“The Leaders Haven’t Come Yet — Don’t Start Selling”

Before We Begin

When I wrote the first draft in October 2024, every line was thick with emotion. I did not understand how the company operated. I was angry, I was resentful. By the time I came to revise it this year, I had been through my own “baptism” at another state-owned enterprise, and looking back, those incidents feel almost routine. The one thing that has not changed is that “playing house” feeling.

I had reservations. I was worried that telling everything exactly as it happened would bring trouble, so I have changed the names of people, places, brands, and the e-commerce mini-program. The story is real; I am showing only one cross-section of it, to share what I can.

“Qingzhou” is a pseudonym. The agricultural brand names that appear in this article are pseudonyms as well.

In August 2022, I joined an agricultural products company in the Qingzhou Development Zone, working in marketing. The marketing department had only just been set up. The company — a state-owned enterprise — had been around for over a decade and had gone through a period of private capital entry. When I joined, it was in the midst of its mixed-ownership reform. Our entire job was running a mini-program called “Qingzhou Taste.” At the time, the company was not truly launching a full-scale e-commerce push. It was simply operating the project on behalf of the Qingzhou District Agricultural Committee. Do a good job with the mini-program, and the company would collect government subsidies.

This business logic — neither truly market-oriented nor genuinely aimed at public service — set the stage for all the confusion that followed.

◉ Qingzhou in daylight and at dusk.

The Qingzhou Development Zone sits along the lower reaches of the Yangtze River. In look and feel, it seems little different from the other small riverside towns in the Jiangsu-Zhejiang region. The land is wide and the soil rich, giving it a natural advantage in agriculture. Yet as a development zone under a city better known for its economic output, Qingzhou’s GDP over the ten years since it was upgraded from a county to a district has been unimpressive. The district government has turned to developing “specialties” — building an online trading platform to extend the national recognition of local agricultural products — which, under the internet economy, counts as a fresh initiative.

On the surface, it sounded like a good job, and I was excited. I had worked as an editor at a food video media outlet, filming cooking tutorials and short food documentaries, travelling to all sorts of places across the country to document local specialities. I had also done product promotion for a chain food brand. This time, I thought I could get to the source of agricultural products — actually delivering on the “from soil to table” concept and giving farmers some tangible help. On top of that, the mini-program had official endorsement behind it, so I figured there would be plenty of extra support and the work would be straightforward.

But a few months into the job, my enthusiasm was extinguished. I gradually realised that this government-driven mini-program operated on no purely e-commerce logic at all. It was tangled up in government directives, the rules of personal-connection culture, and shareholding manoeuvres. How many agricultural products could actually be sold was, in truth, the least important objective.

For the two years that followed, I felt as though I were playing house, until I was eventually forced to leave. Shortly after, the mini-program “disappeared” too.

Where Do the Goods Come From?

After I joined, the first major task in the marketing department was moving the existing group-purchasing business online.

I had assumed that running e-commerce meant dealing directly with ordinary consumers — planning marketing campaigns, producing self-media content to drive traffic, and managing operations within the mini-program. But the company also assigned a senior salesperson, Lao Pu, to our marketing department. Lao Pu had been doing group-purchasing business at the company for over a decade.

Lao Pu’s group-purchasing experience was primarily aimed at enterprises and public institutions. Our company had run a to-B business for years: bulk-buying agricultural products, then wholesaling a portion to department stores and supermarkets and running welfare group-purchase schemes for enterprises and public institutions. Every time a festival came around, these organisations would publish their welfare budgets in advance, and if we won the bid, we would source local agricultural products, transport them, and distribute them to their staff. This kind of institutional festival-welfare business operates on a completely different logic from the fully competitive retail market outside.

Lao Pu was in his early forties. When he came, he did not take part in any online work at all. He simply added two extra steps to the old institutional group-purchase process: listing the agreed product bundles on the platform, and printing paper vouchers to send to partner organisations. Ironically, it was us who had to carry out those two steps on his behalf. And if a partner organisation’s staff ran into problems redeeming their vouchers, we also had to act as customer service and answer their queries.

