Organizational Surgery in a Shrinking Market
Once the market moves from growth to saturation to contraction, companies need to rebuild their channel operating units and capability platforms rather than keep shuffling boxes on the org chart.
China’s consumer goods industry is passing through a historic inflection point.
From the growth era of the past, to the saturation era after 2013, and on to the contraction era of recent years, companies’ strategies in these three phases have pursued entirely different goals.
In the growth era, the strategic goal came down to a single word: grab. Open more outlets, enter more regions, hire more people—as long as you could stake out territory, a certain amount of organizational redundancy was tolerable. That is why so many companies recruited so many trade marketing people.
In the saturation era, many companies saw sales growth peak while total volume held steady, and some shifted their strategic goal to defense plus optimization—gradually trimming headcount, refining how the organization was managed, raising efficiency and cutting costs. But in essence they were still making minor adjustments within the existing framework. No big moves were made.
In the contraction era, however, total volume is shrinking, and strategy must progressively sharpen its focus—operating around your moats, taking market share from competitors, and finding growth within the structure.
But here a problem emerges. The strategic goals changed, yet the logic of the organizational structure did not change with them.
Over the past decade, channel structures have fragmented dramatically. Membership warehouse stores, snack discount chains, instant retail, community group buying—every new channel that emerged came with a completely different operating logic of its own. And the companies? The so-called “organizational transformations” of the past decade amounted to little more than splitting regions, merging departments, and redrawing reporting lines—essentially moving boxes back and forth on the same two-dimensional org chart.
That is “patching things up.” In the growth era, when everyone was staking out territory, a rough-edged organization didn’t matter—growth could paper over every problem. In the saturation era, profits could still hold things up, and patching was good enough. But in the contraction era, with total volume shrinking, profits under pressure, and growth no longer a given, the organization’s lag begins to turn into a fatal wound.
That is why today’s market environment actually offers a window of opportunity. After nearly a decade of dramatic fragmentation, the channel landscape is now relatively stable; essentially all the new channels that were going to appear have appeared, and the operating logic of each has largely been proven out. A stable environment is precisely the time to operate on the organization—not more patching, but reconstruction.
But reconstruct what? That is the real question.
Let’s start with a real case.
Company A had a team of over a thousand people, with far too high a share of staff-function roles and a severe shortage of frontline sales personnel. Management recognized the problem. What did they do? Split. They broke up the original regional units, trying to shorten the management chain through flattening. After splitting, they found it wasn’t working, so they merged; after merging it still wasn’t right, so they split again. Back and forth for a year and a half, the regional units shrank drastically, but performance showed no improvement whatsoever.
Why? Because they only moved the boxes on the org chart.
The boxes changed, the reporting lines changed, but the decision-making chain did not.
Take a concrete scenario. Company A wants to launch a new flavor. A regional salesperson raises the request and first reports it to the regional manager—the regional manager reviews market feasibility. Then it goes to the area director—the area director reviews the expense budget. Then to the business unit—the business unit reviews product positioning. Then to the marketing department—marketing reviews brand consistency. Then to the supply chain—supply chain reviews production scheduling feasibility. Then to finance—finance reviews the margin structure. Then to legal—legal reviews compliance. Finally to headquarters—headquarters has to convene a meeting to study it.
Every layer reviews, every layer holds veto power, and no layer holds final decision authority. From raising the request to completing approval, this new product takes two to three months if things move quickly, half a year if they don’t. By the time approval comes through, the market window has long since closed. How are expenses allocated? Still carved up by the old rules. How does information flow? Still reported up layer by layer, filtered layer by layer, until it reaches the decision-maker unrecognizable.
So restructuring an organization is, at its core, restructuring processes. If the underlying processes don’t move, moving what sits above them means nothing. Positions themselves have no value; merely reshuffling positions only looks busy.
This is the first counterintuitive truth: the core of organizational restructuring is not moving positions—it is moving processes, decision mechanisms, expense allocation, and the distribution of power.
Many companies approach organizational transformation by starting with an org chart and ending with an org chart. The diagram goes through three versions, the PPT through five, the reporting lines through eight—but the decision rights inventory, the approval nodes, the meetings and reports—the things that actually determine how efficiently an organization runs—not a single word gets touched. That’s not transformation. That’s redecorating.
