Abstract Study question What are the types and frequency of errors occurring in the laboratory identified by an electronic witnessing system (EWS)? Summary answer Several types of mismatch categories were identified, and the rate of a potentially catastrophic true mismatch error was < 1%. What is known already Mix-ups in the IVF laboratory have been recently highlighted in the media. These errors can have devastating long-term consequences for the fertility clinics and families involved. Prevention of these errors should be paramount in the laboratory. Study design, size, duration A retrospective analysis was conducted using 15 months of electronic witnessing data from two IVF laboratories in the United States. A total of 11,357 witnessed procedures were evaluated to determine the type and frequency of mismatches that occurred. Participants/materials, setting, methods All procedure match and mismatch data was recorded by an EWS (Matcher, IMT International). The mismatch data from two laboratories was evaluated to determine the category and occurrence of mismatches taking place in the assisted reproductive technology (ART) laboratory. Main results and the role of chance The overall mismatch rate was 0.78% (89/11,357). The types of mismatches were identified as: 1) Program: duplicate barcodes scanned, barcode misread by program, non-Matcher barcode picked up during scan, connectivity issues; 2) Wrong patient selected; 3) Transcription error; 4) Cycle set up: treatment cycle not set up correctly, donor not linked to treatment cycle; 5) True error: dishes and/or tubes from two different patients were in the same workstation at the same time. Program error happened most frequently at 0.30% (34/11,357). The wrong patient selected and cycle set up errors occurred 0.18% (20/11,357), followed by transcription error at 0.11% (12/11,357). A true error occurred 0.03% (3/11,357) and was rectified in < 10 seconds meaning there was no time for actual mixing of gametes to occur. These findings agree with previously published reports worldwide where the occurrence of a true mismatch is < 1%. Limitations, reasons for caution The present study retrospectively analyzed mismatch data gathered over a 15-month period after implementation of EWS. Further investigation is warranted to determine if mismatches occur more or less frequently after long term use of an EWS. Wider implications of the findings Failure to identify a true mismatch error may lead to a catastrophic error. Continued evaluation of EWS may provide insight into how and why a true error occurs. This information can then be utilized to improve laboratory protocols and procedures for enhanced patient protection. Trial registration number Not applicable