Three things to know.
It never decreases when you add a variable, so it cannot be used for model selection. Use adjusted R-squared, or out-of-sample performance.
High R-squared does not mean a useful model - two trending series regress beautifully on each other and mean nothing. That is spurious regression.
In finance, low is normal. A genuinely predictive return signal often has R-squared well under 1%. Candidates who dismiss that as failure reveal unfamiliarity with real return data, which is exactly what the question is probing.