To reduce bad debts by hundreds of millions, a bank launched a satellite into space.
“What we're most worried about internally is that if the high - definition images captured by the satellite cannot be quickly converted into structured data or unstructured data that can generate business value, and thus effectively improve the bank's response speed in credit risk prevention, the investment of tens of millions of yuan will be in vain,” said the head of the financial technology department of a joint - stock bank.
On April 14, 2026, the “Postal Savings Bank Satellite”, jointly developed by the Postal Savings Bank of China and Changguang Satellite Technology Co., Ltd., was launched at the Jiuquan Satellite Launch Center in Gansu.
At the beginning of this year, the “CMB Golden Sunflower Satellite” of China Merchants Bank and the “SPDB Digital Intelligence Satellite” of Shanghai Pufa Bank were successively launched. Both of these two satellites belong to the “Tianqi Constellation”, the first low - orbit satellite Internet of Things constellation with global networking in China.
“Banks launching satellites into space has suddenly become a trend,” Feng Junyu sighed to a reporter from Economic Observer. “It's quite enviable.”
Feng Junyu is the supervisor of the credit risk control business of a joint - stock bank and has long been concerned about the iterative innovation of bank credit risk control measures.
“As remote - sensing satellites have the ability of full - color high - definition imaging with a resolution better than 0.5 meters, as well as the remote - sensing monitoring capabilities of rapid response, high - frequency imaging, all - weather operation, and high coverage, this technology can fill the blind spots in risk control in many credit scenarios,” Feng Junyu admitted. Two months ago, the credit risk control department he works in also submitted an application to the head - office leadership to introduce satellite remote - sensing technology as soon as possible.
As the director of credit risk control of a city commercial bank, Gu Qian is also actively persuading the bank's senior management to build a “broadband + narrowband” satellite matrix as soon as possible.
Broadband satellites can mainly provide high - data - rate and low - latency communication capabilities, and can conduct video monitoring, remote - sensing image transmission, and real - time data interaction for the production in loan enterprises' factories, commodity logistics, and building construction. Narrowband satellites can take high - precision and high - resolution images of crop planting and enterprise operation conditions in remote areas with weak communication signals, effectively expanding the bank's credit risk control coverage.
However, there is still a long way to go for banks to quickly integrate the high - definition images captured by remote - sensing satellites into the credit risk control process and form effective risk control early - warning and risk prevention measures.
“What we're most worried about internally is that if the high - definition images captured by the satellite cannot be quickly converted into structured data or unstructured data that can generate business value, and thus effectively improve the bank's response speed in credit risk prevention, the investment of tens of millions of yuan will be in vain,” said the head of the financial technology department of a joint - stock bank.
Why Are Banks Launching Satellites Frequently?
Before joining this city commercial bank, Gu Qian worked in a joint - stock bank and witnessed the application results of remote - sensing satellites in the field of credit risk control for the livestock industry.
In the past, banks had difficulty in assessing the milk production and specific income of pastures in remote areas and were reluctant to lend to them. However, when banks equip cows in these remote areas with Internet of Things devices and use narrowband remote - sensing satellite images to monitor the number of cows and milk production in real - time, they can accurately calculate the approximate income of the pastures and provide corresponding credit funds to support them in raising more cows to increase revenue.
In the past, after banks issued agricultural loans to farmers and farmer cooperatives, they would send people to conduct irregular on - site inspections of crop pest conditions and yield changes to take corresponding credit risk prevention measures. However, due to the long interval between on - site inspections, banks had difficulty in timely obtaining the latest information on crop pests and yields, and the credit risk control measures were passive and lagging. Now, banks can monitor the growth of crops with high frequency through remote - sensing satellites, accurately judge whether crops have suffered from droughts, floods, pests, and diseases, and quickly take corresponding risk - remediation and credit risk control measures.
