Tang Binsen dives into the deep water area, Yuanqi Forest strives to "make mistakes"

Ask AI · How can Tang Bingsen improve the success rate of product innovation through trial-and-error mechanisms?

How great products get a chance to exist

Fast Reading

■

Through public evaluation mechanisms such as the “Creativity Camp 2044,” Nongqi Forest has shifted product decision-making from private boardrooms to an open field—so that products receive instant feedback before they become results.

■

With the implementation of Nongqi Forest’s Ning’an Innovation Institute, Nongqi Forest has established a tiered R&D structure of “bench testing – pilot testing – mass production.” By building a mini factory and pilot production lines, it creates an intermediate layer covering the span from “idea to mass production.” This lowers trial-and-error costs while shortening the path from proof of concept to small-scale production.

■

In the face of a harsh market environment, on the occasion of the ninth anniversary of its founding, Tang Bingsen proposed “don’t overdo it” and “resist temptation.” He emphasized that after building core capabilities to withstand risks, the company must resist hollow theories and blind expansion, focusing on the main track to deal with the uncertainty of 2026.

Author | Su Yinan

Editor | Huang Yuntao

▌****Peaches at 4:00 a.m.

On the afternoon of March 30, in Ning’an, Hubei, it rained lightly. Nongqi Forest set the stage on a patch of grass outside the lab. A young product manager stood on stage and told a story about eating peaches in the orchard at 4:00 a.m.

Her name was Xu Hui. During early summer, she went to visit a peach orchard and saw peaches everywhere across the hillsides. Just as she was about to pick one to taste, a farmer stopped her. “At this time, the peaches don’t taste at their best. You have to wait until 4:00 a.m., when the dew is still there.”

She really waited until that time. At daybreak, the peaches were a bit green, a bit pink, and a bit hard. But when she bit into one, she remembered that “fresh, crisp” flavor.

Later, the story became the starting point for Nongqi Forest’s bubble water “Burst-juice White Peach.”

That time—4:00 a.m.—already carries a kind of unstable texture: it doesn’t belong to the night, and it hasn’t fully handed itself over to daytime either. Many flavors exist only at this moment.

In a sense, what Xu Hui wanted to capture wasn’t a “better-tasting” flavor, but a fleeting slice of time.

Later, she tried to break that sense of freshness into three dimensions: the color was the faint pink that comes from sun exposure; the aroma was a slightly green, immature fruit fragrance; and the mouthfeel wasn’t a single kind of sweetness—it was sweet with “ups and downs.”

In the end, she summed up her product philosophy: “Not making fruit taste more like a drink, but restoring drink so it’s more like fruit.”

Under the stage, there were hundreds of audience members wearing chest badges that read “lead drinker.” As they listened, they played with the electronic voting devices in their hands. With every press of a button, it felt like pushing a newly formed idea toward some fate: accepted, kept, modified, or quietly abandoned without a sound.

This was the现场 of the second “Nongqi Forest Creativity Camp 2044”—a beverage tasting event blending product debuts, flavor evaluations, and cross-industry conversations. The reason it anchors on “2044” is that the organizers hope the event can continue for at least 20 years.

_ Xu Hui speaking on stage_

No red carpet, no long, drawn-out process—only a group of product managers with an average age under 30, standing on stage to tell their own stories. But this kind of “de-staging” presentation itself becomes a new kind of stage.

Most consumer companies are used to showcasing results under spotlights. Nongqi Forest, however, chose to make the process public too. In traditional consumer companies, this kind of judgment is often made in meeting rooms. Now, it’s laid out on the grass—turning into a public act with a bit of randomness.

According to the company’s founder, Tang Bingsen, it’s more like a “recording session”—recording those ideas that haven’t yet been proven, and recording what a product looks like before it becomes a product.

▌****Enough trial-and-error samples

Tang Bingsen doesn’t like to make this sound complicated. A product manager said that Tang Bingsen’s Feishu signature is: “To do simple things to the extreme is a form of innovation.”

The purpose of this event is very straightforward. At the start, Tang Bingsen said, “Many people like to summarize methodologies after product success. But at the beginning, it wasn’t as great as the outside world imagines. In reality, it’s just some simple, rough ideas.”

