Start with a calendar. Not a strategy document, not an operating model deck. The calendar of one engineering manager in one capability center, on an ordinary Tuesday. Nine one-on-ones this week, because she has fourteen direct reports and believes, correctly, that people deserve time. Four delivery reviews. Two hiring panels, because the team is growing again. A compliance training she will do at night. Somewhere in the white space between meetings, the actual thinking her role exists for.
Nobody designed this calendar. That is the point. It is what a role looks like when an organization doubles around it and the role stays the same size. Her span was five people when the team was founded. It became fourteen one requisition at a time, each one individually sensible, none of them accompanied by the question of whether the shape still worked. She is not failing. She is carrying a load the design was never meant to place on one person, and she is carrying it well enough that nobody has to look at the design.
This is the first surface where a center's strain shows, and the most visible once you know to look. Every center begins with a shape that fits it. Then headcount doubles, scope widens, and the shape does not grow on its own. It gets stretched instead of redrawn.
There is a reason the stretching happens so quietly, and it is worth understanding with some sympathy, because the people involved are behaving rationally at every step.
An organization's shape is a set of fixed decisions: how many layers, how wide the spans, which functions are separate, where the seams run. Headcount, by contrast, is a flow. Work arrives continuously, and the natural response to more work is more people, because a requisition is a Tuesday decision and a redesign is a board conversation. So growth gets absorbed the cheap way, person by person, into a shape drawn for a different volume. The layers stay where they were. The spans widen. Teams that should have split into two stay as one large one, because splitting would require choosing a second leader, and choosing is hard. Roles get invented around whoever is available rather than around the work, because the work cannot wait.
Meanwhile the supporting shape lags even further behind. HR, finance, legal, procurement, the functions that were sized for the founding center, inherit the full weight of the grown one. A recruiting engine built for two hundred hires a year is asked for six hundred. A compliance framework written for one entity and one building meets three entities, two cities, and a hybrid workforce. These functions do not fail loudly. They fall behind quietly, and the line organization experiences the lag as friction it cannot name: offers that take too long, policies that fit last year, questions that have no owner.
The pace makes all of this sharper than it has ever been. India's capability center ecosystem added nearly half a million people in two years, growing from about 1.9 million in FY2024 to 2.36 million in FY2026, by Nasscom and Zinnov's count. Zinnov's mid-market report found that 35 percent of India's mid-market centers were set up in just the last two years, and the cohort of mega centers, those above five thousand people, is projected to grow from 88 today to more than 230 by 2030. The FY2026 landscape report describes what it calls the collapse of the maturity timeline: centers now reach in a few years the scale and scope that once took a decade. Every one of those compressed journeys is a shape being outrun in fast motion.
Inside a center, a stretched shape rarely announces itself in the language of organization design. It shows up as three ordinary, deniable facts of daily life.
One manager, fourteen direct reports, and a calendar that proves it.
Every surge in work answered with a requisition, never with a redesign.
HR, Finance, and Legal hearing about decisions after they are made.
Each of these has a respectable cover story. Wide spans get called flat and empowered. Requisition-first growth gets called responsiveness. Support functions running behind gets called business focus. The cover stories are not lies, exactly. They are the compliments a center pays itself for coping.
The stretching also has a geography. It concentrates at the middle. The top of a center gets designed deliberately, because site leaders and their staff are visible to the enterprise, and the bottom gets refreshed constantly, because hiring never stops. The middle, the layer of managers and senior managers who translate one into the other, is where the founding shape quietly persists. It is the layer that was thinnest at founding, because a small center needs almost none of it, and it is the layer that growth punishes most, because every hundred new hires need three or four new people who can lead, and leaders cannot be requisitioned the way engineers can. Centers that grew fast almost always carry a middle that is one size too small, promoted one year too late, and stretched across teams that should have been split two reorganizations ago.
