Thirty years of collective agreement data, ready to use for any advice
AWVN holds a unique knowledge position in the Dutch labour market: more than thirty years of collective labour agreement (CLA) data covering thousands of organisations and sectors. Appfront developed an application that allows subsidy advisers to search this full history quickly and in a structured way. The result is an advisory process that is faster, more consistent and better founded.
75% faster
Advisers answer CLA questions on average 75% faster than with the previous way of working, as searching, filtering and substantiating now come together in a single tool
10,000+ CLAs
The tool gives access to more than 10,000 CLAs spanning 30 years, so advisers can for the first time structurally draw on the full market history in their advice
A data file that could no longer bear its own weight
As a leading CLA advisory organisation, AWVN holds a unique knowledge position: more than 10,000 CLAs over a period of thirty years, with detailed metadata on sectors, terms, pay agreements and employment conditions. This knowledge was kept in an Excel file that had grown over the years into an unmanageable whole. Searching took time, results depended on who was searching and how, and the risk of errors increased as the file grew.
With this platform, we now reach a substantiated conclusion within seconds that we can use directly in our advice.
Understanding how advisers reason
The core of the project was not building a search engine, but understanding the advisory question. How does a CLA adviser formulate a statistical question? Which parameters are always relevant, and which vary by situation? That analysis led to a structured question builder, in which advisers build a question from fixed elements, without needing technical knowledge or depending on how a colleague might phrase the same question. Working closely with the advisers, the tool was built and refined iteratively, so the interface matched the way CLA expertise is applied in practice.
Structured questions, traceable answers
The application gives advisers access to the full CLA database through two entry points: direct searching and filtering on organisational characteristics, or building a statistical question in natural language. For every outcome, the system shows which CLAs and records the answer is based on, so conclusions can always be traced back. This is essential in an advisory context, where substantiation is just as important as the answer itself.
Our approach in 3 steps
Research & Strategy
We begin with user research and a clear strategy to determine the right direction.
Data modelling and AI
In collaboration with Data Science Lab, we set up a data warehouse and indexed all collective labour agreement documents using a custom-developed AI model.
Development & Launch
Agile development with weekly updates, followed by a careful launch.
A new platform without losing trust in Excel
Advisers have been used to working with Excel as the basis for their advice for many years, so keeping that trust intact is essential. That is why the platform is built around tables, and users can always see the underlying data.
Framework
React | Python | Docker | LLM | Microsoft Azure
Consistent advice, based on the entire market
Where advisers used to rely on their own search routines and interpretations, they now work from one shared tool with a single data source. Questions are answered much faster, but the biggest gain is consistency: every piece of advice is based on the same data and the same methodology.
Let's get started
Together, we create smart digital solutions for your organisation's challenges. Not rushed, short-lived products, but thoughtful, high-quality solutions built on a foundation of UX design and technological expertise, so your organisation is ready for tomorrow.
Arrange an introductory callOr explore our case studies, blog posts or get to know our team.