He’s opposed to excessive packaging. There’s no “super single-product methodology,” no “consumer upgrading curve,” and he doesn’t even emphasize “strategy” very much. In his description, Nongqi Forest is more like a “cook-your-own-kitchen” operation: starting from a thought, constantly trying and constantly changing.

In the kitchen, hardly anyone knows precisely what the finished dish will be at the beginning. More often, you cook while tasting. If the flavor is off, you add something more. If the heat is wrong, you start over again. This process is naturally low-efficiency—but that’s also why it retains a kind of elasticity: it allows for mistakes, and it allows for the unexpected.

For many years, the mainstream logic in the beverage industry has been simple and blunt: find a certain demand, roll it out through channels and amplify it with marketing, and quickly grow a big single product. The product itself is obviously important, but more often than not, it’s a result that needs to be packaged and scaled.

But on that grassy field in Ning’an, the products are not yet “results.” They’re more like ideas that haven’t fully taken shape—repeatedly told, refined, questioned, and praised by teams of product managers.

Another product manager talked about “punk wellness.”

He used to work on health supplements, then moved into the beverage industry and observed a phenomenon: many people buy health supplements, but can’t stick with them, can’t remember them, and in the end they just leave them there—until they expire. “Many nutrition supplements emphasize taking them after meals, on schedule. But in reality, just thinking about what to order for delivery already leaves you stuck in decision paralysis—who remembers to swallow one more pill?” So he tried to blend the idea of taking vitamins into everyday drinking water.

The logic behind the VC water he led sounds simple: a bit more VC and zinc, and a bit less sugar burden; a bit more relaxed “top-up,” and a bit less of a persistence threshold. He said this isn’t a more complex product. It’s an attempt to make nutrition supplementation no longer just something people talk about, but something a little more convenient, a little lighter, and easier to keep up with.

There was another product manager who talked about climbing mountains.

She described a scenario: when people hike in high-altitude areas, their bodies are exhausted but their emotions are excited. When they stop to rest, many people want to drink a sip of soda.

She broke that demand into three layers: the first is the refreshment sensation; the second is replenishment; the third is an emotional reward. Traditional sodas can only provide the first layer. Functional drinks lean heavily toward the second layer. And what she wanted to do was to combine all three. That’s how a “carbonated electrolyte beverage” came about: it has the immediate stimulation from bubbles, replenishment of electrolytes and vitamins, and it can give yourself an ice-cold, invigorating reward in the moment when you’ve been sweating hard. This is a typical process of “reverse-engineering a product from a scenario.”

Over more than two hours, a dozen-plus product managers brought a range of new products: iterations of Nongqi Forest’s mature product lines such as sparkling water, an “alien” electrolyte water, iced tea, and Good to Go; as well as new experiments like cooling tea that tried to make some change on the traditional drinks track. Meanwhile, brand-new faces such as Coke beer, prebiotic vitality sparkling beverages, three great companions of hawthorn (Shan Zhi San Junzi), and Dan Zhen brown rice milk were aimed at different scenarios like gatherings, breakfast, and family reunions.

These small innovations aren’t about creating an entirely new category; they’re fine-tuning within a mature structure. When these stories are put together, it’s hard to form a unified methodology.

If there’s anything in common, it’s that they all tried to continuously turn “fuzzy liking” into “a product that can be manufactured and quantified,” starting from personal experience, specific scenarios, and subtle feelings.

During this process, the product managers are both observers and experimenters. The voting mechanism gives these attempts an instant feedback system. After each round of sharing, the vote counts are locked on the spot. Without complex decision procedures, the outcome is exposed directly to the room.

Tang Bingsen sat in the audience throughout the entire time and rarely provided instant feedback on the product managers’ shares on stage.

What he cared about most wasn’t which product could win. More importantly, it might be somewhere else—how to get more people to propose ideas, how to quickly validate those ideas, and how to keep some parts of them even after failure. Summarized in one sentence: how to give good products a chance to exist.

Nongqi Forest develops a large number of new products every year, but not that many ultimately make it to the market.

Tang Bingsen’s stance is: “Innovation can only be systematized into a methodology through enough trial-and-error samples, and it must rely on the company’s existing systems and talent to produce innovative products.”

