Dorothy had to overcome a lot of challenges following the yellow brick road to reach the Emerald City in the land of OZ. Similarly, organisation’s looking to gain the benefits of putting Operational Analytics (OA) into action undergo a challenging journey.
OA is increasingly becoming a key point of growth and investment for many businesses. Why? Put simply, OA focuses on improving an organisations’ existing processes. It does this by embedding valuable insights to improve decision making for those involved in complicated business processes. These insights are derived through various data aggregation and advanced analytics techniques to automate the bottlenecks and manual touchpoints where viable. The investment in pilots and proof of concepts now needs to be realised and driving these insights into the value chain is essential.
Whilst the benefits of OA are evident, businesses often encounter hurdles when it comes to leveraging it as a tool, but now, more than ever, embedding OA into your business is critical as we are faced with an unprecedented volume, variety and velocity of data. Confidently leveraging this data can help you to understand insights, create a single point of truth on project performance, answer key commercial questions, and help predict and react to future trends.
As operations develop ever-increasing data sets through the deployment of sensors and new technology, leadership teams are discovering they do not have the skills or tools to create insight that helps them make informed and confident decisions. In some cases, leaders are relying on intuition and experience to drive decision making, which may not be the best course of action for the organisation.
So, how can operations prepare? What follows are five key challenges I believe most organisations will need to overcome when they plan to embed insights into operational processes for business benefit.
1. Bringing everyone together: stakeholder engagement and buy-in
To survive in a challenging business environment companies must remain relevant. This requires regular interaction with stakeholder groups. A robust stakeholder engagement model is vital for companies to understand and respond to legitimate stakeholder concerns. Most importantly, there must be alignment on the analytics strategy of the organisation.
All stakeholders need to be engaged to achieve success. Teamwork among a cross-section of departments is essential for evaluating, championing and implementing analytics-driven initiatives. When stakeholders are informed and in-sync they can act as advisors who explain what analytics innovations are available to support the business.
2. Emphasis on iteration and managing expectations
Managing client expectations often focuses on relationships, deliverables and deadlines. But managing expectations within OA is much more complex - it often means preparing the client for an "it might not work the first time" scenario. It may not meet the initial wants of the business, e.g. "I thought Artificial Intelligence (AI) could tell me when to shut down my factory for maintenance, but it only has information on the pumps." Analytics is an iterative process and requires continuous feedback and training to keep the analytical models relevant and scalable.
Another reason that these expectations are not met is that users need to be provided insights in a way that they can easily use to make better decisions. If users are not able to embed these analytics into day to day operations, little value will be provided and adoption of these analytics won’t happen.
3. The necessary in-demand skills: industry knowledge and capability
Globalised, interconnected organisations create a level of complexity that requires sophisticated real-time decision making. However, many teams are not armed with the level of insight required to make confident decisions.
Whilst AI and variations of intelligent software have the potential to generate valuable business insights and automate actions based on the results, it is more so the right people who implement and drive the technology that is needed.
Without the right people who possess the required knowledge and capability, and without clear goals for how to best implement and use the technology and information, companies risk spending large sums and seeing small returns.
The skills market for AI, cloud architecture and data engineering is in such demand that the average tenure is about 2.6 years according to a recent survey by the recruitment firm Burtch Works. I hear this regularly from my peers in other consulting firms and clients alike that this is one of the biggest challenges they all face, finding good people and keeping them. But while there is a shortage and demand in this area, it gets harder when looking for these skills with experience in asset intensive industries. The asset data scientist, with AI skills and the subject matter expertise in infrastructure assets is one such role that is difficult to fill. Working with universities is important to provide the guidance in what the industry is needing. It is also essential to be working on a diverse and challenging range of analytics projects to keep the team engaged and interested.
4. Starting small: scaling analytics
A common approach to delivering OA is to start small with a Proof of Concept (PoC). This approach allows for an early view of what can be achieved and allows for agile development going forward. Getting traction to deliver business value at scale and move beyond the small pockets of success is challenging for most organisations. There are numerous reasons for this such as not having the right foundations in trusted data, technical infrastructure and skilled people to be able to scale across the enterprise.
5. Under pressure: quantifying Return On Investment (ROI)
In today’s competitive environment, businesses are under pressure to reduce costs while improving the effectiveness of their operational delivery, but each organisation’s ROI measures should be as unique as their overarching business strategy. There should be well-defined use cases with input from the business that are able to easily outline both the tangible and intangible benefits. From there, defining distinct measures to a set of agreed success criteria is essential to helping quantify ROI. These criteria might not always be related to the bottom-line, but still provide a quantifiable benefit. Implementing new processes and tools in stages will also help measure success as it will allow the teams to gather and present meaningful feedback, deliver improvements and advise changes. And by being able to show the true value of these early operational analytics projects, it makes the follow on projects so much easier.
There is significant business benefit to be gained by successfully embedding analytics into the operational processes. Those that can set the expectations, take the business on the journey, and chip away at the larger challenges in a repetitive and agile approach will be successful. They must also be able to engage the key stakeholders, gather the right capability and execute the plan. So similarly with Dorothy who had many challenges to traverse across OZ, she knew where she wanted to go and who she needed to bring on the journey to get there.
Stay connected to our expert insights