In the end, quality cannot come in at the end
Where knowledge work still has not caught up with manufacturing
Oh, the irony: The office work of the 20th century was heavily influenced by how factory work was organized. But while manufacturing has evolved decades ago thanks to Lean, the knowledge work of today is still deeply rooted in Taylorism. In other words: People who have to solve problems via creativity and collaboration are treated as if they were workers on a manufacturing belt (before Lean Manufacturing).
My personal example for this is how quality is handled in digital product management: Usually put at the end, with one or two people responsible for quality, no decision power coming with the dedicated role and management pressure to cut down testing time. If later on production the resulting quality is not at the needed level, how much of a surprise is it? As an extra joke: The introduction of measures to improve software quality often comes with a low quality itself.
This is where everyone involved in the discovery and delivery of a digital product could show some humility. In manufacturing, effective ways to improve overall quality (and therefore overall business results) are old and known.
I remember learning about Computer Integrated Manufacturing (CIM) at the university in the late 1990s. It started with Computer-Aided Manufacturing (CAM). Soon a lot of other disciplines were added - Engineering, Design, all named CAx. While they were all nicely lined up, one was different: Computer-Aided Quality (CAQ) was not added at the end, but ran in parallel to the whole production.
Until today, I have heard many times people preaching to introduce quality early on, mostly in the form of "this new tool will fix everything" or "the testers have to step up their game". But I have never seen this fundamental understanding about quality that I learned more than 25 years ago put into practice. It is time to end the arrogance and ignorance.
Queen - Machines (Or Back To Humans)