What Severance gets right about the wrong way to work

Series 2 of my favourite streaming show, Severance, has just reached its climax. It’s unlike anything I’ve watched before. An incredible blend of character depth, humour, intrigue, suspense, and drama, it keeps you hooked from start to finish.

If you haven’t come across it, the story follows Mark, who leads a team of office workers at a company called Lumon, whose memories have been surgically divided between their work (innie) and personal lives (outie). As they begin to uncover the truth about their jobs (and themselves) the mystery deepens in unexpected ways.

What makes Severance particularly fascinating is its striking parallels to our own working lives. Naturally, it got me thinking about how we do Digital Engineering.

Here are my reflections (I've tried to avoid spoilers but no guarantees, so proceed at your own risk):

Deviations from optimal function

• Lumon’s strange, maze-like, sterile office reminds me of enterprise architectures or business processes which can be opaque, overcomplicated, slow, and isolating, a maze of workflows, object types, and permissions where you can’t find anything without help.
• Kier’s Principles, which form the basis for Lumon’s strange corporate religion parallels the gospel of historical company policies, or “the way we’ve always done it” dogma. Legacy doesn't always just refer to data and systems.
• Just like Lumon maintains tight control while giving the illusion of autonomy, many system vendors give companies tools but restrict how they can use them, update them or migrate away from them. The illusion of empowerment often masks rigid structures. "The surest way to tame a prisoner is to make him believe he is free" Harmony Cobel
• The department in which the characters work is called Macrodata Refinement. the work is mind-numbing data processing or cleansing. It's not dissimilar to data administration teams found in large organisations. The team are forced to work on data refinement without knowing the bigger picture, without context and making decisions without insight. A Mr.-Milchick-like data enforcer, or project manager/quality controller, is always looking over their, and Engineering, shoulders.
• Too often, companies operate like Lumon, making big decisions in a vacuum, far from the realities of daily work. The Board or Severed Floor management doesn’t care what the Macrodata Refinement team thinks or feels, and in many businesses, it’s the same story - strategy rolls downhill, disconnected from the people expected to make it happen.
• What Lumon calls influence is more accurately manipulation, enforced through punishment and a system of trivial, often bizarre, rewards. In this case I would be curious as to the impact of Waffle Parties, Music Dance Experiences or finger traps as rewards for great product data management. My experience of punishment for not having releases ready or changes processed on time is more status update meetings. Rather than that a trip to series 1's "The Break Room".
• Communication from management to the team, and even from Outies to Innies, is largely condescending and hierarchical in nature. I'm not suggesting that this is the industry norm however in organisations where multiple layers of management insulate decision-makers, and employees are discouraged from stepping outside narrowly defined roles, communication becomes more about maintaining structure than enabling progress.
• The split between Innie and Outie is by far the biggest silo, but departments are also kept apart, and even encouraged to be wary of each other. You don't need me to spell out the parallels here.
• This isn’t a failure mode exactly, more of a universal dilemma, in and out of work. Do we make personal sacrifices for the greater good, or do we do what serves our own interests? Organisations often expect individuals to act in the company’s best interest say, by putting in extra effort that benefits someone else further down the line. But that expectation runs counter to how people are wired and how most companies are structured.

What happens as a result of this kind of environment shouldn’t come as a surprise. When I think of the rebellious character Helly R, it reminds me of Share PLM's explanation of human nature and resistance to system implementations. In Severance, the fallout isn’t subtle, it’s disruptive, sustained, and ultimately destabilising. And the same goes for real organisations that ignore the people doing the work.

Hope

The world of Severance shows us what happens when people are cut off from meaning, context, and each other, and reduced to dispensable resources, valued only for their function.

"The good news is: hell is just the product of a morbid human imagination. The bad news is: whatever humans can imagine, they can usually create." Harmony Cobel

While Ms Cobel stated it as a grim truth, the idea that "whatever humans can imagine, they can usually create" fills me with hope. It never ceases to amaze me what humans are capable of designing and building. From SpaceX rockets, autonomous cars, UAVs, everything ASML does, Artificial Intelligence, to quantum computers, CRISPR gene editing, fusion reactors, robots, and massive infrastructure projects like undersea tunnels and floating wind farms.

If we are capable of all that, surely we’re capable of creating digital engineering environments where people can collaborate and thrive without silos, cognitive load, bureaucratic drag and operational overhead.

We can create something better by becoming more people-centric, and truly understanding what our own ‘innies’ need to thrive. In the end, great engineering isn’t just about what we build - it’s about how we work together to build it. That’s the future we should be designing for.

I dedicate this article to my fellow series addict and outie wife, Nadine Ferrone ❤