Lao Pu was a loner. He hardly ever stayed in the office, always saying he was out running business. I never really got close to him. At first I did not understand why they had assigned him to the marketing department, until, after some time, I noticed that our daily revenue was only a few dozen to a few hundred yuan, but before holidays it would spike to four or five thousand. That is when it hit me: our results were almost entirely propped up by Lao Pu’s group purchases. If we relied solely on selling to individual consumers, total revenue would be embarrassingly low, and we would certainly fail to deliver results to management by year-end.

◉ A Chinese mitten crab co-operative in Qingzhou.

The second major task in the marketing department was building a regional public brand for agricultural products.

Before our mini-program went live, the Qingzhou government had previously promoted a public agricultural brand called Fengyue Haichuan (pseudonym). I heard they had originally wanted to register the name “Qingzhou” directly, but since trademark law prohibits the direct use of place names as trademarks, they decomposed the characters by their visual components, and that is how Fengyue Haichuan came about. Many local farmers had used the brand’s packaging boxes, and local media had run quite a few stories about Fengyue Haichuan. But when the new district leadership took office, they came up with a new brand name — Qinghe Three Brothers — and declared that everything would start fresh, with Qinghe Three Brothers fully replacing Fengyue Haichuan.

The Qinghe Three Brothers logo was designed by our marketing department. The part-time graphic designer spent half a day producing an icon. There was no brand story, no colour guidelines — and the District Agricultural Committee simply approved it as was. By February of the following year, the name had already appeared in the government work report: “Improve the public agricultural brand system, launch the unified Qinghe Three Brothers identifier, and build the Qingzhou Taste platform.”

The new brand made its official debut. On the surface, it felt like I was part of some momentous event — with official backing behind it, things should have been sailing smoothly. Yet during the first six months after I joined, the most effort I spent was going all over the place looking for suppliers and convincing them to sell on our mini-program, and no one seemed to give this official platform the respect it deserved.

◉ White-cut chicken made with locally free-range chickens, served at a Qingzhou restaurant.

It was only after joining that I gradually learned how many varieties of local specialty agricultural products Qingzhou actually had. At Jiangsu-Zhejiang restaurants, cold starters commonly included Qingzhou golden melon shreds and Qingzhou mutton. Qingzhou rice was well regarded, and the Qingzhou rice wine made from it was a regular sight on restaurant menus. Chinese mitten crabs, chicken, duck, pigeon and other poultry, all manner of vegetables and fruit — Qingzhou produced them all. Even saffron, which I had assumed was only grown in Tibet, had Qingzhou as one of its production areas.

With so many categories, there were plenty of individual growers and breeders. Local smallholders typically produced and sold their own goods. Where individual output was too small for a partnership, we approached co-operatives and enterprises that had already reached a certain scale. Some of these had built up their own reputation in the city — Chunhe-brand pork and Jinmu-brand mutton (pseudonyms), for example. If they would come on board, their presence alone would boost our promotion.

The general manager and Lao Pu were local to Qingzhou, with wide connections. They knew many suppliers and helped us set up WeChat groups one by one. We put together an information-collection form and invited partners to provide their business licence, product descriptions, wholesale prices, and suggested retail prices.

◉ Orchards in the Qingzhou countryside often also keep a few poultry.

The form would go into the group and often draw no response at all. The operations team had to follow up every few days just to get back a “half-empty” form, and then another two or three rounds of asking before they could barely scrape together complete information. Some suppliers simply called: “I’ll just tell you the prices, write them down, all right? I don’t have photos — just download them from the internet, they’re all pretty much the same.”

Sluggishness was the least of our problems; what was worse was the perfunctoriness. Hengda Stainless Steel (pseudonym) was well known locally and treated us with politeness and friendliness, responding quickly. I guessed this was largely because they manufactured for top international brands and had experience dealing with outsiders. What they wanted to list this time was their in-house brand, Youchu, a range of stainless-steel cookware. But the prices they quoted looked suspicious. I checked their Tmall flagship store to see sales figures, and found that for several items, the cost price they had given us was actually higher than the retail price on the flagship store. We pointed this out in the group, and they promptly resubmitted more reasonable prices. The reason they had originally thrown out random wholesale prices was simply that they had already worked out our platform would not bring in much volume.

Progress on onboarding partners was painfully slow. The District Agricultural Committee organised a dedicated mobilisation meeting to ask each enterprise and co-operative whether they were willing to join Qingzhou Taste. Our company sent Lao Pu. When he came back, his report was: “Everyone raised their hand, of course, but whether they’ll actually come on board — that’s still up in the air.”

