Our Mission
Decision-making power should never belong to platforms
There are 300 million households in China making food decisions every day. They scan QR codes, read ingredient lists, study components, search for reviews, and ask friends. They spend a great deal of time, yet still fail to obtain an answer they can truly trust. Not because they are not intelligent enough, but because the information system of this market was never designed for them in the first place.
Why has this market failed for so long?
It is not due to a lack of information. It is because a true decision-making mechanism has never been established. Shelf-based e-commerce platforms generate revenue from merchant fees. When a platform simultaneously serves as both a merchant’s sales channel and a consumer’s shopping advisor, it structurally loses the ability to reject its own merchants. This is not a moral issue—it is a mathematical one. You cannot serve two principals with conflicting interests at the same time. Content-driven commerce earns from brand advertising spend. Platform algorithms optimize for clicks and conversions, not truth. Behind every recommendation, consumers can never be certain: is this a genuine opinion, or a paid placement? General-purpose large models are trained on internet content. And the internet is dominated by brand websites, marketing articles, and product pages. The “food knowledge” these AIs learn is, in essence, a product of brand marketing. Traditional testing reports are owned by brands. Institutions conduct the tests, brands pay for them, and brands own the reports. Standing in front of a shelf, consumers can never know whether a product has been tested, what the results are, or who is accountable. These four existing systems each have their own structural flaws. And they all share one root cause: their revenue comes from brands, not users. When a system’s survival depends on brand satisfaction, it is structurally incapable of truly standing with the consumer. This is an inevitable outcome determined by incentive structures, not a failure of effort by any single company.
Therefore, this problem cannot be solved by optimizing existing systems. It requires a new system that, starting from its revenue structure, stands entirely on the user’s side.
Suangoo is the first attempt at building such a system.
We have written our revenue structure into our corporate charter: 100% of revenue comes from users, a maximum fulfillment profit margin of 5%, no acceptance of brand payments, and no advertising. This is not a promise that can be easily changed. Amending the charter requires formal legal procedures and shareholder approval. The significance of this design is not that “we are principled,” but that it transforms independent judgment from a value into a structure.
What are we building?
Suangoo is an AI-driven decision system for the human body, starting with food. A user points to a product barcode or ingredient list, and within seconds, Suangoo completes ingredient recognition, regulatory comparison, and risk grading. Combined with the household’s health profile—such as allergies, chronic conditions, and dietary goals—it delivers an actionable verdict: Safe, Caution, or Reject. Not a suggestion. Not a reference. A verdict. Each decision is written into system memory, becoming the foundation for future judgments. The more it is used, the better it understands the household—and the harder it becomes to replace. Our AI is not trained on the internet. It is trained on real testing reports issued by 120+ CMA/CNAS-certified institutions, over 500 rejection cases with full evidence chains, and real household health data from users. Our testing reports are commissioned and paid for by Suangoo, and owned by Suangoo. Every rejection record includes specific detected values, applicable regulatory limits, the issuing institution’s name and certification number, and complete supporting evidence. Anyone, at any time, can verify it.
Why is rejection our most critical asset?
For most platforms, rejecting a supplier means losing revenue. For Suangoo, rejecting a supplier is the accumulation of an asset. Each rejection record is a training signal that teaches the system how to identify problematic products. Every “no” makes the system harder to deceive. This creates a moat that deepens automatically over time. Currently, we have established evaluation coverage over 3,000+ upstream suppliers, with a final pass rate of less than 1%. When everyone else says yes, the institutional will to say no becomes a barrier in itself.
Why are we doing this?
Some problems are not recognized until they are experienced firsthand. A person lying in a hospital bed wants to buy a bowl of truly safe porridge. They have money, time, search tools, and purchasing channels—yet still cannot find a trustworthy answer. Not because resources are insufficient, but because the system was never designed for this need. This person is not an ordinary consumer. She has managed real capital, seen countless business models, and understands how to evaluate whether a system’s incentive structure truly holds. She knows better than anyone: this is not about any single brand being inadequate, nor any platform lacking effort. It is that all existing systems are structurally incapable of being honest with users. This is why Suangoo exists. Not because it is a market opportunity, but because it is a structural problem that must be redesigned.
