Two ways to write the same formula, one clearer path forward
The first approach is better, and it's not a close call. When every cell in a chain references the same source cell instead of the one directly above it, you build resilience into your spreadsheet. One broken cell won't cascade into a column of errors. That single design choice can save you from chasing failures downstream, failures that often look like random blanks until you trace each dependency back to its origin.
Consider what happens in practice. The second method, where each cell points to its immediate predecessor, creates a fragile chain. If cell A2 breaks, maybe a manual edit, a deleted row, or a corrupted reference, then A3, A4, and every cell after it will also break. You now have a row of errors that all look the same but only one root cause. Debugging that means checking each link until you find the first one that failed. In a long column, that's time you shouldn't have to spend. The first method isolates the problem. If A2 breaks, A3 still works because it checks A1 directly. You fix one cell, not a dozen.
This isn't about coding elegance or Excel trivia. It's about how you protect your own work. Spreadsheets grow. What starts as a three-row formula can become a hundred-row dependency map within a week. The choice you make today, reference the source or reference the neighbor, determines whether that growth is manageable or brittle. The source-reference pattern scales cleanly. Each cell stands independently, and errors stay contained. That's the kind of design that lets you trust your data at a glance, instead of needing to audit every link.
So when you write your next chain of formulas, ask yourself which outcome you want: a single point of failure or a single point of repair. The answer shapes how your spreadsheet behaves under pressure. And in a world where data moves fast and changes faster, that clarity is worth building into every cell.