Introduction Primary endoscopic and percutaneous drainage for pancreatic necrotic collections is increasingly used. We aim to compare the relative effectiveness of both modalities in reducing the duration and severity of illness by measuring their effects on systemic inflammatory response syndrome (SIRS). Methods We retrospectively reviewed all cases of endoscopic and percutaneous drainage for pancreatic necrotic collections performed in 2011–2016 at two hospitals. We assessed the post‐procedure length of hospital stay, reduction in C‐reactive protein levels, resolution of SIRS, the complication rates, and the number of procedures required for resolution. Results Thirty‐two patients were identified and 57 cases (36 endoscopic, 21 percutaneous) were included. There was no significant difference in C‐reactive protein reduction between endoscopic and percutaneous drainage (69.5% vs 68.8%, P = 0.224). Resolution of SIRS was defined as the post‐procedure normalization of white cell count (endoscopic vs percutaneous: 70.4% vs 64.3%, P = 0.477), temperature (endoscopic vs percutaneous: 93.3% vs 60.0%, P = 0.064), heart rate (endoscopic vs percutaneous: 56.0% vs 11.1%, P = 0.0234), and respiratory rate (endoscopic vs percutaneous: 83.3% vs 0.0%, P = 0.00339). Post‐procedure length of hospital stay was 27 days with endoscopic drainage and 46 days with percutaneous drainage (P = 0.0183). Conclusion Endoscopic drainage was associated with a shorter post‐procedure length of hospital stay and a greater rate of normalization of SIRS parameters than percutaneous drainage, although only the effects on heart rate and respiratory rate reached statistical significance. Further studies are needed to establish which primary drainage modality is superior for pancreatic necrotic collections.