Our mission:
Let decision-making power truly belong to the user
Reject 99% of mediocrity, and deliver the 1% certainty — for humanity
Why has this market failed for so long?
It is not due to a lack of information. It is because a true decision-making mechanism has never been established. Shelf-based e-commerce platforms generate revenue from merchant fees. When a platform simultaneously serves as both a merchant’s sales channel and a consumer’s shopping advisor, it structurally loses the ability to reject its own merchants. This is not a moral issue—it is a mathematical one. You cannot serve two principals with conflicting interests at the same time. Content-driven commerce earns from brand advertising spend. Platform algorithms optimize for clicks and conversions, not truth. Behind every recommendation, consumers can never be certain: is this a genuine opinion, or a paid placement? General-purpose large models are trained on internet content. And the internet is dominated by brand websites, marketing articles, and product pages. The “food knowledge” these AIs learn is, in essence, a product of brand marketing. Traditional testing reports are owned by brands. Institutions conduct the tests, brands pay for them, and brands own the reports. Standing in front of a shelf, consumers can never know whether a product has been tested, what the results are, or who is accountable. These four existing systems each have their own structural flaws. And they all share one root cause: their revenue comes from brands, not users. When a system’s survival depends on brand satisfaction, it is structurally incapable of truly standing with the consumer. This is an inevitable outcome determined by incentive structures, not a failure of effort by any single company.
Therefore, this problem cannot be solved by optimizing existing systems. It requires a new system that, starting from its revenue structure, stands entirely on the user’s side.
Suangoo is the first attempt at building such a system.
We have written our revenue structure into our corporate charter: 100% of revenue comes from users, a maximum fulfillment profit margin of 5%, no acceptance of brand payments, and no advertising. This is not a promise that can be easily changed. Amending the charter requires formal legal procedures and shareholder approval. The significance of this design is not that “we are principled,” but that it transforms independent judgment from a value into a structure.
What are we building?
Suangoo is an AI-driven decision system for the human body, starting with food. A user points to a product barcode or ingredient list, and within seconds, Suangoo completes ingredient recognition, regulatory comparison, and risk grading. Combined with the household’s health profile—such as allergies, chronic conditions, and dietary goals—it delivers an actionable verdict: Safe, Caution, or Reject. Not a suggestion. Not a reference. A verdict. Each decision is written into system memory, becoming the foundation for future judgments. The more it is used, the better it understands the household—and the harder it becomes to replace. Our AI is not trained on the internet. It is trained on real testing reports issued by 120+ CMA/CNAS-certified institutions, over 500 rejection cases with full evidence chains, and real household health data from users. Our testing reports are commissioned and paid for by Suangoo, and owned by Suangoo. Every rejection record includes specific detected values, applicable regulatory limits, the issuing institution’s name and certification number, and complete supporting evidence. Anyone, at any time, can verify it.
Why is rejection our most critical asset?
For most platforms, rejecting a supplier means losing revenue. For Suangoo, rejecting a supplier is the accumulation of an asset. Each rejection record is a training signal that teaches the system how to identify problematic products. Every “no” makes the system harder to deceive. This creates a moat that deepens automatically over time. Currently, we have established evaluation coverage over 3,000+ upstream suppliers, with a final pass rate of less than 1%. When everyone else says yes, the institutional will to say no becomes a barrier in itself.
Why are we doing this?
Some problems are not recognized until they are experienced firsthand. A person lying in a hospital bed wants to buy a bowl of truly safe porridge. They have money, time, search tools, and purchasing channels—yet still cannot find a trustworthy answer. Not because resources are insufficient, but because the system was never designed for this need. This person is not an ordinary consumer. She has managed real capital, seen countless business models, and understands how to evaluate whether a system’s incentive structure truly holds. She knows better than anyone: this is not about any single brand being inadequate, nor any platform lacking effort. It is that all existing systems are structurally incapable of being honest with users. This is why Suangoo exists. Not because it is a market opportunity, but because it is a structural problem that must be redesigned.
Our mission:
Let decision-making power truly belong to the user
Reject 99% of mediocrity, and deliver the 1% certainty — for humanity