Chunhe pork was a major local brand. I had seen its dedicated pork counters at the wet market and in supermarkets. In earlier group chats, they had pointed out that if it was just scattered individual orders, each one requiring separate courier dispatch, the cost did not work. They never explicitly refused, but they also never filled in the form. After the Agricultural Committee’s mobilisation meeting, the operations team thought things would finally move forward. They reached out again, only to find nothing had changed — both sides were still at a standstill.

After a while more, with no one giving any extra push, Chunhe suddenly one day filled out the form on their own initiative and sent it back to us. Two or three other pork brands also signed up enthusiastically.

We later worked out that this was not the result of industry competition. Word had been passed down from above: if Chunhe did not join Qingzhou Taste, then the next time official promotion introduced Qingzhou’s representative pork brands, Chunhe would no longer be mentioned.

A directive like this would naturally leave no written trace. The “invisible hand” delivered the message in perfect silence. The companies chose to join not because they believed in the e-commerce platform, but simply to pre-empt whatever the authorities might do next.

Our e-commerce platform became, in effect, a beneficiary of this semi-coercive mechanism. Over the course of six months, some thirty to forty co-operatives and enterprises gradually joined Qingzhou Taste. The number of products on sale exceeded a hundred, so you no longer scrolled two or three screens on your phone and reached the bottom. The mini-program finally started to look like a proper e-commerce platform.

Striving for Legitimacy

As the platform’s product range gradually expanded, we received a fresh directive: remove all traces of the Fengyue Haichuan logo from the pages.

This was a legacy issue. Fengyue Haichuan had been the regional public brand for several years before. The District Agricultural Committee had designed packaging for a handful of pillar categories — vegetables, white goats, Chinese mitten crabs, Cuiguan pears, rice — and in many of the product photos suppliers submitted, the Fengyue Haichuan logo was still clearly visible. The operations team had no choice but to erase the branding from each image, one by one.

But physical packaging was hard to eradicate. Too many boxes had been printed before, and no enterprise was willing to simply toss them aside. We could still catch glimpses of Fengyue Haichuan in the occasional consumer unboxing photo.

We would politely remind suppliers to try not to use the old packaging any longer, but it stopped at the reminder: Everyone was willing to turn a blind eye — nobody knew whether, a few years down the line, after the next government turnover, policy might shift again. Who could say the Qinghe Three Brothers being pushed so hard today would not itself become the next legacy problem.

With no proper brand guidelines in place, we at least tried to make our marketing strategy look the part.

Our department head, Sister Kong, was a former colleague of mine. She was the one who had invited me to join the company. A Qingzhou local herself, she had met the company’s general manager — also from Qingzhou — in an MBA class, just as he was about to set up the marketing department. It was the general manager’s personal connections, after all, that had earned the company the chance to become the District Agricultural Committee’s operator, and set in train the whole chain of events that followed.

◉ The operations colleague and I photographing products in the park outside the office.

Sister Kong had spent over a decade in marketing and carried with her a methodology she had picked up at big companies. She believed that any promotional push first needed a “marketing calendar,” dividing the year into distinct promotional periods. The 618 sale, Double 11, the New Year Shopping Festival — these were the major sales events virtually every industry participated in. On top of that, you had to “invent festivals” tailored to your own product characteristics, so that the mini-program would have different campaigns rolling out one after another all year round.

So we started with an annual plan, then broke it down into a sprawling period-by-period marketing spreadsheet. For each period, the sheet had to clearly list the theme, start and end dates, featured products, marketing tactics, promotional mechanics, member benefits, communication plan, and performance targets.

In my previous role I had handled only the communications side — once the marketing team had finished their planning, I would match content and external placement to the brief. But here, planning, communications, and execution were all on me.

I dug out the period-marketing spreadsheet Sister Kong and I had used at our previous company and, working off its broad structure, began laying the groundwork for at least the first half of the year.

Like any e-commerce platform, we set up spend-threshold discounts, limited-time deals, flash sales, and a five-yuan no-minimum-spend coupon for new registrations. Then, following the agricultural seasons, we contrived sale after “sale.” Almost every period saw a fresh marketing theme conjured on the mini-program.

