I think I have done exactly what he suggests with the Ikea mirror with mediocre results but will do another test as he claims it works on an unheated bed.
I thought that the guy in the video said that he ran his bed at 70C. Maybe he said both.
It shouldn't be that hard to get your print to stick to the bed. If the bed is warm enough and free of contaminants, the molten plastic should deform to conform to the bed surface and stay there until pried away. If the bed is too cool, it will quench thr molten plastic before it can conform to the surface. If the molten plastic is too cool to conform to the bed surface, the effect will be the same.
If the room the printer is in is too cold or too drafty, it may adversely affect both the bed and melt temperatures, keep the printer warm and out of drafts.
Wait for the bed and hot end to warm up before starting to print. Note that the actual melt and bed surface temperatures may lag behind the reported hot end and bed temps. Print a test line at the bed periphery immediately before starting the print. Think of this as a combination prime and nozzle check. You might choose to add a (short) skirt or even a brim in the slicer. Inspect the skirt for defects.
Try printing the first few layers slow and thick.
Your hot end has two fans, one to cool the heatsink and the other to cool the printed object. The former should always be running. The latter should be under firmware control. Check to see that they are wired properly and turn off the latter if the bed is too cold.
Do you have a comprehensive assembly/setup/troubleshooting guide for your printer? If you do, go through it carefully to spot possible defects such as poor electrical connections, bowed bed or axes, etc.
Sorry. I have to stop here. Good luck.