The Agile methodology has been touted for years as a software development approach. Since its inception, various industries have adopted Agile principles beyond that original scope.  For these not-so-traditional undertakings, each organization must apply Agile principles in the context of its organization and selectively jettison those characteristics of the method that aren’t fit-for-purpose.

Just as Agile practice has diverged from its origins, there has also been a shift in Analytics: specifically, predictive and prescriptive analytics from the Technology-to-Energy industries.  Where many industries  have utilized Big Data platforms and data science algorithms for years, oil & gas is only beginning to realize their power.


To get the most from advanced analytics, these oil & gas newcomers must adopt non-waterfall practices, fail fast, and push minimum viable products to their customers quickly.


Here are seven tips on how you can effectively deliver use cases to your consumers, modify/optimize your organization, and achieve your long-term, advanced analytics goals:

  1. Validate Adaptability – Understand that Agile is not a cure-all solution. Use cases that have well-defined work-flow steps, dependencies, and time-frames will not benefit as much as those with unknown, variable, or flexible characteristics that can leverage the iterative nature of Agile more fully.  That’s not to say the former can’t be streamlined, just that it’s important to validate the type of case you’re working with; typically, advanced analytics use cases follow the latter, and are solid candidates for delivering via Agile.
  2. Institute Executive-Level Agile – Be aware that the majority of leadership operate in the status quo (read waterfall), even while teams are operating with Agile. Therefore, ensure executive expectations for delivery are in line with the Agile approach. When your leaders are on board, you achieve higher levels of support and communication.
  3. Combat Tension – When an advanced analytics team begins cutting-edge work, tensions can rise between the traditional organization and the team. This may stem from fear of obsolescence for current technology or mere pushback on new methods for running projects. Regardless, transparency and communication can be used to convert the naysayers.  An unexpected, but welcome, benefit from sharing the goings-on of your Agile team is that productivity generally increases.  As one would imagine, with greater transparency comes higher visibility/scrutiny.  The self-organized Agile team can rise to this occasion.
  4. Time Devotion – When adapting Agile for analytics, a team must narrow its focus on a single project and will suffer if a team member’s time is fragmented across many initiatives. Sprints require resources for a short period of time (1 to 4 weeks), but that period is most productive when resources are dedicated during that time-frame.  Unlike waterfall projects that can last 6 to 8 to even 12 months, we’re looking at the 12 to 16 week range for Agile, which mandates dedication and focus during sprints.
  5. Open for Change – Inspection and adaptation are key on an Agile project, and especially so with Agile for analytics. Don’t make the mistake of creating a plan, sticking to the plan, and ignoring the output of each Sprint.  Incremental development is advantageous when minimum viable products (MVPs) are tested, commented upon, and modified per Product Owner request, time and time again.  You may have an idea of what the final product looks like, but don’t let that initial vision blind you to new information.
  6. Relinquish Decision Ownership – A Product Owner is the individual (usually in the business) who is responsible for guiding the development/delivery team to the product the business needs. This role needs relative autonomy to guide the product, provide business requirements, modify and add features based on MVPs, and promote the product within his or her realm.  Don’t micromanage or second-guess the Product Owners.
  7. Yes to the Socratic Method – The key for Agile members, especially Scrum masters and analytics Managers, is to learn to guide with questions. Self-organizing teams do not respond well to command and control leadership style.

Keeping these seven steps in mind, you can achieve VICTORY through successful Agile analytics initiatives.  As a closing thought, Agile isn’t a black box of magic where you input a use case and out comes an MVP.  It takes effort, collaboration, communication, transparency, and diligence, just like any other project.  While there are significant differences between Agile & Waterfall, these are guiding principles and should be fit-for-purpose in your organization.


Click here to learn more about our Data & Analytics practice area. Want to continue the conversation? Contact us at