When trial-and-error becomes something encouraged—sometimes even institutionalized—in a company, what it truly accumulates isn’t just products, but a capability about “how to generate products.”

This capability is hard to explain clearly, but it will gradually take shape over time.

▌****The Ning’an factory

About 100-plus meters away from the “Creativity Camp 2044” stage is the newly completed Nongqi Forest Ning’an Innovation Institute—another major reason for Tang Bingsen’s trip to Hubei.

Put into use at the end of 2025, the Ning’an Innovation Institute has a total investment of 200 million yuan in Phase I. It is jointly funded by the Ning’an municipal government and Nongqi Forest, and it is positioned as a one-stop R&D innovation testing base—specifically serving the incubation of innovative categories and the landing and transformation of new processes. Nongqi Forest places great hope in it, saying it is an “important platform for moving the company’s R&D investment into deeper waters.”

Tang Bingsen said that over many years of making beverages, Nongqi Forest has always wanted to explore many processes and make many attempts. “The deployment of the Ning’an Innovation Institute follows our ideals—built according to the experimental lines we’ve planned for the future.”

It isn’t a traditional laboratory in the usual sense—it’s an intermediate layer covering the span from “creative concept to mass production.”

The Ning’an Innovation Institute is equipped with pilot production lines, mini factories, and other equipment, connecting multiple key stages from proof of concept to small-scale production. The mini factory can produce small batches of products at extremely low costs. Further up the chain is the pilot factory—closer to mass production—equipped with more complete process capabilities.

Starting from a wild and imaginative idea, a product here can complete the whole closed loop from testing to production. It lowers trial-and-error costs and also shortens the path from ideas to products.

If you put these clues together, you’ll find that what Nongqi Forest is building is more like a new production mechanism.

The first layer is the reconfiguration of how products are “generated.” Products are no longer the result of decisions made by a small number of people; instead, they’re proposed, discussed, and filtered in a more open environment. The second layer is the industrialization of R&D capability. By using a tiered structure of bench testing, pilot testing, and mass production, it accelerates the path for product implementation. The third layer is a change in the organizational structure. Product managers are pushed to the front stage—so everyone has the opportunity to propose their own products, instead of operating around a single predetermined big hit product.

When these are layered together, they form a pattern between the iteration model of internet products and the manufacturing model of traditional consumer goods.

Nongqi Forest doesn’t shy away from the fact that it’s becoming “heavier” and heavier. Counting the seventh self-built factory it’s currently preparing, its total cumulative investment in factories has nearly reached 8 billion yuan. The number of offline outlets has surpassed one million, covering more than 800 cities nationwide.

Build high walls and store up grain. But that doesn’t mean you can relax.

In February this year, in an internal letter Tang Bingsen released to mark the ninth anniversary of Nongqi Forest, he wrote: “We still have a long way to go before becoming a truly excellent beverage company… The market environment remains tough. The risk-resilience we’ve built is only a foundation—we’re nowhere near a time for optimism.”

In that letter, he shared stories about Tang Seng and the turtle. The former may look weakest, yet because it holds a firm mission, it has the most unshakeable power. The turtle in the race between the hare and the tortoise is also a typical example of single-minded devotion—each step is solid and effective.

How to face 2026? Tang Bingsen told everyone: there’s no such thing as a brand-new grand plan. Resist temptation, resist all kinds of blind actions and hollow theories. Don’t overdo it—focus on the main track and stay true to the初心 of making good products.

▌****Epilogue

It didn’t rain hard. It just kept falling, like a background sound that wasn’t quite confident. The grass was wrapped in a layer of dampness; when your shoe soles stepped onto it, there was a slight sinking sensation.

By evening, all the voting activities for this session of the camp were finished.

Some products received high scores, while others were relatively plain. That means some products may be massively mass-produced in the future, while others face being re-adjusted—or even completely overturned.

It’s hard to say which of these products will definitely become the next breakout hit. But during this process, what remains isn’t just a few beverages. Even more important are those ideas carrying “fresh, crisp energy”—the peaches at 4:00 a.m., the sip of soda during a climb, or a home-style drink with honeysuckle and snow pear flavor.

Here, they’re told, challenged, and recorded.

_ Cover source: photographed by Snow Leopard Finance and Economics Society _