While I agree with the suggestion proposed by Davenport regarding centralizing analytic talent, I believe it depends on the organization. Due to each organization’s unique reporting structure, talent must be focused in a manner best fit for the intended processes.
Some of the pros include:
Centralized point of contact
Creation of bottlenecks for IT critical functions
Some of the cons include:
Work being done in silos
Duplicate employee counts working on similar project types
Me personally, I have seen matrix-style reporting at many clients and while this creates silos, it allows projects to move with velocity and flexibility. I believe balance is difficult to create in most situations, because there are so many variables to focus on – for example, talent, training, projects, proper tools, incentivized learning - the best thing an organization do is work on all of these aspects at parallel on a continual basis.
The organization I work for is large, thus, it is divided into smaller organizations based on service, location, industry, and specializations. A centralized analytic talent group exists for all the business units, products, services, etc. This is simply because the depth of data requires an acute focus in an area. Organizations who sell less products, on the other hand, could benefit from a centralized analytic talent base for the biggest reason of cost savings. These savings come from not duplicating work and talent.
An article in the Harvard Business Review addresses why businesses aren’t getting value from their analytics teams – and suggests to focus on the four areas as a starting block: (1) stick with simple models, (2) explore more problems, (3) learn from a sample of the data, (4) focus on automation of tasks. To me, a centralized analytic team would make doing all four of these for each department very difficult.
Resources
Davenport, T. H., (2013). Enterprise analytics: Optimize performance, process, and decisions through big data. Upper Saddle River, NJ: Pearson FT Press.
Veeramachaneni, K. (2017, February 22). Why You're Not Getting Value from Your Data Science. Retrieved from https://hbr.org/2016/12/why-youre-not-getting-value-from-your-data-science
Comments