The second counterintuitive point is even more surprising: compressing management layers is not the same as compressing decision layers.
Many companies assume that if they compress five layers of management into three, decisions will speed up. Not necessarily. With fewer administrative layers, the number of decision nodes may actually increase. Why? Because while the layers were compressed, the approval rights, veto rights, and rights to be informed at each layer were never redefined. Before, with five layers and one signature per layer, at least the chain was clear. Now it’s compressed to three layers, but at every layer someone says “I need to take another look at this” or “I need to double-check this expense”—and decisions end up slower than before.
Fewer boxes, but the chain didn’t get shorter. That’s the true picture of organizational transformation at many companies.
So where exactly does the problem lie?
To answer that question, we have to go back to the channels themselves.
In the past, channels were two-dimensional. What does two-dimensional mean? One region, one team, managing every retail outlet in that region. Whether it was a hypermarket, a supermarket, a convenience store, or a mom-and-pop shop, the same crew covered them all, managed with the same logic. The organization was arranged along a geographic axis, from macro-region to province to city to county—one straight line, crystal clear.
This logic was efficient in the growth era. Back then, channel types were relatively uniform, retail formats didn’t differ much, and a regional manager overseeing hundreds or thousands of outlets used roughly the same playbook regardless of store type. As long as the organization ensured broad enough coverage and enough people, growth followed.
But over the past decade, things have changed.
Membership warehouse clubs arrived, discount snack chains took off, instant retail exploded, and community group buying muscled in too. Each new channel is not simply another place to sell goods—it comes with an entirely different operating logic of its own.
Membership clubs demand large packs, low SKU counts, and high turnover; to do business with them, your supply chain rhythm, product specifications, and pricing strategy all have to change. Discount snack chains run on a private-label logic, where brand premiums don’t carry much weight and the game is extreme value for money. Instant retail requires minute-level fulfillment, so your warehousing and delivery system has to be redesigned. Community group buying plays on pre-sales plus collective procurement, with an entire chain completely different from traditional distribution.
The channels are still the same channels, but the operating logic behind them has completely changed. This is the shift from two dimensions to three—no longer spreading outlets across a flat plane, but different channel operating types stacked on top of one another, forming a three-dimensional structure.
But the organization? It’s still lined up along a single geographic axis. One macro-region manager oversees the traditional distribution people, the KA people, the e-commerce people—and often the discount snack and instant retail people too. Each channel’s operating logic is completely different, yet everyone reports to the same person, expenses are carved from the same pie, and approvals travel down the same chain.
Conflict is inevitable.
Traditional distribution demands deep distribution—boots on the ground, grinding it out store by store. Discount snack chains demand rapid response, product customization, and supply chain flexibility. These two channels place completely different resource demands on the organization, and even compete with each other for resources. The traditional channel people feel the new channels are stealing their budgets and manpower; the new channel people feel the traditional channels are dragging down the pace and hogging headcount. This isn’t anyone’s fault—it’s a structural contradiction born of organizational logic failing to keep pace with channel change.
Looking back at the evolution of channel coverage models over the past twenty years, that trajectory itself is the best evidence.
Take the mainstream approaches to distributor management: from traditional agency around 1990, to deep distribution after 2000, to the return of the large-distributor system, to distributor mini-boss contracting, to the partnership model that emerged around 2020—each iteration was, in essence, allocation rights sinking downward, from brand owner to distributor, and then to frontline business units.
But note: every one of these iterations was an optimization on a two-dimensional plane. Allocation rights were sinking, channels were getting thicker or thinner, but the organization’s fundamental logic never changed—still arranged along a geographic axis, still one macro-region managing all channels. Each iteration was “patchwork,” tuning parameters inside the existing framework without ever touching the framework itself.
Today, channels have gone three-dimensional, and the two-dimensional optimization logic has run its course. You cannot solve a three-dimensional problem by tuning parameters inside a two-dimensional organizational framework.
So what’s the way forward? The answer is not to start by drawing a new org chart, but to first get one thing clear: what is your smallest operating unit?
The smallest operating unit here is not the store, not the outlet, but the channel operating type.
What is a channel operating type? Membership clubs are one operating type, discount snack chains are another, instant retail is another, and traditional distribution is yet another. Behind each operating type stands not just one more outlet for selling goods, but a complete rearrangement of the value chain.