The same is true in the field of cross - border trade financing. In the past, banks mainly relied on the “three forms and one certificate” (purchase order, payment order, logistics order, and customs clearance form) submitted by enterprises to provide trade financing. Once the cargo ship could not dock at the port or the goods were detained in the customs warehouse for too long, banks could only be anxious about the trade financing risk. Now, banks can use the images captured by remote - sensing satellites to timely understand the latest progress of commodity logistics. Once a problem occurs, banks can urge enterprises to find ways to speed up the delivery and collection of goods and require enterprises to add collateral for trade financing loans, effectively reducing the credit risk of trade financing.
Banks' involvement in remote - sensing satellite technology to improve credit risk control capabilities began in 2020 and can be roughly divided into two categories: one is to seek cooperation with remote - sensing satellite operating agencies at a relatively low cost; the other is to invest tens of millions of yuan to launch satellites.
In 2020, MYbank was the first to apply satellite remote - sensing technology in the rural financial scenario. By using deep - learning AI (artificial intelligence) technology to analyze satellite images, it can identify the planting area, types, and growth conditions of crops, evaluate farmers' assets and future income, and provide them with credit services.
At the end of 2020, Ping An Bank launched its first Internet of Things satellite, “Ping An 1”, kicking off the prelude of banks directly launching satellites to strengthen credit risk control capabilities.
After that, many banks followed suit.
In 2021, the Industrial and Commercial Bank of China used satellite remote - sensing technology to verify land ownership in Heilongjiang, Inner Mongolia and other places, identify crop types, obtain business information of crops such as corn and sugar beets, and accurately monitor the growth of crops to improve agricultural credit risk control capabilities.
In 2022, Xingye Digital Technology, a subsidiary of Industrial Bank, launched a satellite remote - sensing application system for credit risk control in businesses such as crop planting and forestry carbon sinks.
In the same year, Ping An Bank launched its second Internet of Things satellite, “Ping An 2”.
From 2024 to 2025, China Merchants Bank launched two satellites successively to monitor the construction progress of real - estate projects, etc., as a new tool to enhance the post - loan management capabilities of real - estate loans.
Since 2026, Shanghai Pufa Bank, China Merchants Bank, and the Postal Savings Bank of China have successively launched satellites to enhance their credit risk control capabilities in respective business scenarios.
Seeing many banks introducing satellite remote - sensing technology to strengthen credit risk control capabilities, Feng Junyu is also eager to catch up.
After applying to the head - office leadership to introduce satellite remote - sensing technology as soon as possible, he began to look for a cooperation model that can achieve the best input - output ratio.
The Game of the “Economic Account” of Input - Output
There are three options in front of Feng Junyu.
The most cost - effective option is for the bank to spend several million yuan each year to purchase satellite image data from remote - sensing satellite operating agencies. However, its disadvantages are quite obvious. Some remote - sensing satellites do not have the ability of full - color high - definition imaging with a resolution better than 0.5 meters, and their orbits may not match the bank's business scenarios. As a result, a considerable proportion of satellite image data cannot be used by the risk control department, causing a certain degree of waste of investment.
Feng Junyu believes that a practical method is to spend tens of millions of yuan each year to rent a launched remote - sensing satellite. The advantage is that this satellite can serve the bank whole - heartedly, but the disadvantage is that its set orbit, remote - sensing imaging accuracy, and resolution may not match the bank's post - loan management requirements.
The method that Feng Junyu most desires but is also the most difficult to implement is for the bank to spend tens of millions of yuan to jointly develop and launch a satellite with an aerospace enterprise. The advantage is that both the orbit and the remote - sensing imaging accuracy and resolution of this satellite are “tailor - made” around the bank's credit risk control needs, but the disadvantage is that the initial investment cost is quite high, and the bank may not be willing to pay for it.
Feng Junyu admitted that there have been continuous disputes within his bank over which cooperation model to choose in the past two months. After all, since last year, banks have started to cut the R & D budget for financial technology in consideration of cost - reduction and efficiency - improvement. If the satellite usage expenses are included in the financial technology R & D budget, it will inevitably lead to a further reduction in the digital transformation investment of other business departments, causing dissatisfaction among them.
At the beginning of April, at an internal meeting, the head of the marketing department of Feng Junyu's bank said bluntly that the risk control effect brought by using satellites may not be better than “manual operation + ground Internet of Things technology”.
