Youknights
Generate unified insight from different accounting packages and revenue models.
A generic data model with a custom ingestion.
First understand the business before building a platform.
In 2021, Vincent co-founded Blenddata with two partners. After a project for investment group YouKnights, he knows for sure: Blenddata has a lot to offer the industry and its customers.
“During my graduation, the prevailing thought was that you can achieve a lot with a cool predictive model, while little attention was paid to the quality of the data. In a proof of concept, you work with datasets that have specifically been prepared for it. I soon realised that real-world practice is far more challenging than what is portrayed at university; good data collection and a good data infrastructure involve a lot of work. There was still a lot to be gained if you want to bring a good data model into production.”
“The companies dealing with data and cloud engineering did so more in the role of database administrator, which has a dull image. This can be more exciting, full of fresh energy. We want to trigger peoples’ excitement about this. With cloud technologies, hardware is no longer a limiting factor, so you have a lot of resources at your disposal.”
Co-founder | Lead engineer
Co-founder | Lead engineer
“YouKnights’ challenge was to get a universal view of the different companies in the investment pool. Each of the companies under YouKnights has an accounting package to monitor finances. As an investment company, you want quick insights without switching contexts.”
“In implementation, we discovered how complicated it is to manage different types of companies: there are labor-intensive firms, service companies, and manufacturing businesses, each with completely different revenue models. You don’t know if your initial setup is sufficient when onboarding a new company, so you keep discovering new things during the implementation.”
“For instance, on the input side, people also played a role; even if two companies use the same accounting software, their users interact with it differently. Then you end up in an organisational discussion, because every administrator has a different vision. So are you going to equalise it software-wise or organisation-wise? The solution turned out to be a mix of both. Talking to the customer about how administrative staff conduct business, gave us valuable insights. How people work affects the platform you build. In systems where manual data entry is a source, the person entering it remains the weakest link.”
“To start a project like this, you need domain knowledge. Especially with financial data, because the numbers have to balance out. For example, you need to have a good understanding of how a balance sheet and profit and loss accounts are constructed, because this has to be recreated in the backend. As a consultant, it is fun to dive into a domain. You dive into it to learn something from it again. This time finance, with another client it was batteries and solar panels.”
“Our added value is substantive knowledge about our field. As a consultant, you need to know what you are coming to implement, that will continue to be the case in the future. The ratio of engineering, hands-on code writing and domain knowledge varies from assignment to assignment. There is also project and expectation management involved in my work. We also teach clients about data management at the input stage so that they enter everything correctly.”
“During this project, we made a shift. We initially started with Azure Functions and are now working with our in-house developed and standardised Modern Data Stack. During the implementation, we found that easier to work with for ourselves. In fact, this was one of the projects that made us switch to Modern Data Stack. Building the platform has several phases.”
“In the beginning, YouKnights guided us through the business requirements. Technically, we were able to capture that better and better. You aim for a result that is not quite perfect immediately, and then another iteration follows. This approach delivered results immediately. A financial controller used to draw up the balance sheet every month. Now that happens daily and completely automatically. As a result, new insights are available every 24 hours. With daily updates, the time-to-value is much shorter. Moreover, we have integrated data quality modules that provide insight into how accurately administrators handle figures, such that errors surface sooner.”
Co-founder | Lead engineer
Co-founder | Lead engineer
Growing in engineering and consultancy.
Working for SMEs and multinationals.
Blend your benefits!