Voyance Vision is a Document Management System (DMS) that uses computer vision, machine learning, and artificial intelligence to extract, store, manage, and retrieve documents for businesses. With the global demand for DMS reaching billions of dollars, companies need efficient solutions for handling daily documentation. Voyance Vision addresses common challenges like low image quality and poor scanning while also automating processes to save businesses time and resources.
As the lead designer on the project from February to November 2021, I was responsible for driving the overall design direction and creating all significant deliverables. This included creating prototypes to visualise and test different design concepts and research documents to inform the design process. I also developed a design strategy document that outlined the overall design vision and principles for the project, which helped to guide the various teams and ensure that everyone was aligned towards the same goals.
In addition to my design responsibilities, I worked closely with the diverse in-house teams to ensure that the design elements were effectively integrated into the project. This included collaborating with project managers to ensure that the design was aligned with the project roadmap, working with data scientists and engineers to ensure that the design took into account the technical limitations and requirements, and working with frontend and backend developers to ensure that the design was implemented accurately and efficiently.
Throughout the project, I also had the opportunity to work directly with the CEO, who is a software and machine learning engineer. This allowed me to gain a deeper understanding of the technical aspects of the project and ensured that my design work was grounded in a strong foundation of technical knowledge and expertise. Overall, my work on the project helped drive the project's success by ensuring that all design elements were carefully considered and integrated seamlessly into the overall project plan.
During the design process for Voyance Vision, I collaborated with project managers to research and understand document handling challenges businesses face, particularly in the fintech sector. This research led to a solution to streamline KYC processes for fintech businesses, addressing speed, accuracy, automation, cost, and security.
The fintech space in Nigeria is rapidly growing, and businesses must process numerous documents daily to comply with regulations. Manually extracting data is time-consuming, error-prone, and expensive. Voyance Vision uses machine learning algorithms to automate data extraction, making it faster, more accurate, and cost-effective.
The project involved a cross-functional remote team, and I coordinated their efforts to achieve our goals. I divided the project into two phases, focusing on improving speed and accuracy first, then reducing cost and increasing automation. I utilised rapid prototyping to gain feedback and ensure alignment across the team, resulting in a high-quality solution for our target users.
Voyance Vision allows companies to train their models or use our pre-built models that our Data Engineers have trained. Pre-trained or prebuilt models help you extract essential information from documents without gathering data and then build and train your models.
Using pre-trained models is an excellent option for companies that need to conduct an analysis quickly and don't have the time or resources to build their models.
Documents can be uploaded easily from local and cloud storage without disrupting your workflow. Data extracted can then be downloaded in various formats.
Vision allows its users to annotate documents - this is a process of labelling the data accessible in various layouts like video, text, or images. This process also improves accuracy.
Workflow integration reduces the rate of human interaction, instructing the system on what to do to extract data during certain events.
During the development of the Voyance Vision prototype, we enlisted the help of a few data scientists, data engineers, and machine learning experts from outside our company to test the prototype and provide feedback. One of the key insights we gained from this testing was the need to provide an option for users to upload PDF documents. We had initially assumed that all documents submitted would be in image format, but we quickly realized that this was not the case. Many users and businesses had their documents saved as PDFs, and not being able to upload these documents would significantly limit the usefulness of the solution.
Another feature that was requested by the testers was the ability to use the trained models as an API. This would allow other organizations to integrate the Voyance Vision technology into their own systems and processes, expanding the potential reach and impact of the solution.
The feedback we received from the testers was invaluable in helping us to improve and refine the Voyance Vision prototype. It allowed us to quickly iterate and include new features and capabilities that we had not originally considered, making the solution more useful and effective for our target users.
On August 11th, 2021, the project was successfully launched to the delight of the entire company. All of the hard work put in by the team had paid off as the launch went smoothly.
The implementation of Vision reduced the need for manual intervention by 80%, resulting in a 70% reduction in the dependence on human resources and a 90% reduction in processing time. This advanced system was able to efficiently and accurately process documents, eliminating any errors that could cause delays or financial losses. The quick and accurate processing of documents through Voyance Vision delighted customers with instant approvals and activations.
Fraud detection: Vision can identify fraudulent documents, fake images, and other forms of invalid data. Vision can also detect unstructured data such as messages, claims, and customer feedback and then notify humans about incidents of suspected fraud. In addition to these benefits, Vision improves forecast accuracy, allowing loss control units to cover more cases with fewer false positives.
Vehicle damage assessment: Assessing damage to a vehicle takes much time and human resources. Vision can scan and analyse the damage to a vehicle in seconds. This allows the assessor to determine which vehicle elements should be fixed or replaced based on information from manufacturers.
Claims management: Machine learning and artificial intelligence are revolutionising the insurance industry, allowing businesses to automate claim management, detect incremental fraud patterns, and minimise false positives that differentiate between real and fake data points.
The major challenge was identifying the specific solution we would offer at launch. Because computer vision is so robust, you can do a thousand things with infrastructure like Vision. So my work was to cut out many parts to create what our “right-now” clients need. Also, what we are building, an AI infrastructure, is more complex than your everyday tech product. I had to learn fast and have a clear picture of the product to help the team understand it. I am still doing the learning aspect because people are always coming to me with questions, and I should have the answers:- Jennifer Okafor, Product Manager.
The disparity in understanding of the problem was also a challenge. This was tackled by sharing concise product documentation across the team and having regular sessions with the product team to avoid assumptions. Formalisation of the engineering tasks, going from product description to an engineering breakdown. Architectural decisions had to be made, options had to be weighed carefully. - Temi Babalola, Senior Software Engineer.
Working on this project, I did not have background knowledge in OCR technology, Computer vision, Machine learning, and Artificial Intelligence. I had to study and research to develop an intuitive design that solved the problem. Working with the team on this project was terrific. - Sebiomo Aanuoluwapo, Product Designer.
The information in this case study is my own and does not reflect the views of Voyance. I have intentionally omitted confidential data where necessary.