Why say that? Take membership clubs: they demand large packs, low SKU counts, and high turnover. That means your R&D has to redesign specifications for large packs, your production has to adjust scheduling rhythms to accommodate large batches with fewer product runs, your logistics has to support full-pallet shipping, and your pricing has to leave enough room for the retailer’s low prices. From R&D to production to logistics to pricing to the retail terminal, the entire chain has to be rearranged around this channel’s operating logic.
Or take discount snack chains: they run on a private-label logic and compete on extreme value for money. That means your brand premium doesn’t work there, and you need a product strategy built specifically for this channel—perhaps custom specifications, perhaps a standalone brand line, perhaps a flexible supply chain. That’s another, different value chain.
And instant retail? Some players use national mega-warehouse distribution, but the flash-warehouse model requires direct delivery to stores. Your warehousing and delivery system has to be redesigned—how to position forward warehouses, how to allocate inventory, how to deploy delivery capacity—yet another independent value chain logic.
So the core issue becomes clear: every channel operating type corresponds to a different value chain behind it. The starting point of organizational transformation is not drawing box diagrams, but first mapping out each of these value chains.
Mapping out what, exactly? Mapping out who along this chain is responsible for R&D, who for production scheduling, who for pricing, who for expenses, who for terminal negotiations, who for fulfillment. Then look at how these links are carved up in the current organization—which ones are tangled together, which ones are broken apart, which ones take the long way around.
This is “Step Zero”: before touching the organizational box diagram, first build the decision rights inventory. Break down the decision chain for each channel operating type, and see clearly who holds the decision rights, approval rights, veto rights, and rights to be informed at each link. Skip this step, and any org chart you draw afterward is a castle in the air.
Compare them and you’ll see exactly where the gap lies.
At Company A, launching a new product takes more than a dozen layers of approval, from the region to the business division to headquarters. Every layer has veto power, and no layer has final decision power. Why? Because no one has ever taken this decision chain apart and examined it. R&D thinks it’s not their responsibility, production thinks it’s sales’ business, sales thinks they need to ask marketing, and marketing thinks it has to be reported to headquarters. A new product approval turns into a relay race across every department—everyone is passing the baton, and no one is running.
How does Company B do it? Facing demand from a new channel, it puts R&D, packaging, production, and finance in one shared office. When a request comes in, R&D produces the solution, packaging sets the specifications, production schedules the run, finance calculates the cost—within two or three hours the sample is made, the numbers are worked out, the plan is drafted, and the decision is made on the spot.
In other words, Company B isn’t fast simply because it seats these people together. What it figured out first was: what is the decision chain for this channel operating type, and who should make the call at each link—only then did it bring the people from those links together. What got shortened wasn’t physical distance; it was decision distance.
So, the proper sequence of organizational transformation is: first identify the channel operating types, then clarify the value chain behind each type, then redefine the operating rights—who makes the call, how expenses are divided, how information flows—and only last come the operational-level items like approval nodes, decision lists, and meeting reports.
Get the sequence backwards, and everything is wasted effort. Drawing the org chart first and filling in the processes afterward is like doing the interior decorating before laying the foundation.
There’s one more approach worth mentioning. Company D set up a staff office under the general manager that only does research, analysis, and solution output—it doesn’t participate in decisions, doesn’t control expenses, doesn’t lead teams. Why? Because when frontline business units draft proposals, they naturally write them in ways that favor themselves, so by the time information reaches the decision-maker it’s already distorted. The staff office’s role is to provide “clean input”—unentangled with frontline interests, responsible only for laying market facts and viable options in front of the decision-maker.
This isn’t a new concept, but it matters especially today. Once channels become multi-dimensional, the complexity of information facing decision-makers is several times what it used to be. Without an information hub that stands above frontline interests, every piece of information the decision-maker receives comes filtered through departmental interests, and decision quality cannot possibly be high.
Having covered the direction, let’s look at how different companies actually do it. There is no standard answer here—only solutions matched to one’s own channel structure.
