The re-leveling review always has a satisfied feeling to it. A room, a spreadsheet with names against grades, a set of decisions about who moves up a band this cycle and who holds. Someone is confirmed at the next tier, the paperwork clears, the system updates, and a certificate or a badge or a title change makes it official. Everyone treats it as a small act of progress, which is how it is described in the room, this person has grown into the harder work. And in a narrow sense that is true. The person did get better. The problem sits one level up from the person, in a fact nobody at the table is looking at. By the time the promotion clears, the work that used to define that tier has already stopped being the harder work. It has become the entry expectation. The floor moved while the paperwork was in flight, and the organisation just certified someone into a level that no longer sits where the certificate says it does.
This is the quiet mechanics of almost every talent system right now, and it is worth naming precisely because everyone is living inside it and calling it something else. The work is re-tiering continuously. What was advanced last year is routine this year, because the tools underneath it absorbed the routine part and pushed the whole ladder up a rung. What was a specialist judgment is now a baseline expectation, assumed rather than rewarded. And the machine the organisation uses to re-tier its people, the banding review, the competency refresh, the certification cycle, runs once a year, maybe twice. A streaming problem is being managed by a batch process. The gap between where the work actually sits and where the people are formally banded does not stay constant. It widens every quarter the two clocks run at different speeds.
A Tier Was Always A Snapshot.
It helps to be honest about what a tier ever was, because the honest answer explains why the whole apparatus is now quietly failing. A tier was a snapshot. It froze a moving thing, the actual difficulty and value of a body of work, into a fixed grade so the organisation could pay it, staff it, and plan around it. Tier 1 was the entry work, Tier 2 the work that took a few years to be trusted with, Tier 3 the work only the deep ones did. That snapshot was useful for exactly as long as the thing it photographed held still. And for a long time it did. The difficulty of the work drifted slowly enough that a grade set this year was still roughly accurate three years on, which is why nobody questioned building an entire career architecture on top of a photograph.
The tools broke the stillness, and they broke it without asking permission. The moment the routine core of any tier could be handed to a machine or a copilot or a workflow that did not exist eighteen months ago, the work left in human hands shifted upward. The person doing Tier 2 work is now doing what was Tier 3 judgment two years ago, because everything below that got absorbed and the expectation rose to fill the space. Nobody re-photographed the tier. The grade on the org chart still says Tier 2, the pay band still says Tier 2, the competency framework still describes Tier 2 as it was defined when it was written. But the actual work wearing that label is a level harder than the label admits, and the person doing it is being measured, paid, and developed against a description that expired sometime last spring.
So the organisation is running a live, continuously rising ladder through a set of grades that were fixed to a ladder that stopped existing. It has a talent model that can see the label and cannot see the drift, because the label is what it was built to store and the drift is a thing it has no field for. Ask that model whether a person is current and it will check their tier, their last certification, their completed learning path, and report back that they are exactly where they should be. It will not tell you that where they should be moved six months ago, because the model does not know the floor rose. It is reading a photograph and reporting it as a live feed.
The Treadmill Nobody Names.
Watch what this does to a career and the cruelty of it becomes clear, though it is nobody's cruelty in particular. A person is developed, over months, toward the next tier. They learn the thing, they demonstrate it, they are certified into it. And in the time that took, the thing they learned became the baseline that the tier below them is now expected to have. They did not climb. They kept pace with a floor that was rising underneath them, and the organisation, unable to see the floor move, recorded their standing still as an advance. This is the treadmill, and it is the defining experience of competent people in a re-tiering economy. They run hard, they genuinely improve, and they arrive at roughly the same relative position they started from, because the whole surface was moving the entire time.
The organisational cost is worse than the personal one, because the organisation cannot even see it to grieve it. A function looks fully staffed and fully certified against its own framework, every seat filled by someone at the appropriate tier, every tier signed off. And it is quietly a level behind the work, top to bottom, because the framework it certified everyone against was a description of last year's difficulty. The people are not underperforming. They are performing exactly to a specification that no longer matches the job. The organisation mistakes a re-leveling exercise, moving names between grades, for capability building, closing the distance between people and the rising work, and these are not the same act. One updates a photograph. The other has to run at the speed of the thing being photographed, and almost no talent system is built to run at that speed.
Then there is the cost that surfaces late, when it is expensive. The gap between the banded tier and the real tier is a debt, and it compounds silently until a moment of genuine difficulty calls it in, a crisis, a new mandate, a piece of work that needs the Tier 3 judgment the org chart swears it has three of. And the organisation reaches for its certified people and finds they were certified into a definition that drifted, that the label promised a level the work outgrew, that the three Tier 3s on the chart are doing what Tier 3 meant when the chart was drawn and not what it means now. The bench was never as deep as the grades claimed. It only looked deep because the measuring stick had stopped growing while the work kept going.
The Reframe The Org Keeps Missing.
Here is the part that is hard to sit with. The organisation does not have a training problem, and it does not have a talent shortage, which are the two diagnoses it reaches for because they are the two it knows how to fund. It has a clock problem. Its people are learning, often faster than they are given credit for. The work is simply re-tiering faster than the system that re-tiers the people, and no amount of additional content poured into an annual cycle closes a gap that is being reopened continuously by the pace of the work itself. The thing being measured, a person's position against a fixed grade, is the wrong thing. The right thing is the distance between how fast the person is climbing and how fast the floor beneath them is rising, and that number lives in no framework, gets reviewed in no cycle, and is owned by no one.
Rebuilding this is not a slogan about upskilling faster, which is where most organisations stop and why most of them stay on the treadmill. It means re-instrumenting the talent architecture around trajectory instead of position, measuring people by the rate at which they close on rising work rather than the tier they were parked at last review. It means defining a tier by the judgment it currently demands and re-photographing it on the cadence the work actually moves, not the cadence the HR calendar prefers. It means treating the delta between real work and banded grade as the core metric of a function's health, the thing you watch weekly, because a function whose delta is widening is going obsolete in place no matter how green its certification dashboard looks. And it means building people toward the floor that is coming, not the floor that is here, because by the time you have developed them to today's tier, today's tier is already the entry expectation.
This is the work SSUNDAR does, and it starts before anyone opens the banding spreadsheet. We treat the tier structure as a moving system to be tracked rather than a grade ladder to be maintained, because in nearly every organisation we see, the framework was competent when it was written and has been silently drifting from the work ever since. We measure the distance between where the work now sits and where the people are formally held, so the organisation can finally see the debt it has been accumulating and stop mistaking a full certification dashboard for a current one. It is harder than running the annual review. What it produces is a talent architecture that moves at the speed of the work, instead of one that keeps photographing a floor that has already risen and filing the picture as the truth.
The work re-tiers continuously and the system that re-tiers the people runs once a year. Every certificate you print is an accurate photograph of a floor that has already moved, and the gap between the picture and the room widens every quarter you keep filing it as the truth.
You did not promote them into the harder work. The harder work became the baseline while the paperwork was in flight, and you certified them into a tier that had already become the floor. Tier 2 is the Tier 1 of tomorrow, and your talent system is still congratulating people for arriving at yesterday.