The head gave an example. In the agricultural credit scenario, the remote - sensing satellite takes a photo every month to compare the growth of fruit trees (to evaluate whether the farmers' future income meets expectations). Instead of letting the satellite “do this”, it is better to require the branch customer managers to conduct on - site inspections in the countryside every two weeks, which can achieve the same risk control effect and save the bank several million yuan in satellite usage fees each year.
At the same time, the head of the bank's finance department believes that if the satellite remote - sensing technology can only help the bank reduce the bad debts of agricultural loans by several hundred thousand yuan each year, it cannot offset the annual expenditure in the millions, and the input - output ratio is relatively low.
In April, Gu Qian went to an aerospace enterprise for research and found that with the decline in the cost of aerospace satellite networking, banks can spend about 30 million yuan to lease “narrowband + broadband” satellites or purchase high - precision and high - resolution satellite data services from them.
However, the senior management of Gu Qian's city commercial bank has set a pre - condition for cooperation. That is, through this “narrowband + broadband” satellite matrix, the bank can reduce about 200 million yuan in credit bad debts each year, and the balances of special - mention loans and sub - standard loans should both decrease significantly year - on - year.
Lobbying Strategies
Facing the questions from other departments, Feng Junyu has his own coping strategies. He is trying to let the bank's leadership personally witness the specific effects of this technology in improving credit risk control capabilities, so as to make up their minds to introduce satellite remote - sensing technology.
In May, Feng Junyu has arranged a research trip. He will take the bank's leadership to a joint - stock bank that has launched a satellite to exchange practical insights on the effectiveness of satellite remote - sensing technology in improving credit risk control.
Currently, this joint - stock bank has integrated high - resolution satellite remote - sensing images into its self - developed post - loan risk management system for real - estate projects. By using remote - sensing satellites to capture the latest construction progress of these real - estate projects irregularly, it can achieve high - precision construction monitoring of real - estate projects across the country and effectively improve the efficiency of post - loan inspections.
There is a “little plan” behind Feng Junyu's choice of this joint - stock bank as the research object. Currently, bank executives are highly concerned about the credit risk of real - estate projects.
In the past, banks would send people to the construction site every two weeks to inspect the construction progress. Some developers with tight capital chains, after figuring out the bank's post - loan management behavior, would arrange the construction team to work one or two days before the bank employees' inspections. Once the bank employees completed the on - site inspections, the construction team would “rest”. This made it impossible for the bank to timely understand the real construction progress of real - estate projects, increasing the credit risk accordingly.
Therefore, Feng Junyu hopes that this joint - stock bank can share how to use satellite remote - sensing technology to understand the real construction situation of real - estate projects and the corresponding credit risk control measures, including strictly controlling the subsequent loan fund disbursement and using legal means to urge the developers to deliver the houses within the agreed time, so that the leadership of his bank can deeply feel that “satellite remote - sensing technology can bring better credit risk control effects”.
During the research, he also hopes that this joint - stock bank will specifically introduce the specific application of satellite remote - sensing technology in the field of supply - chain finance risk control.
“Once the bank's leadership recognizes the unique value of satellite remote - sensing technology in enhancing credit risk control capabilities and agrees to introduce this technology, we plan to build a broadband + narrowband satellite matrix by leasing and purchasing satellite data services by the end of 2027,” Feng Junyu told the reporter.
Gu Qian admitted that it is almost impossible for the bank to spend hundreds of millions of yuan to develop and launch “narrowband + broadband” satellites. Since this year, the bank's senior management has twice rejected this proposal on the grounds that the initial investment cost is too high.
Recently, Gu Qian has started to change his lobbying strategy. He emphasizes that “other banks have started to build narrowband + broadband satellite matrices. If we are a bit slower, we will lag behind in the iteration of credit risk control measures”, hoping to promote the bank to implement this technology as soon as possible.
(At the request of the interviewee, Feng Junyu is a pseudonym)
This article is from the WeChat official account “Economic Observer”. Author: Chen Zhi. Republished by 36Kr with authorization.