The distinction that governs everything
Debt is a tool for bridging a gap that will close. If a business is short of cash because customers pay in sixty days and suppliers want thirty, borrowing bridges a real timing gap and the loan repays itself out of the cycle it funded.
If a business is short of cash because its margins do not cover its costs, borrowing does not bridge anything. It converts an operating problem into an operating problem plus a debt service obligation, and it buys time at a price.
Nobody else will make this distinction for you. A lender's job is to assess whether you can repay, not whether you should borrow.
Specific cases where it usually goes badly
Funding losses without a specific, dated, costed plan for reaching breakeven. Borrowing to make payroll for the second month running. Taking an advance to service another advance. Financing a permanent working capital need with a short-term facility, which guarantees a refinance at whatever terms exist on that date. Borrowing to buy time on a decision you already know you have to make.
Two questions worth answering in writing
First: what specifically will this money do, and what will be different in twelve months as a result? An answer that is not a number is not an answer.
Second: if the plan does not work, what happens? With a personal guarantee, the answer usually involves your personal assets. That is a manageable risk when the plan is sound and a bad trade when it is a hope.
What to do instead, sometimes
Renegotiate payment terms with suppliers, which is free and frequently possible. Chase receivables properly. Cut the cost base before it is forced on you. Sell or lease back an underused asset. Take a smaller facility than offered, because the amount someone will lend is not a target.
And where the business needs restructuring rather than capital, a conversation with an accountant or a turnaround advisor is cheaper than any loan.