Amazon sellers actually fear
The first is hijackers. A listing you built over six months, and one morning there is another seller on it, two dollars cheaper. The later you notice, the worse it gets.
The second is reviews. Especially one sitting near the top. A single review with a photo can bury ten good ones. Amazon’s review rules are tight and there is only so much you can do, but you need to know when it happens.
The third is stock. Selling out with no inventory costs you listing weight. Dead stock keeps taking up space.
All three share one property: they need to be noticed quickly, not handled quickly. Noticing is repetitive. Handling is judgement. That distinction matters.
What to hand over, what to keep
Order sweeps, stock threshold checks, delisting sold-out items, relisting arrivals, sorting buyer messages，all of it can go to scripts.
Hijackers are different. A script can help you notice one, not deal with one. The detection logic is fixed: open the product page, check for other sellers, record it if there are. Whether to cut price, send a warning, or go through Brand Registry is a human call, and it may differ per store.
Buyer messages work the same way. Most Amazon Buyer-Seller Messages fall into three buckets: when does this ship, can I return it, is the address right. Those can be templated. Claims, A-to-Z cases, and the messages that are clearly fishing for something need a person. So the script’s job is sorting, not replying.
One detail people miss: Amazon expects buyer messages answered within a day, and missing that hurts account health. Nobody can watch five dashboards. A script can.
Devices and connections
Five stores means five phones on a desk that never move. A data cable is enough: console 10.7.0 or newer with phones on iOS 17 or newer, one cable each, no Bluetooth board, no OTG board, no jailbreak, no signing.
If the store count passes a dozen or so and the phones no longer fit on one desk, look at WiFi. To drop cables entirely, a Bluetooth board (ESP32C3) covers it; desks out of cable reach are the only case for an OTG board (ESP32S3, iOS 17 and above). The question is whether you have a distance problem, not which link is faster.
Past a handful of devices, two physical things come first: powered hubs, and spacing between phones. Neither is a software issue. Connection options has the detail.
Binding stores to accounts
One store, one device, physically bound. Do not rotate two store accounts through one phone.
Amazon’s account linkage checks more than login details，device environment and network exit are in there too. Switching two store accounts on one phone links them deliberately. What you save on hardware, you bet the store on.
Group and name devices as you grow. Splitting by marketplace (US in one group, Europe in another) with the marketplace, store and number in the name works. The slowest part of running many stores is not the tapping, it is remembering which phone belongs to which store.
Keep the dashboards from moving at the same minute. Two batches, ten or fifteen minutes apart, reads more like a person than five at once.
How many stores before it pays off
One or two stores by hand is fine. Around five, with every store needing a look every day, it starts to show.
Start read-only if you are testing: an order sweep and a stock threshold script, run for two weeks. Getting those wrong only means missing something. Only then add the writing side，delisting, relisting, repricing，where a mistake touches real business.
Liang now lets scripts run orders and stock each morning and summarise the result. Hijackers and reviews are detection only; he still handles them himself. Time spent on five dashboards went from over two hours to around twenty minutes.