This study investigates the usefulness of business tendency surveys collected at the KOF Swiss Economic Institute and aggregated in the form of the KOF Employment Indicator for short-term forecasting of employment in Switzerland. We use a real-time dataset in order to simulate the actual predictive process using only information that was available at the time when predictions were made. We evaluate the predictive content of the KOF Employment Indicator both for nowcasts that are published two months before the first official release, and for one-quarter ahead forecasts published five months before the first official release. We find that inclusion of the KOF Employment Indicator leads to a substantial improvement in prediction accuracy of both point and density forecasts compared to the performance of a benchmark autoregressive model.