Somewhere in your head is a picture of designers producing slick app screens. The reality of early design looks more like arts and crafts. Boxes drawn in marker, buttons that are sticky notes, screens shuffled by hand. This is low-fidelity prototyping, and it is a deliberate professional technique, not a lack of budget.
The logic: the earlier you catch a bad idea, the cheaper it is to kill. A paper sketch costs five minutes, so you can afford to be wrong ten times before lunch. Code costs weeks, so being wrong in code is expensive enough that people start defending bad ideas rather than fixing them. Rough prototypes have a second, sneakier advantage: when you show someone a polished screen they comment on the colours, but when you show them marker boxes they comment on the idea, because there is nothing else to look at.
Low fidelity (paper, whiteboard marker, grey boxes) is for exploring: fast, disposable, honest. High fidelity (pixel-perfect, clickable) is for validating details and convincing stakeholders. The skill is knowing where on the dial you are, and the classic mistake is going high too early, then defending your beautiful wrong thing.
The magic trick: Wizard of Oz
The best-named method in the whole field: to test an app that does not exist, you fake it. The user taps your paper screens, and a human behind the curtain plays the computer, swapping pages in response. It sounds like a game, and it is, but it is also how real teams test voice assistants and chatbots before the technology is built. The user's confusion is real even though the software is not, and real confusion is exactly the data you came for.
Invent a small app. Draw six screens on six pieces of paper: home, a couple of task screens, a success screen. Recruit a sibling or friend, give them a mission like "order a pizza in this app", and you play the computer, swapping papers when they tap. Two rules: stay silent, and note every place they hesitate or tap something that is not a button. Each hesitation is a design flaw you found for the price of paper.
Search for "paper prototyping" videos, watching one real session for three minutes teaches more than any chapter. Greenberg and colleagues, Sketching User Experiences: The Workbook, is the classic exercise book.