How to make Data Lab profiles work together


You may have experienced how complex it can be to have all the people involved in the Data Lab coexist and – even harder – work closely together on a common project.

Results are not what you expected, deliverables lack quality, prototypes are not being launched… Don’t freak out yet, it happens!

Why is your Data Lab unproductive?

You should consider your Data Lab as an independent startup operating within the firm, whose only agenda is to create value from data.

Managing this startup can be hard for multiple reasons:

  • continuous technological and scientific breakthroughs
  • unclear roles and missions
  • technical and cultural differences that affect communication


You may face many troubles, and not only internal ones like those mentioned above :

  • A tug-war between constraints, desires and needs from all parts (Business, IT, legal department)
  • The developers, the Business part and the Data Lab are not on the same timeline / rhythm
  • Data Scientists, Data Engineers, Data Architects and DevOps engineers are stuck in business, IT or organization related constraints: they don’t have the proper access to data – sometimes they don’t know if the data they need even exist – and often are not familiar enough with the problematics the project is supposed to address


Knowing who you are dealing with (Data Lab profiles), adopting the right organization to ensure communication and collaboration, are key to overcome all the troubles you may face. Another tip: don’t let the business side be excluded, it is crucial to involve them from day 1 as the business vision should be the main guideline to your project.


The main project stakeholders

That’s a fact: all the people involved will have to interact at some point. Some of them will even have to work closely together.

The Chief data Officer

The Chief data Officer (commonly called CDO), makes sure the company is data driven. His role is mostly strategic. He knows the different profiles involved in a project, their mission, and helps them collaborate toward a common goal.

In practice, he gets an input related to a customer need and has to share it with the different parts of the team. In order to do that, he has the ability to translate business needs into technical features to implement. That’s why you may hear him talk about budget and business needs in the hallways.

Our piece of advice: even if he is familiar with data technologies, he is not necessarily up to date on every new feature. So, as technical as he is, don’t push him.  

The Data Scientist

The Data Scientist lies at the heart of the data team – both delivering models to Data Engineers and make it understandable to clients.

He needs to keep in mind what the Data Engineer wants and what the customer asks for, which he learns through the CDO.

Our piece of advice: you can talk about Python or R, neural networks and how to present results to clients, but don’t dive too deep into business chat or data architecture because he may not know much about that.

The Data Engineer

the Data Engineer masters the technical tools that are used in data processing.

He closely works with the Big Data architect on infrastructure management.

In practice, the Data Engineer provides an easy access to data to the Data Scientist and helps deploy his models.

Our piece of advice: try not to talk budget or business too much, but don’t hesitate bringing up databases or APIs.

The (Business) Data Analyst

He is the link between the Data Scientist and the client and he translates customer needs into technical instructions.

As a business communication specialist, he also shows and explains results to clients.

Our piece of advice: he sometimes asks the Data Engineer for dataset analyses so he knows one thing or two about data processing. On the other hand, avoid being too technical.

You could easily simplify or define clearer roles depending on your needs, your company’s maturity or size. Remember: making these profiles work together is the goal.


Set up the appropriate team organization

Start by asking every team member what he would ask the client to have the clearest view on the project and how to develop it – and share the results with the team.

Choosing tools and following a common guideline will be much easier with a global view of the project. All you will have left to do is define the right framework and start developing.

The Saagie Projects & Analytics feature allows you to centralize the technologies you chose and make it accessible for the team.


Remember that roles are not locked and it’s important for every member of the team to be interested in their fellow co-workers’ mission. Focus on understanding the challenges you may face before trying to find solutions, use the right vocabulary and learn to ask the right questions.

To conclude, find a tool to automate and ease those interactions, you will be more efficient in developing the project both technically and in accordance with the business vision.


Can we use some cookies?

This site uses cookies. An explanation of their purpose can be found below. To comply with new EU regulation, please confirm your consent to their use by clicking "Accept". After consenting, you will not see this message again.

Know more about tracers