The radial velocity technique is commonly used to classify transiting exoplanet candidates. However, stars are intrinsically noisy in radial velocity. No good description of this noise has yet been proffered, although activity in general has been suggested as the source, making it impossible to evaluate its effect on signal detection. In this poster, we propose an activity-based model that incorporates both light and dark stellar spots, capable of producing both photometric and radial velocity time series. We demonstrate its consistency with both SOHO/VIRGO photometry and SOHO/GOLF radial velocities. We then use this model to establish lower and upper limits on the effects of intrinsic stellar noise on the metal lines used to follow up transit candidates, making use of Monte Carlo simulations. Based on these results, we can suggest an optimal observational sampling rate.