The problem was, the results of all these campaigns remained modest.

A large part of the problem lay in communications. In name we were a marketing department, but we had no discretionary budget to speak of. Forget paid advertising — even the most basic product swaps, asking a blogger to post about us on their WeChat Moments or on Xiaohongshu, were off the table.

The only channels we actually had to work with were the Qingzhou Taste Official Account (WeChat) and its Video Account. The Official Account had been inherited from the Fengyue Haichuan era, with around twenty thousand followers, yet each article barely scraped past five or six hundred reads. Later, when Sister Kong shared articles on her own WeChat Moments, reads would bump up by a hundred or two — and honestly, that may have been our most effective promotional tool.

It was not until April 2023 that I finally got my first offline event with an actual budget.

It was a “Benefits for the Community” market organised by the District Agricultural Committee. Besides setting up a stall, we were also expected to broadcast in real time via the Qingzhou Taste Video Account.

◉ Sister Kong doing the livestream; I am the floor manager.

We had done livestreams before, right in the office — Sister Kong as the host, the rest of us as floor managers, with only a dozen or so viewers online at a time. This time the production value was considerably higher. We had specifically hired an advertising agency, which brought external recording equipment and even two experienced streamers. And because the company wanted the final numbers in their report to look good, they allowed us to buy paid traffic.

The problem came quickly. Using external livestreaming equipment on the Video Account, and running paid traffic, both required the account to have a certain minimum number of followers. The few hundred followers we had accumulated over time through genuine engagement were nowhere near enough, so we had to resort to “manual pollination” first.

This was the first piece of money I ever spent on marketing at the company: 150 yuan, to buy 300 Video Account followers.

No sooner had the advertising agency people arrived than they started organising something else. They figured the livestream could not just show raw ingredients — it needed finished dishes too, with the streamer taking a few bites, which would stretch the broadcast a little longer. So we ran upstairs to find a restaurant that could do contract cooking. We knocked on a few doors, and in the end one agreed: seven or eight small dishes, cooking fee totalling 600 yuan.

Once the livestream started, the host’s on-camera pitches and guest interviews alternated. For each product category introduced, we brought a guest from the corresponding industry into frame for a fifteen-minute conversation.

When it was mutton’s turn, I had already invited President Li, whom I had recently met. Before the May Day holiday, I had worked with the District Agricultural Committee on a “New Farmers” series of profiles, and in mine I had described him as a veteran of the entire white-goat supply chain. Taking advantage of the fact that we were still on fairly warm terms, I went to ask him, and he agreed without hesitation.

Just before his segment, I ran through the livestream talking points with President Li. He suddenly got all coy, repeatedly suggesting that someone else go on instead. I assumed he was just nervous before going live, or being polite out of habit, so I did my best to encourage him, saying the segment really needed him. Hesitantly, President Li took the script. I figured that settled it.

When the mutton segment was due to start, I discovered President Li had already left. I had no choice but to grab a staff member from another mutton co-operative right there at the stall to bail us out.

This was so very Qingzhou — in a society built on personal relationships, a casual “I’ll be there” does not have to be taken at face value. And the timing of President Li’s departure was rather convenient: it was just after the attending leaders had left the venue.

◉ The goat pens at President Li’s place.

Back to the market: it was called “Benefits for the Community” in name, but whether residents could actually buy anything was never the point.

Early in the morning, around eight or nine — prime time for residents to shop for groceries — the stalls were already set up and the goods fully displayed, but selling was not allowed to begin. The moment anything was sold, the display stacks would shrink, and when the leaders came for their inspection, it would not look presentable enough.

By the time we were finally allowed to sell, it was already the afternoon. The ice packs had melted, and the remaining meat and poultry had been sitting out for half the day, slowly weeping blood. It had not quite spoiled, but the appearance was such that nobody wanted any of it.

All the District Agricultural Committee contact wanted was “enough” — enough quantity, enough grandeur. How much was wasted was simply not a consideration.

My disgust with all this window-dressing reached its peak that day. Miraculously, just a few days before the event, my voice had given out. For the entire day I could only manage a breathy whisper to handle work calls, and by the end I could not speak at all.

Silence was my last line of defence — the only form of spiritual self-preservation I could hold on to.

The Regulars Vs The Scrappers

◉ Rice paddies in Qingzhou.

