They Moved Up
Every time a machine took over the repetitive work, the people who did it rose to better work. Operations is reaching that moment now.
By Kelly Breakstone Roth, Co-Founder & CEO of Prysmic · June 2026 · 5 min read
Dan Bricklin and Bob Frankston, creators of VisiCalc, the first spreadsheet (1981).
In 1979, two engineers named Dan Bricklin and Bob Frankston shipped a small program called VisiCalc. It was the first spreadsheet, a grid of cells that recalculated itself the moment you changed a number. For anyone who had ever modeled a business by hand, on ledger paper, in pencil, erasing and recopying a full column because one figure moved, it was close to magic.
The reaction was not all excitement. As the spreadsheet spread into every office through the 1980s, a worry traveled with it: that it would put the people who did this work out of it. Why keep a room full of bookkeepers when a machine could do the arithmetic in an instant.
That worry missed the bigger story. The clerical work did shrink. Since around 1980, the United States lost roughly 400,000 bookkeeping and accounting-clerk jobs. Over the same stretch it added about 600,000 jobs for accountants.
The arithmetic got cheap, so the appetite for everything above the arithmetic, the analysis and the judgment and the advice, grew. The people did not vanish. They moved up.
The spreadsheet did not empty the profession. It opened it.
Supply chain operations is now standing roughly where those clerks stood. The work that has defined the job, the chasing and reconciling and stitching of systems together, is becoming something software can carry on its own. And it brings back the same worry that followed the spreadsheet into every office.
The question under every conversation
The first thing people ask me when they hear what we build, an autonomous workforce for supply chain operations, is whether we're helping companies cut jobs. It's a fair question. We are building toward a future where people direct and decide above a fleet of agents, instead of spending their days inside the operation themselves. That future will reshape the work people do, and pretending otherwise would be dishonest.
The worry is widespread. BCG's research on AI in supply chains finds that the hardest barrier to adoption is not the technology, it's people, with social resistance and labor concerns named by nearly three quarters of companies as something holding it back. It deserves a real answer.
When operators actually talk to us, though, the conversation is rarely about headcount at all. It is about growth.
They want to know how AI helps them move faster, scale without piling on drag, and get more out of the operation they already run. The question underneath is always the same: how do we keep growing without the operation becoming the thing that holds us back.
Out from under the work
For decades, running an operation has meant running it by hand. A skilled person reads the state of things across a dozen systems, decides what matters, moves between tools, and takes the next action, again and again, all day. When that is the shape of the job, the day fills with it, and the work that actually moves the business keeps waiting.
The deeper cost is structural. In most operations, almost nothing moves until a person touches it. A shipment slips and sits until someone catches it. A reorder waits on an approval. An exception holds until there is a free moment to think it through. None of this is a failure of the team, it is the design of the work: every decision is routed through human attention, and no team, however sharp, can be in a thousand places at once. So the operation moves only as fast as people can get to each thing, and across thousands of daily decisions that adds up to a business running a step behind its own potential.
Agents change that the moment they take the routine work. They handle each decision as it arrives, in parallel, around the clock, so nothing waits in line behind a single person. The shipment reroutes itself, the reorder goes out, the exception gets resolved or escalated while it still matters. The operation stops waiting on anyone.
The people are not pushed out of it. They are pulled up, out from under it. Instead of being the point everything funnels through, they become the ones deciding where it should go.
The work that reaches them is the work that only a person can do: the judgment calls, the supplier relationships, the tradeoffs that decide where margin comes from, the question of what the business should do next. They go from running the operation to directing it, with a team of agents doing the running. That is harder and more strategic work, and it is far more satisfying. Nobody got into operations to reconcile spreadsheets and chase status updates. That work goes to the agents, and what comes back is the reason they took the job, the part they were excited about in the first place.
This is where the whole industry is heading, not just where we are pointing it. BCG projects that the way supply chain teams spend their time will flip over the next decade, from roughly 70% spent running the operation today to a majority spent designing and improving it, the strategic work, by the mid-2030s. The machine ends up running itself, and the people end up shaping where it goes.
Ahead of growth
There is a second effect, and it changes the math of scaling itself. Ask anyone who has grown a brand past eight figures what made it hard, and much of it comes down to one problem: you have to build operational capacity ahead of the growth that pays for it. The team, the systems, the processes all have to be in place before the volume arrives, which means committing capital and hiring against demand you do not have yet. Build too slowly and the operation buckles under the growth. Build too far ahead and you carry weight the revenue has not caught up to. Either way, the operation is the part that makes scaling risky.
That is the bet that is changing. When agents carry the operational load, capability scales with the business instead of being built ahead of it. A brand can take on the next tier of volume without first standing up the entire apparatus to support it.
The same shift that lets an established brand scale without that drag lets a smaller one compete that never could before. The capability that used to live behind the walls of the largest companies is becoming something a focused team can wield, and fewer good ideas die for lack of an operational machine behind them. More people get to build something serious.
The operation used to be the price of admission. That price is falling.
The bigger story
It is easy to look at all of this and see only what is being taken away. That instinct is familiar, and it has been wrong every time. The grind of running the operation goes to the agents, and the people do what the bookkeepers did before them. They move up. The future of the job is larger than its past, and it's much more human.