To date, singlet oxygen (1O2) luminescence (SOL) detection was predictive of photodynamic therapy (PDT) treatment responses both in vitro and in vivo, but accurate quantification is challenging. In particular, the early and strongest part of the time-resolved signal (500–2000 ns) is difficult to separate from confounding sources of luminescence and system noise, and so is normally gated out. However, the signal dynamics change with oxygen depletion during PDT, so that this time gating biases the 1O2 measurements. Here, the impact of gating was investigated in detail, determining the rate constants from SOL and direct pO2 measurements during meso-tetra(hydroxyphenyl)chlorin (mTHPC)-mediated PDT of cells in vitro under well-controlled conditions. With these data as input, numerical simulations were used to examine PDT and SOL dynamics, and the influence of various time gates on cumulative SOL signals. It is shown that gating can underestimate the SOL at early treatment time points by ∼40% and underestimate the cumulative SOL signal by 20–25%, representing significant errors. In vitro studies with both mTHPC and aminolevulinic acid-photosensitizer protoporphyrin IX demonstrate that rigorous analysis of SOL signal kinetics is then crucial in order to use SOL as an accurate and quantitative PDT dose metric.