Perhaps one of the biggest untapped resources owned by trade marketplaces is their potential to monitor and even predict trade trends. While most sites do not support online transactions, trends in inquiries through their platforms and shifts in the listed products can tell us a lot.

On a macro basis the data could be mind to see where higher volumes of transactions might occur; both geographically and by product or product category. A basic report might read something like this, “We have seen inquiries for pakistani leather goods double in the last month.” Unlike the port-to-port data available from sources like the UN Comtrade database, this data could be mined and made available in real-time and in advance of the transaction. In other words, the trend could be observed as or before it occurs not while it is underway or after has ended.

On a micro level sites could also refer individual users to new opportunities as they happen. Real-time reccomendations might be possible, “Dear John, we noticed you sell wood flooring, there have been 100,000 new inquiries for your product from Tunisia in the last month”. It can be hard for manufacturers (especially those from developing countries) to identify new markets or regions to sell in. John probably never knew he would become the wood flooring king of Tunisia, he may never have even heard of Tunisia.

Admitedly, the data collected by most marketplaces is less than perfect. To have any degree of accuracy data fields like company size, years in business, number of product units available or requested, and a very well managed product category system are all required. Sites have to be able to differentiate between a million inquiries for ten units and one inquiry for a million units.

Ideally these sites would also be able to monitor actual transaction volumes and values. Monitoring such numbers doesn’t necessarily require managing the actual transaction. If you give users solid tools to better manage their sourcing/sales online then they will have incentive to upload/input such information. At worst, occasional surveying or sampling of actual transactions would increase the accuracy inquiry-based predictions. Port data from sources like the Comtradedatabase could also be integrated and used to enhance predictions as well.