Apple cluster control is one computer managing many iPhones: screens in one place, tasks sent by group, results collected together. It does not use jailbreak. External input drives the devices.
Three places you can save, and four you cannot
Someone in cross-border work ran two setups that did the same job. The first cost over 6,000 yuan. The second cost 2,800 yuan. The gap came from things he did not buy because he would not use them.
The device tier is the largest price gap. iPhone 8 through 12 is enough for automation. Second-hand units are 800 to 1,500 yuan, and the same budget covers two or three times as many phones as new ones. When you choose, look at the OS version. HID needs iOS 17 or above. Skip phones whose battery health is too low. Be careful with a replaced screen, because touch can be affected. USB HID starts with EC iOS USB 10.7.0 (2026-09-08). It does not add a development board. The phone is iOS 17 or above, and it is mutually exclusive with “Start Automation” in the console. During testing, iOS 15 and above can use the offline main program v7.2.0 and above as the USB automation program. The download page has an IPA that is already signed, so you do not have to buy a signing service first.
Do not buy the whole fleet at once. Fifty phones at once has a lower unit price, and when the direction is unproven, the unit discount does not cover the loss of buying wrong. Buy three to five, get the flow running, then add to the scale.
Buy only the licence you will use. A USB device licence covers debugging, running scripts, and recording. A USB mirroring licence covers the picture inside the cluster-control interface. If you only run scripts and do not watch the screen in the console, the mirroring licence can wait. If you later need to see the picture, that spend still happens. The timing moves later.
A USB hub with its own power supply cannot be cut. Idle current is very small. When a batch task is sent, dozens of phones light up at once, the inrush current exceeds an ordinary hub, and devices go grey one by one. A cheap hub is often fine during a connection test, and it drops once tasks are running.
Certified data cables cannot be cut. Off-brand cables work and then fail. Today they are fine, tomorrow one phone will not connect, you suspect the software first, and half a day later it is the cable. A poor cable used long enough can also damage the port.
Keep spare devices at five to ten percent of the total. With no spare, one failure stops one production line, and the output of the repair days is gone.
Your own time takes a large block as well. Debugging, troubleshooting, and reconfiguring are heavy in the first two months of rollout. If time is expensive, a plan with a wider service scope is the better fit.
Saving on the hub shows up as a batch of drops halfway through a task, fine one moment and failed the next, and fine again during a connection test. Saving on cables shows up as one device dropping over and over, recovering when you change the cable, failing again later, until you find it is the third cable on the same line. Saving on backups shows up as no replacement on the day a device fails, and the task stops. Saving on service shows up as nobody to ask when something breaks, two days of figuring it out yourself, and those two days often exceed the price difference you saved.
| Item | Can be trimmed | Leave it in |
|---|---|---|
| Phones | Second-hand, a model that is enough | None |
| USB hub | A small test can make do | A live batch run needs its own power supply |
| Data cables | None | Certified cables, and off-brand stays out |
| Spare devices | The ratio can be discussed | Leaving no spare at all |
| Licence | Buy only the one you will use | When it is time to buy, do not delay |
| Computer | Use the one you already have at first | Replace it once the scale is there |
| Service scope | Can be cut if you can operate it yourself | Leave it in when nobody on the team can carry it |
Saving on the hub trades away stability. Saving on spares trades away whether you can keep running when a phone fails. Saving on service trades away the time you spend diagnosing it yourself.
Our own pricing
EasyClick Android is 199 yuan a year, with no device-count limit. iOS licences are charged per phone, split into a USB device licence and a USB mirroring licence, and you buy the one you use. A licence covering one thousand devices sits between 2,000 and 3,000 yuan a month. There is no extra charge by person, by session, or by feature module. Run three to five licences for a week first, and add more once the direction is confirmed.
A personal developer account is a separate bill, paid to Apple, 688 yuan a year, up to 100 devices, supporting iOS 12 and above. It is separate from the software licence above. TrollStore supports iOS 15 through 16.6.1. A restart drops the signature, and you sign again. Offline 7.0.0 and above: a single-app setup signs only the main program, and a dual-app setup also signs the proxy.
The 3,000-plus yuan saved on the second setup came from buying phones second-hand and starting with five to validate. Those five confirmed the direction worked. If it had been equipped for fifty at the time, a wrong direction would have been more than 30,000 yuan. Devices look the most certain, and they are also the easiest to overbuy before the scale is proven. If you are still negotiating with a vendor, what one set costs, and where the gap is covers how to align two quotes before you compare them. The full budget list splits hardware, subscriptions, and hidden costs item by item.
About EasyClick: A mobile automation script development platform covering three phone ecosystems: Android without root, iOS without jailbreak, and HarmonyOS. It also offers Apple script writing, Apple cluster control, Android cluster control, remote cloud control, and AI phone cluster control. → Explore all products
Ready to build it for real?
Every approach in this article can be built on the EasyClick phone automation platform — full documentation, developer tools and cluster/cloud-control products, free to try.