The goal of high-throughput quantitative proteomics is to precisely quantify every protein in a sample in the shortest possible time. Here, we compared data-dependent acquisition (DDA) versus data-independent acquisition (DIA) on a Q-Exactive HF with the aim of achieving the greatest proteome depth and quantitative accuracy/precision. We initially generated spectral libraries containing 135,075 peptides representing 10,119 proteins by fractionating digested HeLa lysates. Unfractionated single-shot analysis was subsequently performed in triplicate with either DDA or DIA and analysed with Match Between Runs in MaxQuant or Spectronaut, respectively. DIA quantified significantly more peptides in all three triplicates than DDA (~85,000 vs ~57,000), and quantified more peptides with less than 20% coefficient of variation. At the protein level, DIA quantified slightly more proteins however surprisingly; the reproducibility of DDA was slightly better presumably due to the more advanced label-free quantification (LFQ) features of MaxQuant. We next modified the Single-Pot Solid-Phase-enhanced Sample Preparation (SP3) approach using paramagnetic beads with 96-well plate magnets to perform proteomic analysis of 1 mm cardiac organoids. Combined with spectral libraries and DIA, this approach enabled us to quantify >6,000 proteins per 2 h run. We applied this approach to; i) characterise organoid maturation, ii) compare matured organoids to human myocardium and, iii) perform a screen of novel cardiac regeneration compounds in pre-clinical trial.