Anything dead comes down to five checks, in order, and a twenty dollar multimeter does all of them.
Check one, power in. Wall, adapter, cable. Measure the adapter's output, don't trust the label. A surprising share of dead devices have a dead adapter.
Check two, the switch and the fuse. Continuity across the fuse, and across the power switch in the on position. Both fail more often than boards do.
Check three, the input reaching the board. Open the case with the power unplugged, then measure at the board's input with it live if the design allows safe probing. If power reaches the board and the board stays dark, the fault is on the board now.
Check four, the first regulator. Find the biggest regulator near the input, read its part number, find its datasheet, and measure what comes out. No output where the input is good is a cheap, real fault.
Check five, the rails downstream. Work from the regulator outward, comparing against the datasheet values or, better, a working twin.
Two rules over all of it. One: mains voltage is not a learning environment; if the device plugs into a wall and the fault might be before the transformer, that is a different discipline. Two: write every measurement down. Memory is a bad oscilloscope.
What are you chasing right now? Post the device and which check caught it, and be honest about which one stumped you.