Youdu Capital · Primary-Market Financing
On 2026-10-19, Radical Ventures published an article about the forecasting system of its portfolio company Mantic. The article describes Mantic checking every system change against questions whose outcomes are already known. This article does not adopt the competition performance promoted in that source, and we have not tested the product first-hand. The source is more than seven days old and is used as background within the fourteen-day window. Overseas material is for background reference only. Youdu Capital serves domestic financing for Chinese technology companies and does not provide cross-border financial services.
In China, on 2026-10-21, the Ministry of Science and Technology published the Opinions on Promoting Deep Integration of Technological Innovation and Industrial Innovation. The document encourages state-owned enterprises to establish R&D reserve arrangements that include exemption from assessments of preserving and increasing asset value, together with provisions for tolerating and correcting errors. This is a policy direction for specified entities. It does not mean that private companies have received funding, assessment exemptions or a universally applicable management rule. It raises a distinction worth considering: should information used to understand change also carry the function of evaluating responsibility?
The following discusses a hypothetical organizational situation, rather than evaluating the institutions or companies cited. A founder asks a business manager for an operating forecast and directly uses the same number to determine performance rewards and penalties. A forecast answers what may happen given current information; a target describes what the team intends to achieve. They may be close, but do not necessarily need to be set through the same reporting exercise. Otherwise, the person supplying information must also consider how changing the number will affect them personally.
The first causal layer concerns reporting incentives. Suppose a customer postpones procurement and a manager lowers the expected contracting schedule, only for this to be immediately interpreted as voluntarily reducing their work expectations. The update then carries an additional personal cost. Even without false reporting, the manager may retain an optimistic explanation that has not been fully disproved and wait for a clearer outcome before revising it. The issue is not whether the manager is sufficiently ambitious, but whether the organization puts additional pressure on someone who reports adverse information early. Accurate updating and striving for better results may consequently come into conflict.
The second layer concerns dependencies between departments. If sales continues to report the original contracting schedule, the delivery team may keep reserving implementation staff, while the product team schedules customer-specific adaptations accordingly. By the time the forecast changes, other tasks have already been pushed back and schedules that could once have been adjusted are harder to move. More departments share the cost of late information, although it may not appear in the original reporter’s results. The founder may see execution problems across several departments when the starting point was an assumption that was not updated promptly.
Targets and current forecasts can be retained separately, making clear that updating a judgment neither automatically resets rewards and penalties nor removes the need for explanation. Whether a target changes remains a separate discussion for someone with the appropriate authority. The forecast should change promptly with new evidence. The manager needs to explain whether the revision reflects the customer’s process, delivery conditions or a corrected judgment, and which arrangements may be affected. This retains the question of responsibility without requiring every information update to begin with negotiating a concession on the target.
The distinction must not become an excuse for continually lowering expectations. If a forecast falls repeatedly as an assessment approaches without new facts, the reasons still need checking. If a change was known early but not passed to dependent departments, the interruption in communication also needs explanation. The relevant questions are what was known at the time and what action was taken, rather than using final target attainment alone to infer whether the earlier judgment was honest. A forecast that ultimately misses does not by itself establish a manager’s failure of duty; a forecast that happens to be correct does not replace understanding the process.
An operating meeting can first address actions arising from a forecast change and then examine responsibility for executing the target. For example, if customer acceptance conditions remain unmet, first let delivery and product managers know which work needs reordering, instead of first asking who will take responsibility for the original date. Then discuss what the team can control and what still needs checking. This does not abandon the target; it changes the order of discussion. Useful information enters resource decisions first, while the performance discussion has its own factual basis, reducing the risk that the two obscure each other in the same argument.
Over the next three to six months, if a company enters a period when targets are being allocated and schedules for the next stage are being updated together, it can select a business line with clear dependencies across departments. Retain the original target and latest forecast, and observe when adjustments follow adverse evidence and when other departments change their arrangements accordingly. The point is not how often forecasts fall, but what choices remain when information arrives. If forecasts update while dependent departments continue to use old assumptions, first correct how the information is transmitted and used, rather than immediately changing the performance assessment criteria.
If practical use shows that the team is using forecast updates to avoid action, return to the evidence and matters within its control. If changes are reported earlier and resources can be rescheduled promptly, continue observing whether the distinction is useful without immediately turning it into a new scoring standard. These are conditional operational scenarios. They do not assert that the companies cited conceal information or have incentive problems, and they promise no operating or financing outcome. Consult qualified professionals on performance arrangements, employment relationships and specific agreements. This article does not constitute specific investment, legal or tax advice.
Sources The Era of Superhuman Forecasting Is Here (2026-10-19) https://radical.vc/articles/the-era-of-superhuman-forecasting-is-here/ 关于推动科技创新和产业创新深度融合的意见 (2026-10-21) https://most.gov.cn/xxgk/xinxifenlei/fdzdgknr/fgzc/gfxwj/gfxwj2026/202610/t20261021_197540.html
Content note: This article was drafted with AI assistance and reviewed through an independent AI process, using the listed public sources. It contains no original interviews, and no first-hand testing of the products, services or methods discussed was conducted. Checking public sources is not a field investigation or first-hand experience.