Magnetic activity affects the observed properties of solar p modes: mode frequencies and linewidths increase with solar activity. Using VIRGO/SPM data from solar cycles 23 and 24 we show that solar-cycle variations are measurable in the temporal autocorrelation function of the p modes. Following a method developed for local helioseismology, we measure the p-mode travel times for multiple skips and propose to average these to enhance the signal-to-noise ratio. This method is robust to noise, simpler to implement than peak bagging in the frequency domain, and promising for asteroseismology applications. We estimate that the activity cycles of Sun-like stars may be detectable in long photometric time series.