Crisis of Sponsorship for BI teams.

 Crisis of Sponsorship for BI teams.

There is an accelerated explosion in the domain of data. While, a decade ago, BI was central to the data strategy of an organisation, there has been a mushrooming of data teams with their own mandates. We have seen the rise of Big data teams, Analytics teams, data science teams, data platform teams and many other flavours of data teams. While earlier one could easily call oneself a data generalist, that is not true any more. 

While there are decentralising forces there are also centralising forces where many organisations have moved to calling all data professionals as data scientists. 

This has created a fog of war providing many opportunities and pitfalls. Many professionals have moved to largely similar positions with different designations, while a number of skills have lost their currency. These clashes are visible in data warehouse vs data lakes or RDBMS vs python libraries, where concepts that were considered essential skills not so long ago are gasping for breath. 

A core problem that this vagueness creates is the crisis of sponsorship. Data is not the end in itself but a means to it. Thus data projects and teams need strong sponsorships to sustain the development cycles and bloom and lead to fruition. This sponsorship can be provided only with the vision of what the results of this investment is going to be. Interestingly this vision must come from the business side for whom data understanding may not be a core skill. This the task for communicating and convincing this vision, falls on the shoulders of data teams themselves.

A bit hit during the process has been taken by the BI teams. By traditional definitions BI is responsible for the procedures and technologies by which data is collected, stored and processed into insights and information. This looks like a very broad definition but in reality the people implementing these are only human. They are prone to drawing imaginary lines of what they think their responsibilities are. For example, when big data technologies were on the march, many BI teams made the mistake of sticking to traditional RDBMS and ignoring the contributions that Big data could make to the BI processes. Similarly, with the advent of convenient python libraries, the importance of executive dashboards took a hit. 

The common theme separating the success stories from failures are what are traditionally called the three Vs “velocity, variety and volume” with an added factor which I would like to call “vision”. Vision caters to the way a BI team looks at its mandate and utility. A team that keeps its vision inline with the demands of larger business will constantly expand its horizons taking in new roles and responsibilities to stay relevant. If they fail at this, then there will be a mushrooming of new teams that will happily grab the ever expanding needs. It is in this context that we return to sponsorship for BI team.

Sponsorship is a handshake between two different ideas. On one side are the BI teams offering projects and ideas for sponsorship. If the projects and ideas on offer are stale, there will be few takers for them. They need to be inline with the expectation from the other party. 

The other party is the business who have raging fires to manage and critical questions to answer. They can generally be found to be willing to try out an idea if it offers to help them meet their goals. In a situation where offers are lacking they may back the idea with most promise even if it's exaggerated. This can often be their downfall. The business needs to have certain basic abilities to evaluate the promises for their ability to back it up. 

If you may imagine these two in the form of a vast space, there is a goldilocks zone where things thrive. This is the zone of correct sponsorship to viable ideas.

A stretch is required from both ends to reach this goldilocks. For example from BI team, moving beyond traditional warehouse and data mart concepts onto dealing with data blobs, clickstreams, and unstructured data in general. Similarly, in their tool set expanding ETL to include Big data technologies and expanding traditional reporting to integrate with python and R to make analysis and integrated experience. And finally taking the next lead to Managed self service solutions which stretch expansively to exploit traditional ETL and analytical resources as a whole to create business centric solutions.

In summary, no data tram can thrive without sponsorship from the executive team. All data teams and especially BI teams need to redefine themselves, focussing more on business than on technology to be able to thrive in the new era. Only then would they be able to maintain their relevance in a new data world.