Envisioning a framework to realize the business case of a data project

Getting people on board

Once you have established a culture where successful dataprojects are überhaupt possible, you can start planning the journey. If you are tasked with setting up the data function in an organization, or you have an idea for a project in the data sphere you will have to get buy in from the rest of the organization to go further with this.

In an ideal world everyone is on board from day one and you have enough time and money to realize all the ideas you can dream up – not sure if that really is the ideal world though, keeping in mind Parkinson's law that states that work expands to fill the time available for its completion. But even in an organization a with clear purpose and a culture of empowerment and accountability, people will have to agree on how to spend the budget. Moreover, it could very well be that the outcome of the data project will significantly change the way certain things are done in the organization.

This means you have to convince people of the value of data and you have to think about how you can communicate this value in the most convincing way. In some way, this could be approached with a change management mindset: you want to change the way the budget was distributed the last few years and you want a bigger cut to realize the ambitious vision you have for the data projects you have in mind. You also may want to question and change existing processes in order to make the more efficient or even redundant. These type of changes affect peoples jobs up and down the chain of command and you need to be aware of this to communicate with the necessary tact about the project you would like to see succeed.

Source of the picture: cornerstone.rutgers.edu

In a best case scenario you will have the typical change management curve: early adopters convince laggards. In a worst case scenario, in a culture that allows for it, you will have backstabbing and turf wars. Whatever scenario you will get into, a certain degree of office politics is unavoidable.

A compelling business case

A solid and compelling business case is an indispensable communication tool in every scenario: show the value. How will the organization benefit from the data project. This can be higher revenue or margin, lower costs or risks, a better reputation or service level. Or it can be the implementation of a mandatory regulation to be compliant with legislation, and so on. The business case provides all the necessary information to make the job of the decision maker possible.

A lot of articles on why data projects fail that we read, stress the importance of a business case. We suspect a lot of progress can still be made in this area: the data people cannot work on their island tweaking models, they have to know very well where the value for the business is coming from and it may very well be that this is not in the first place by developing a state of the art machine learning algorithm. You have to learn how to walk before you can run.