Not everyone who works the land can count on help from “upstairs.” More often than not, people rely on their own means of survival.

Sister Wei is exactly that kind of person.

Sister Wei originally joined Qingzhou Taste as a pastry supplier. Everyone in the office had a vivid impression of her — she was incredibly active, always online on her phone. Whether it was new product listings, after-sales issues, or some last-minute announcement, Sister Wei was the first to respond in the group. Many times we could not see a single other supplier chime in; the whole WeChat group felt as though she were the only one there.

A colleague would often send me a short clip, saying it could be re-edited for promotional use. No need to guess the source — it was always from Sister Wei’s WeChat Moments.

After adding her on WeChat, I got a much clearer sense of her posting frequency. At a minimum, she posted eight to ten Moments posts a day — close-ups of products, grid layouts linked to her Kuai Tuan Tuan sales pages, or short clips of about fifteen seconds. And her range was far from limited to pastries. Whatever vegetables, mutton, local free-range chickens and ducks, Chinese mitten crabs … that we carried on Qingzhou Taste, she had on her end, too.

The tool Sister Wei used was a group-buying mini-program: the merchant supplies the goods, sets the price and commission, while the group leader manages their own private audience, shares the link, and earns commission once a sale goes through. What this channel tests is not platform traffic but the group leader’s eye for products and personal influence. That is when Sister Wei’s agricultural produce business really took off.

In truth, Sister Wei had always been an excellent salesperson.

Before selling agricultural produce, she had run a mother-and-baby lifestyle store in Qingzhou for over twenty years and was among the earliest batch to open a shop on Taobao. From 2016, the mother-and-baby business grew harder; Sister Wei kept her store open, but she also joined Amway, travelling to unfamiliar cities like Shenzhen to develop new clients, and her performance was quite solid.

Sister Wei truly has a born salesperson’s constitution — boundless energy, a speaking pace so quick she could go on forever. Her slightly husky voice and the faint Jiangsu–Zhejiang accent in her cadence, somehow, made consumers less likely to feel put off.

The real starting point of her pivot into agricultural produce was the city-wide static management imposed during the 2022 pandemic response. Everyone was locked at home, unable to buy food. Then Sister Wei suddenly realised that the supplies were all around her: her male cousin had a 100-mu nursery with free-range chickens foraging under the trees; her female cousin grew pears and peaches; her other male cousin knew how to make Qingzhou cake … All of these relatives became her suppliers.

In one go, Sister Wei joined about fifty local WeChat groups and started posting the group-buying mini-program links there, covering over forty residential compounds around her small town. To this day she can “point to any of them with her eyes closed,” because she has personally delivered goods to each one.

It was in these WeChat groups, too, that Sister Wei met our salesperson, Old Pu, and subsequently became a Qingzhou Taste supplier. Pastries, fruit and vegetables, mutton — she gradually listed the local produce she was already selling onto Qingzhou Taste. By her estimate, the platform accounted for roughly five per cent of her total revenue. A small share, but for a salesperson of her calibre, no channel is ever left on the table.

This “born salesperson” also demonstrated a remarkable product sense.

◉ Visiting Sister Wei; she shows me the “imperfect crabs.”

Take Chinese mitten crabs, for instance. Beyond the high-ticket gift boxes, Sister Wei opened a separate SKU on her own mini-program for “imperfect crabs” — ones with a missing leg, too ragged for a gift box but cheaper and better suited to eating at home. The official e-commerce platform had to think about presentation and aesthetics; Sister Wei’s channel, by contrast, was far more nimble.

From our perspective, Sister Wei was not exactly the ideal supplier. After all, she was a middleman — her supply prices were no bargain, and tracing quality back to the source was difficult. But she more than made up for it with speed. When a product was just coming into season, she would almost always spot it before we did. That Excel new-product form of ours? She was the one who filled it in most diligently. And so the partnership just kept going, that way.

It was not until a year later, when I visited her in person, that I truly grasped how large Sister Wei’s business had grown.

By then, Sister Wei had close to 9,000 WeChat contacts and eight customer groups of her own — the larger ones with three or four hundred members, the smaller ones with over a hundred. On top of that, she had recruited 181 downline group leaders. Through the mini-program’s “One-Click Resale” mode, these downline leaders did not need to handle shipping; they simply managed their own WeChat Moments and collected commission directly.

