The excessive search for parking, known as cruising, causes pollution and congestion. To mitigate these negative effects, cities are exploring new methods. However, accurately measuring the number of vehicles searching for parking is challenging and typically requires sensing technologies. In this paper, we propose an alternative method that eliminates the need for such technology by leveraging parking meter payment transactions to estimate parking occupancy and the number of cars searching for parking. Our estimation scheme is based on Particle Markov Chain Monte Carlo (PMCMC). We validate the PMCMC approach using data simulated from a 𝐺𝐼/𝐺𝐼/𝑠 queue, demonstrating that it produces asymptotically unbiased Bayesian estimates of parking occupancy and key model parameters such as arrival rates, average parking times, and payment compliance rates. Finally, we apply our method to parking meter data from SFpark, a large-scale parking experiment, and subsequently compare the Particle Markov Chain Monte Carlo parking occupancy estimates to the actual data from parking sensors. Our approach is easy to replicate and scalable, relying solely on historical parking payment transaction data that cities already have.