Once again we’re hearing the drumbeats of promises that the face of business intelligence is changing. As tool sets become increasingly intuitive and powerful and technical training and education are easy to find, some industry experts are once again heralding the end of IT as we know it, vis a vis “self-service BI.”
Readers who have been in the industry for some time realise that this is hardly a new phenomenon. BI and analytics have gone through cycles of user-focused versus IT-focused capabilities. And though the last iteration of self-service BI was reaching its apex a few years ago and has not ebbed to the extent it has previously, we as BI architects and owners owe it to our stakeholders and leaders to understand the impetus behind this change and to help prepare our organisations for the next phase of self-service BI delivery modes.
Historically, this change from IT-focused to self-service BI activity was more pronounced. In the early days of BI, reporting simply didn’t happen unless IT built the data structures and the ETL, created and formatted the reports, and even printed or emailed them to users finally, sometimes months after the initial request for data. Now that horizon is much shorter, and users are able to create ad-hoc reports and analyse cubed or unorganised data with an array of tools from their desktops or even mobile devices all in minutes or seconds!
But this begs the question of whether IT is still part of the BI equation for companies. That is, can non-IT business units gather a few more-technically capable colleagues to wrangle the data, slap some software on top of it all, and get the job done just as easily as they could have if IT had done it for them? It’s tempting to say yes, and cheaper and more quickly to boot!
There are two real problems with this conclusion however: it ignores the very real limitations of tools (which vendors might be loathe to explore), and it over-simplifies the work needed to deliver real BI and analytics capabilities that can deliver any meaningful insights, especially in complex organisations with extensive, complex, and disparate data stores.
In the first place, those who believe that tools can eliminate the need for IT entirely are either overstating the capabilities of tools, which inevitably need either heavy customization in order to work effectively on top of a company’s existing systems, or else deliver only part of the reporting landscape. Why not just buy each part then? That’s possible of course, but once again IT involvement is mandatory to make them work together, whether it’s through integration partnerships or in-house staff participation. So still IT has a place at the table. Said differently, you can buy all the wood for a house, but you still need a carpenter to build it.
And in the second place, ignoring the contribution of IT (and specifically trained data warehouse and BI technical staff) assumes that everyone will simply know how data relates and integrates, whereas nearly every BI project since the beginning of BI as a discipline has proven otherwise! It takes non-trivial skill and experience to understand how and why different data elements from various systems are related in order to deliver meaningful insights (especially cross-disciplinary ones). Moreover effective BI practice requires the kind of neutral stewardship of data integration activities to identify discontinuities, arbitrate disputes, and facilitate effective data sharing across the enterprise.
All-in-all, the hazard warnings of an IT-less BI landscape seem intriguing, and we have to admire the way in which IT’s contribution to the delivery of BI services to the organization have shifted from that of ownership to that of facilitation and stewardship. That’s exactly the kind of transformation that prepares our BI customers for revolutionary insights and capabilities.
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->