I often hear from business owners our rejection rate from OEMs suddenly shot up and I don’t know why. Welcome to my talk series Mai Kya karu specially recorded for you. I’m Nalin Mehta
and today we will talk about why rejection spikes happen and exactly where to look in your data to find the real culprit. A sudden increase in rejection is never random. It’s a signal. The key to fixing it fast is knowing what data to examine and in what sequence. Start with a three layer breakdown. First segment rejections shift by shift, not just week by week or even hour by hour if your business
requires that. I have seen 80% of rejections come from night shift or even the 4:00 a.m. to 7:00 a.m. part of the night shift because one operator was on leave and his replacement skipped a crucial inspection step or they just lost focus at the most tiresome part of the night shift. If rejections spike on specific shifts and timing that points to people or process gaps, not equipment. Second, map rejections to specific defect types and which operations actually created them. Trace backwards from OEM reports. Is the defect from the first operation or the third stage? One supplier discovered 70% of rejections were surface contamination during washing and not machining. They have been adjusting cutting parameter for weeks trying to solve the wrong problem. Third, very critical. Compare current rejections against the same product from some periods in the past, say 3 months ago. What changed? Often it is something subtle. A new coolant batch, a replacement cutting tool, or a calibration drift. One client’s spike started when they switched to cheaper cleaning chemical that left micro residues. The OEM caught it and they didn’t. If you use bought out components, cross reference rejections with the material lot numbers. If they cluster around specific batches, maybe your supplier changed something. Once data pinpoints the source, act with precision. Shift specific means training. Stage specific means process control. Time based means input quality or supplier quality. Don’t implement generic program. fix the exact failure point and then share your root cause analysis with the right people. Sometimes it’s good to share with the OEM. Also, a data-backed response shows capability, vague promises show panic. So dig into hourly or shift patterns, trace defects to operations, compare historical data, and check material badges. The answer is in the data. Stay connected on LinkedIn for more insights. Thank you very much.
