"...like, 99% accurate."

Is 99% accurate good enough in a medical business?  Let's see what 99% means.

For every 100 labs filed, one gets lost = 4 per month = 50 per year = out of business

Imagine doing the wrong procedure only 1% of the time.  I'd make this mistake 10 times per year.  This is unimaginable.

In correctly entering 1% of insurance data causes payment problems on 100% of those patients.  So for 1% of 100 patients per week = 1 patient per week = $ 100 per charge on average x 52 weeks = $5200 per year lost. (This is a very conservative estimate.  Would actually be much higher.)

How about recording the callback number wrong on 1% of people who call for test results.  I'd have difficulty calling 2 per week or 100+ per year.

An on and on...

How about in life outside of medical practice?  Is 99% good enough?

Nope.  Not here either.

Examples:

An average drivers puts 12000 miles per year on the car.  If they have an accident 1% of the time, they crash the car every 120 miles.  Not too good.

Would you get on a plane that had a 1% chance of not landing?

So 99% is a good grade on a test or a good GPA.  Otherwise, not so good.