Alone, Sister Wei did nearly the work of an entire marketing department: sourcing products, producing content, managing customers, running communities, handling after-sales.

By comparison, Qingzhou Taste’s membership numbers and community size paled next to hers. Or more bluntly: Qingzhou Taste was, in effect, one of Sister Wei’s “downlines.” The power dynamic I had assumed between platform and supplier may have been entirely the other way round.

Looking back at Sister Wei, I felt it for the first time with real force: over the past two years, much of what I had been involved in had been nothing but playing house.

We took the marketing calendar seriously, designed membership schemes, planned seasonal campaigns, and celebrated when the Official Account read count nudged up by a hundred or two. But the people who actually sold agricultural produce did not rely on any of those systems. They built their businesses up, bit by bit, on their own supply lines, personal networks, WeChat groups, and Moments.

We thought we were building a platform. In reality, we were simply borrowing momentum from sales networks that others had long since built.

Two years into my tenure, the company’s shareholding structure shifted. The general manager, who had been banking on a funding round, was ultimately pushed out by the capital side, his management authority stripped along with his seat. The marketing department he had spearheaded was also ordered to relocate to the parent company’s offices.

Before long, Sister Kong resigned. Old Pu left, too.

The two operations colleagues and I had almost nothing to do: one Official Account post a week, a mini-program promotion adjustment every fortnight. Sitting in the new office, we were as good as invisible — three ghosts.

But it was also during that period that I began to notice subtle shifts.

That January, the district government work report called for strengthening the “Qinghe Three Brothers” regional public brand for agricultural products, but for the first time in years, it did not pair that with language about strengthening the mini-program platform Qingzhou Taste. The name of our platform had simply vanished from the official plans.

Did the company end up securing the government subsidy that year? I have no idea. With the key figures gone, would the District Agricultural Committee keep pouring support into Qingzhou Taste? I doubt it. The parent company clearly saw this coming, too, and naturally stopped adding headcount to the project.

Half a year later, the three of us left, one after another.

To this day, Qingzhou Taste has not been taken offline. The Official Account bio still reads “Official Designated Sales Platform for the Qingzhou Agricultural Products Regional Public Brand.” It is only that the hidden layer of connection that once kept it running has long since disappeared.

While writing this piece, I came across a newer district government work report.

A year after we left, the report noted: “Strengthen the building and promotion of the regional public brand, and continue to expand channels for agricultural product exhibition and sales.”

Two years after our departure, it had shifted again to “Deepen the building of the Qingzhou Fresh brand and platform.”

A familiar plotline surfaces once more. You see, there is always a new generation to take its turn.

– This is the 825th original from Foodthink –

Foodthink

Frankie

Struggling through life, writing on the side.

About the Lianhe Creative Programme

To understand the state of food and agriculture today, and to support more people in exploring the complexities behind food and farming issues, Foodthink, together with a number of public-interest and media partners, launched the Lianhe Creative Programme in 2024 and 2025 to fund research and public-facing content creation by media creators and researchers in the food and agriculture field.

After multiple rounds of interviews by the judges, the Lianhe Creative Programme supported 18 creative projects in 2024 and supported 20 creative projects in 2025. To date, some of the supported works have been completed and published:

A Cleaner Named A-Mei Just Wants a Proper Meal | A Worker’s Table

In Malaysia, Chinese Buyers Only Want Grade A Durians

“Fake Meat” Drives Out Real Meat: Herders, the Table, Amazon

Watermelons Guaranteed Sweet, Growing Them Guaranteed Bitter

From the Shangshan Yao to the “Heaven-Chosen Mushroom Forager”: A Foraging Frenzy Sparked by a Single Termitess Cap

Ma Lan in Shenzhen, Without a Meal Buddy

Why Has the Sweetness of Childhood Disappeared?

Why Can’t Guizhou Do Without Sour Soup, and Why Can’t Sour Soup Do Without Guizhou?

Who Drove the Wet Market Out?

Will Fresh-Food E-Commerce Make Wet Markets Disappear?

When Drones Become New Farming Tools: Who Gets to Define “Scientific Farming”?

A Braised Goose in the Qiaopi Remittances: 100,000 Yuan a Bird