Data visualisation has been a hot topic for long enough that many companies are well into their forays developing sophisticated and powerful data presentation applications. But as data and BI architects gain greater familiarity with the data science that underpins the powerful decision-making graphics and analytics that visualisation packages showcase, it’s important to make sure that the organisation they serve understands the need for the same kind of deliberate and careful approach to data governance, accuracy, and consistency.

In other words, don’t fire your data architects just because you’ve now got some data scientists on staff.

The industry’s consensus is that the skill sets for these two roles are not the same. Data Architects Plan, supervise, govern, and sustain the organisation’s practices around data, from how it’s gathered and transformed to how it’s stored and used. Data Scientists are the modern-day data wizards, accomplishing miraculous feats of analysis and observation by applying sophisticated (might we say arcane?) algorithmic and visualisation tools to data in order to uncover hidden patterns and correlations.

The two paths are clearly distinct, and data science is just new (and shiny) to us right now. And until the novelty of data science wears off and the industry begins to standardise the manner in which data science techniques are brought to bear on business problems, it’s going to seem as though data science is the sexy job and data architecture is the drudge-work.

What’s more, the trap that companies can easily find themselves falling into is taking their data architects (they are data experts after all, right?) and stretching them into data science roles. With some classes in R and a brush-up on university-level statistics, plus a few books on experiment design, they can do the job right?

Maybe — if your data architect happens to be the analytics equivalent of a superhero.

This isn’t to imply that they’re not intellectually capable of learning the tools in the data science toolbox. Far from it; data architects need prodigious intellectual and analytical capacity in order to provide effective strategic and tactical oversight to a company’s data, a task that grows more complex and demanding each year. Instead, this observation is a reminder that just as not every member of the BI team is just right for the data architect role. And just so, not every data architect would make an equally-effective data scientist, or vice versa.

The most effective staffing decisions for the BI team take into consideration the skills necessary to be most successful in each role: extensive business knowledge, the ability to shift seamlessly between detail-oriented work and high-level data planning, and a methodical, holistic approach to data for architects versus inquisitive and exploratory instincts, a high degree of mathematical or programming capability, and a kind of relentlessness in seeking new and transformative insights for data scientists.

DataHub Writer: Douglas R. Briggs
Mr. Briggs has been active in the fields of Data Warehousing and Business Intelligence for the entirety of his 17-year career. He was responsible for the early adoption and promulgation of BI at one of the world’s largest consumer product companies and developed their initial BI competency centre. He has consulted with numerous other companies about effective BI practices. He holds a Master of Science degree in Computer Science from the University of Illinois at Urbana-Champaign and a Bachelor of Arts degree from Williams College (Mass)..
View Linkedin Profile->
Other Articles by Douglas->

Data visualisation has been a hot topic for long enough that many companies are well into their forays developing sophisticated and powerful data presentation applications. But as data and BI architects gain greater familiarity with the data science that underpins the powerful decision-making graphics and analytics that visualisation packages showcase, it’s important to make sure that the organisation they serve understands the need for the same kind of deliberate and careful approach to data governance, accuracy, and consistency.

In other words, don’t fire your data architects just because you’ve now got some data scientists on staff.

The industry’s consensus is that the skill sets for these two roles are not the same. Data Architects Plan, supervise, govern, and sustain the organisation’s practices around data, from how it’s gathered and transformed to how it’s stored and used. Data Scientists are the modern-day data wizards, accomplishing miraculous feats of analysis and observation by applying sophisticated (might we say arcane?) algorithmic and visualisation tools to data in order to uncover hidden patterns and correlations.

The two paths are clearly distinct, and data science is just new (and shiny) to us right now. And until the novelty of data science wears off and the industry begins to standardise the manner in which data science techniques are brought to bear on business problems, it’s going to seem as though data science is the sexy job and data architecture is the drudge-work.

What’s more, the trap that companies can easily find themselves falling into is taking their data architects (they are data experts after all, right?) and stretching them into data science roles. With some classes in R and a brush-up on university-level statistics, plus a few books on experiment design, they can do the job right?

Maybe — if your data architect happens to be the analytics equivalent of a superhero.

This isn’t to imply that they’re not intellectually capable of learning the tools in the data science toolbox. Far from it; data architects need prodigious intellectual and analytical capacity in order to provide effective strategic and tactical oversight to a company’s data, a task that grows more complex and demanding each year. Instead, this observation is a reminder that just as not every member of the BI team is just right for the data architect role. And just so, not every data architect would make an equally-effective data scientist, or vice versa.

The most effective staffing decisions for the BI team take into consideration the skills necessary to be most successful in each role: extensive business knowledge, the ability to shift seamlessly between detail-oriented work and high-level data planning, and a methodical, holistic approach to data for architects versus inquisitive and exploratory instincts, a high degree of mathematical or programming capability, and a kind of relentlessness in seeking new and transformative insights for data scientists.

DataHub Writer: Douglas R. Briggs
Mr. Briggs has been active in the fields of Data Warehousing and Business Intelligence for the entirety of his 17-year career. He was responsible for the early adoption and promulgation of BI at one of the world’s largest consumer product companies and developed their initial BI competency centre. He has consulted with numerous other companies about effective BI practices. He holds a Master of Science degree in Computer Science from the University of Illinois at Urbana-Champaign and a Bachelor of Arts degree from Williams College (Mass)..
View Linkedin Profile->
Other Articles by Douglas->