The toy in fashion needs to be supported and developed carefully to avoid it becoming a broken toy

In the age of data, dashboards and data observatories have become essential decision-making tools. Yet, like broken toys, we often find outdated platforms and abandoned projects. Creating an effective data observatory is not just about designing attractive graphics; it requires a solid strategy, from understanding the business to sustaining it over time. In our experience, we have identified six keys to a successful observatory:

1/ Defining its purpose: The first step in developing an observatory is to thoroughly understand the business and its objectives. This involves answering key questions: What is the dashboard trying to achieve? Who is it for? Who will manage it? These answers will define the functions and purpose of the dashboard. It is not just about collecting data, but about delivering tangible value, whether that is supporting strategic decisions, informing the public or facilitating operational processes.

2/ Sizing the scope: The Observatory must be properly sized. This means defining the areas it will cover, the products it will generate and the services it will provide. Will it be a real-time monitoring tool? Will it produce detailed reports on a regular basis? Defining these elements avoids overloading the dashboard with irrelevant information and ensures that the dashboard is useful and manageable.

3/ Working on visual structure and navigability: A good observatory combines clarity and functionality, so it should include drop-down menus, filters and navigation buttons that are designed to be intuitive. In addition, accessibility criteria and help buttons should be considered to make it easier for users to interact with the platform without feeling overwhelmed. A clean and well-structured design can make the difference between a dashboard that is used and one that is abandoned.

4/ Use external data (with caution): Not all data need to come from internal sources. The increasing availability of open data makes it possible to enrich observatories with external information. But be careful because this requires special attention in terms of data quality and governance to ensure that external sources are reliable and sustainable over time. Otherwise, the system could collapse if one of the external sources stops providing data.

5/ Temporary sustainability: As we said, finding abandoned dashboards is like finding a broken toy. To prevent this from happening, an observatory should not be designed to offer a static photograph, but should be updated and maintained over time. This would imply endowing it with certain economic, human and technological resources, institutional backing and a commitment to guarantee its continuity.

6/ Flexible planning and openness to innovation: The success of an observatory depends on planning that is both robust and flexible. From launch to maintenance, milestones need to be set and processes automated. However, the world moves fast and needs to change, so it is important to leave room for innovation and unpredictable opportunities. Incorporating tools such as artificial intelligence for advanced analysis or dynamic calculations can be key to keeping the observatory relevant.

In our work on the creation of observatories, we have come across some of them that have served as inspiration, such as:

Netherlands Climate Viewer: is remarkable for its volume of information and structure, although its size can be overwhelming at first.
Daily Mortality Monitoring System (MoMo): simplifies complex information with pre-calculated values designed to provide consistent, real-time data updates.
Urban Accesibility: shows urban data integrated into a 3D map with aggregated values by area and street, very useful for assessing inequalities and facilitating decision making.

A well-designed data observatory is not just a technological tool; it is a strategy that links data to decisions. To do this, it must be aligned with the objectives being pursued, accurately scoped, structured to facilitate user navigation, and sustained over time. In an environment where data is the new oil, the key is to turn its potential into real value, betting on the creation of an audiovisual and dynamic platform that drives innovation and knowledge.