I learned to distrust the compliments during the years I spent inside one of these centers, watching it grow past two thousand people. The tell was never the org chart, which always looked reasonable on a slide. The tell was the reorganization rhythm. When a shape fits, reorganizations are rare and boring. When a shape is being stretched, reorganizations become a season: every eighteen months or so, boxes move, dotted lines appear and disappear, and everyone loses a quarter to re-forming teams. A center that reorganizes constantly is not restless. It is trying to solve with motion a problem that can only be solved with design, because each reorganization moves people around inside the same undersized architecture.
The costs are real, and they compound, but they land in budget lines nobody connects to structure.
The first cost is managerial. A manager with fourteen reports is not managing; she is triaging. Coaching disappears first, then career conversations, then the early detection of problems, which is most of what management is for. The people under her experience this as absence, and some of them leave over it, and the exit interview will say career growth, and the retention program will offer a course. The structural fact, that no human being can develop fourteen people at once, never enters the record.
The second cost is decision speed at the working level. A stretched shape has too few points where a call can be made, so small decisions queue behind the same overloaded people. This is different from the authority problem this series turns to next, where decisions travel upward out of the center entirely. This is congestion inside the center, and teams learn to route around it, which produces the shadow structures every stretched organization grows: the informal network of who actually gets things done, running quietly alongside the chart of who officially should.
The third cost is the cruelest, because it looks like a virtue. A stretched shape selects for heroes. The people who thrive in it are the ones willing to absorb what the design cannot hold, and the center comes to depend on exactly those people, which is how a structure problem matures into the continuity problem that closes this series. The strain does not stay in the shape. It moves into the bodies of the people holding it.
There is a fourth cost, harder to see because it is an absence: the work that never happens. A shape with no slack has no capacity for the redesign that would create slack, which is the trap in its purest form. Improving how work is done requires exactly the kind of protected, thinking time that a stretched shape consumes first. So the center runs harder every year at the old way of working, and the gap between what it does and how it is built widens on a schedule nobody set.
None of this happened because anyone built badly. The founding shape was right for the founding center; that is why it worked. It was kept past the size it was designed for because keeping it was free every single day, and redrawing it was expensive on any given day. Organizations, like people, defer exactly the maintenance that is invisible until it is urgent.
The honest question for a leadership team is therefore not whether the shape was ever right, but whether it fits the size the center is now, and whether it has room built in for the size the center will be in three years. A shape with room in it looks specific: spans a person can actually carry, a management layer that exists before it is desperately needed, support functions sized to the center's future rather than its past, and a standing habit of redesigning how work is done when work grows, instead of only adding people to the old way.
It is worth being honest about why the redraw gets deferred even after everyone can see the strain, because the reasons are human before they are financial. A redesign names winners and losers, and a growing center would rather keep everyone winning. A redesign requires the founding leaders to acknowledge that the thing they drew is finished, which lands close to acknowledging that an era is finished. And a redesign has no natural sponsor: the global function heads see only their slice, the site leader is measured on delivery this quarter, and the enterprise architecture team, where one exists, was built for systems, never for centers. So the shape persists by default, which is the most durable way anything persists.
I have also seen the other side, and it is worth describing, because a shape that has been genuinely redrawn feels different within a quarter. Managers reappear in their teams' lives, because a span of six or seven leaves room for actual management. The reorganization season ends, replaced by small, boring, continuous adjustments, which is what design maintenance looks like when it exists. Support functions stop being the place where things slow down, because someone finally sized them to the center's future. And the heroes get quieter, not because they matter less but because the design has started carrying what they were carrying. None of this is exotic. All of it is a decision.
There is a simple test I offer leaders who want to know where their center stands. Take your five best managers and look at their calendars, the real ones, for two weeks. If what you find is triage, the shape is speaking. It has been speaking for a while. The calendar was just the first place it could get anyone to listen.
Nasscom & Zinnov, GCC Value Orbit: From Delivery Engine to Enterprise Nerve Centre, India GCC Landscape Report FY2026, July 2026. zinnov.com
Nasscom & Zinnov, India GCC Landscape Report: The 5-Year Journey, 2025. nasscom.in
Zinnov & Nasscom, Mid-market Global Capability Centers Report, 2025. zinnov.com
Zinnov, The 5 Shifts That Defined India's GCC Story in 2025, 2026. zinnov.com