5% margin of error: the sample size
±5% is the most used margin of error in quantitative research: fine enough to settle most marketing decisions, wide enough to stay affordable. The number to remember: 385 respondents at 95% confidence (large population, cautious 50% proportion). That's exactly this calculator's preset.
Concretely, ±5% means a score measured at 60% reflects a reality between 55% and 65%. Enough to separate 60% from 40%; not enough to separate 52% from 48% — in that case, tighten the margin to ±3% (1,068 respondents) or ±2% (2,401). The chart under the result shows this precision/cost trade-off at a glance.