PandaTS Brings A.I to Document Verification with New Google Collaboration
Clunky and time-consuming document verification procedures have been a persistent pain point in online trading for both brokers and traders. Brokers would like their on-boarding processes to be frictionless, ideally tying up as few members of staff and as little time as possible per account. Traders normally want to get down to the business of trading as quickly as they can after completing the registration form.
In our experience, one of the main causes of bottlenecks in the document verification process is when outmoded CRMs are being relied on to do more than they were originally intended for, and when different components of a brokerage’s infrastructure (for example the client area and CRM), don’t play well with each other, requiring time-consuming workarounds for back office and compliance staff. Another common problem is when CRM navigation is not intuitive or user friendly, forcing staff to dig around back and forth through different menus in order to get at what they need. Having been repeatedly called in over the years to overhaul brokerage registration and verification workflows, a few years ago we decided to solve the issue once and for all by creating our own CRM that would optimise this specific process and many more besides.
Bringing AI to Document Verification
The feedback we’ve received over the years has been very positive and has helped us to improve our systems by alerting us to how they perform in real-world scenarios. Recently, with the huge interest in online trading during the global lockdowns, the industry has observed that beyond a certain volume of incoming traders, there’s a definite upper limit to how many applications and documents human staff can physically process throughout the course of a working day. For this reason, we decided it was high time that hard working back office and compliance teams receive a non-human helping hand in the form of artificial intelligence.
In partnership with the good people at Google, we’ve harnessed Google’s incredibly powerful Google Vision AI and put it to use in our Panda CRM. Our CRM verification module has been updated to make the process of finding, reviewing and comparing verification documents against CRM entries much smoother and easier for human members of staff. This allows them to move through records in a much quicker way and not have to perform unnecessary actions. We have also improved our system of triggers for automated document requests. But that’s not all. Back office and compliance teams using our software now also have Google Vision significantly lightening their workload by performing various time-consuming duties.
Final Thoughts
We expect the new verification module to reorganise the way in which our partners handle this aspect of their daily business. Firstly, we anticipate a massive reduction, if not the complete elimination, of uploads that have to be rejected for poor quality or improper cropping. Secondly, we believe that the optimisations we have made in terms of how records are made available to back office and compliance staff, will make the process of verifying large numbers of documents much quicker, easier and far less tiresome. Finally, the groundbreaking automated reading, comparison and scoring of incoming verification documents, will completely reinvent how online brokerages handle their KYC/AML commitments. Furthermore, using machine learning, our document verification module will actually get smarter over time as it is trained on more and more uploaded documents. In other words, the AI’s accuracy will increase the more it is used, to the point where hopefully in the future it can be relied upon to completely shoulder this burden for its human colleagues.
For free consultation
Request a Call