I do performance / app triage work, but see the same thing. Often I walk in to a supposed "emergency" only to discover the problem has been occurring for months, if not years. Often, there is a significant cost (IE: in the millions) but either the organization isn't willing to remediate, or isn't even aware of the full scope of the cost (IE: "It's not my budget so I don't care"). In at least one case, I came across a security problem where the response was "oh yeah, we've known about that for years". Sigh. Sadly, too often unless companies have a very large customer who gets angry with them, or they are publicly shamed for a problem, they just let it magically go.