Quality Control
Traditional Quality Control relies on customer feedback (CSAT/NPS) and random selection. It’s a very manual process. In the end, it also creates a very large focus on the negative. When you go over mistakes when teaching, it can create discomfort and fear. Speed and efficiency can dip before recovering for the final improved result. While this should be part of the process, I don’t believe it should be the main focus.
When going through issues, it’s possible to check how long certain issues take, assess how long they should take, try to find areas where things are moving slow, and address them. A lot of “getting a feel for where we can improve”
In the example below, I automated all of that as a way to scale training with larger teams. It takes the average time to resolve any given issue type for everyone, and the average time for each individual. It then targets and highlights who to train, and what issues to train them on to ensure they’re comfortable. It essentially says “train these people on these custom issues for each of them, if they reach average resolution speed, this will free up the most time with the least effort based on speed and volume of issues.
Doing this targets areas people are uncomfortable with and are likely to make a mistake on but haven’t necessarily made a mistake on yet. It also makes it quick to spot people who are very fast, and by checking their work, you can quickly find out if they’re taking shortcuts, or just that good! By implementing this strategy, I increased volume resolved by nearly 50% in the first month on a large project, and the accuracy of traditional QA increased.