Company C: Using Organizational Creativity to Solve the “People Problem”
The thorniest part of organizational transformation is often not process, but people. Senior veteran leaders with high rank and heavy influence—you can’t move them and can’t use them well. Company C’s solution was to establish a new products business division and put seasoned veteran employees in charge of new product promotion there. This wasn’t a retirement home; it was giving them a different track to compete on. The result: they produced a blockbuster food product, and in the e-commerce segment they actually became the most profitable department.
The lesson of this case isn’t the innovation business division itself, but the logic it reveals: don’t try to force-solve people problems with HR measures—solve them with organizational creativity. Switch the track, and the experience and resources of veteran employees may be exactly what a new product needs most.
Different Brands, Different Solutions
Compare several typical organizational models in the industry, and you’ll find that different channel structures lead to completely different organizational solutions. Note: the difference between them is not in quality but in the channel structures they match.
One bottled water brand takes the decentralized approach. Provincial-level amoebas, with distributors and sales reps forming profit-accounting units, lump-sum expense budgets, and bare-price supply; headquarters keeps only an inspection team reporting directly to the general manager. Why decentralize? Because the channels are mature, terminals are dense, and per-bottle margins are thin—operating rights must be pushed to the very front line, and money is made on scale and efficiency. Headquarters doesn’t make decisions for the front line; it does only two things: set the rules and investigate violations.
Another brand competing with it takes the exact opposite path—digitalized centralization. A large middle-platform digital system provides enablement, headquarters coordinates, regions operate, with internet-style high-frequency iteration. Why centralize? Because it started as a new brand in new channels, with fast product iteration—it must rely on a data hub for unified dispatch, and only centralized decision-making at headquarters can guarantee agility. The front line is not the decision-maker; it is the execution terminal.
These two models are nearly polar opposites, yet both are right. Because the channel structures they face are completely different. There are also companies taking the traditional centralized approach—central authority plus regional division of labor, headquarters negotiating with national key accounts, large regions supervising execution, delivering unified output through brand momentum and management standards. And there are companies that adapt by market maturity—decentralize in growth markets, delegate in mature markets, control in dominant markets: one organization, three playbooks.
Among these four models, none is right or wrong. In essence there is no good or bad, right or wrong—only match or mismatch. Whatever your channel structure looks like, that is what your organization should look like.
External Reference: How the Big Brands Do It
Procter & Gamble’s approach is called Focus Markets—thick resource allocation in key markets, lean operations in ordinary markets. Not every market follows the same template; instead, organizational depth is matched to market maturity and strategic value.
Coca-Cola takes the franchising-plus-headquarters-enablement model—headquarters builds a shared services platform providing data analytics, supply chain coordination, brand advertising, and other services, but stays out of the decision chain. Frontline operating rights sit with the bottling plants; headquarters is an enabler, not a commander.
What do these two cases have in common? Both first think through “where the operating rights should sit,” and only then decide “how the organization should be built.” They don’t draw the org chart first and divide up authority afterward—they settle the operating rights first, then draw the chart. Get the sequence right, and the organization falls into place naturally; get it wrong, and no amount of adjustment will fix the awkwardness.
Three Market Facts
Finally, coming back to the present, there are three market facts that are the premise of all organizational transformation:
First, growth has shifted from dividing up total volume to fighting over structure. The overall market is no longer growing—every extra bite you take is one bite less for your competitor. In this environment, an organization’s speed of response matters more than its size.
Second, every channel carries its own operating logic; you cannot force one organization onto all channels. The differences between channels are not differences in terminal format—they are differences in value chain logic.
Third, the frequency of market fluctuation now exceeds the frequency of organizational decision-making. By the time your layers of approval are complete, the window of opportunity has already closed. The core goal of organizational transformation is not to make the organization bigger or more complete, but to make the decision chain keep pace with the market’s rhythm.
So back to the question at the beginning: restructure what? Not the org chart—the operating rights. Not redrawing the boxes, but redefining the decision rights, expense authority, and information rights along the value chain behind each channel operating type. The org chart is the last diagram you draw, not the first.
Finally, here is a test. To judge whether your organization needs to change, and what to change, ask just one question: Does each of your channel operating types have its own independent decision chain? If it does, the organization is following the channels; if it doesn’t, the organization is dragging the channels along.
In the era of growth, it was enough for the organization to follow scale. In the era of contraction, the organization must follow the channels. When the strategic goal changes, the organizational logic must change with it. This is not a multiple-choice question—it is a question of survival.