Ok, my takecwould be to really think about root cause for the error. Whilst it was “human” error, if you’ve had 3 more recently and none before is there a reason for the change? Really ask why, then why, and why again..drill down to how mistakes could have been possible
then you need to look at “fool proofing”. What is sometimes reffferred to as poke yoke. (Japanese lean terminology). It is completely unrealistic to expect humans not to make errors, so as a company you need to engineer your ways of working to prevent errors. If You do the analysis of how it was possible to make the errors, it should then yield up fool proofing changes.
so an example, a need for 2nd person verification if data is that critical, or to change the visual appearance the entry screen to make worng digit entry more difficult (eg spacing, forcing decimal points etc). Or it might be you need breaks and therefore role needs to be split between 2 people in a job reorganisation so you’re not concentrating for long periods. Or even how the data can be entered automatically so you’re not concentrating are just verifying and confirming. All these things cost money, so it needs be be balanced with the costs of those errors.
do the analysis therefore on cost of those errors - is it customer perception or actual lost revenue?
I would go into the meeting then armed with this- a clear suggestion of how to stop errors in future. It gives them a clear understanding you have taken this seriously, you aren’t denying but are identifying weaknesses in system and how to prevent.
I would also calculate my level of errors, how many errors per 100 transactions and % level of accuracy. They’re asking for 100% which is frankly, for humans, unacheivable. We are all flawed by being human. But if your rate overall is in 99% plus range you are in a good position to point out that to expect 100% is flawed and ways of working need to be changed.
And here is a paper you could use if it comes to disciplinary hearing..ts bonkers and unrealistic to expect a human to work like a machine
https://www.sciencedirect.com/science/article/abs/pii/S0747563211000707