Analytics - An Evolution from Inform to Insight

I would like to discuss in this article a viewpoint of the effect of maturity of analytics and reporting in a business context and how the maturity changes I have implemented have led to significant business improvement. Also to discuss the continuous struggle as a business leader of information availability and beyond that actionable information.

When researching new innovation in Supply Chain often transparency in finance and performance is within the Top 5. This being the case it is usually a very hard task to achieve and requires fundamental investments.

I have found that information, in general, is “available” but is either in a state of inaccuracy or provided in such a way that it does not support effective decision processes.

Let me explore for a moment my view on what this means. Firstly, as a principle, to be able to advocate change or to make decisions, information needs to be presented in a way that is clear, and all agree that the “number is real”. It is the first challenge faced by most leaders. I have lost countless opportunities and time discussing the accuracy or believability of the number versus spending value adding time what to do about it.

Often leaders and their people are lost in an endless cycle of providing numbers out of context, numbers customers and stakeholders do not agree with, subsets of data unknown to the receiver or become unclear on relevance. Time wasted on having multiple sources of data, numerous similar reports, none of which can be easily verified or reconciled. This leads to even greater confusion, mistrust and indecision. Worst case scenario is when these sources are used to decide in good faith and trust only to find latter flaws have generated critical business issues.

Often these problems extend beyond an area within a company but can be the same for all stakeholders, suppliers and customers. All caught up in this cycle of poor information.

In simple terms, if I use a practical example in business, often we are presented with information in the form of a report or a chart. Practically this is not useless but limited in its usefulness. A simple chart or report can display a snapshot of information. Often the report will be a result. Relevant to a business performance requirement but a result of actions that connect to it.

I have a management philosophy that drives the need for better analytics. A simple approach to knowing. KNOW where you are, KNOW why, KNOW where you are going. In other terms What was the result? Why did the result occur? What can be done about the result and what will the next result be? Simple but powerful in execution.

If we relate this back to the example, often the report will answer the first KNOW only. It displays the result relating to where we are. If it is trusted, that is ok, but in reality, often the result is challenged from an accuracy or relevance point of view. Time is wasted trying to verify the accuracy and relevance. For this example, let’s assume that it is accurate.

The use of this information in context generates a fundamental question. If I understand that is the result, why is it so? The Second KNOW. “Why” is such a hard thing to answer sometimes. Reflect how long you and your teams invest in trying to nail down what is the cause of the result. This is where the complexity and investment truly begins. A tragic cycle of creating report after report, massive amounts of analytics and data to determine an answer. Sometimes it is found, or as a fallback a position of assumption based decision making is implemented.

The key problem is the investment and return. And the ratio of time connected. I want to be spending my time with a future focus not trying to figure out where and why I am where I am.

If this is a problem statement that what is a potential solution?

The way I have addressed this is by focusing on the three KNOW process and ensuring that the analytical processes and reporting structures are built around it. But there are a few things I need to be confident around beyond the accuracy of data.

The key to the problem is understanding drivers and how they can empower what to do with a result. Understanding all the different variables and the relationship of the variables to each other starts to create a map. A map of the data requirements, analytical work and reports to understand the why. Creating these up front will create an environment where the key variance to expected outcomes can be easily and quickly understood.

Often subject matter experts fall back on their knowledge/experience and rightly so. The problem though is the KNOWING. When discussing possible causes of a result with my management teams, they can construct valid and reasoned arguments to why a problem or result occurs. All based on some factual input and solid assumption analytics. The problem is they are unable to define clearly, or scientifically back up the assumption. Insight analytics breaks through this and any needs for the assumption is removed.

Analytics and reporting maturity driven by a 3 KNOWS principle and purpose is the way to solve the problem. By investing in not only the appropriate forms of analytics but taking an approach to how to create an information set in the right way can break the cycles and drive improvement. Analytics maturity can be mapped as below, starting with data availability and management. As each level of maturity is achieved the empowerment of the business and yourself as a leader grow exponentially.

Data

Data is the first maturity level and is the foundation of all that follows. The saying “crap in and crap out” is never truer than with data. The trouble is I hear this statement far too often. Even to a point where an acceptance of poor data becomes the new norm. To be able to begin on the maturity path the first actions are to ask questions and to plan/implement change using a master data management approach.

Where is my data?

What is the accuracy of it?

What are the master data management (MDM) processes and compliance in place?

Are they sustainable and consistent?

How often and I checked the data sources?

What is the security requirements I need to be aware of?

How is my data structured and how could it support the analytics/reporting I need?

What are the gaps and how will they be addressed?

Don’t get trapped in it though. Data management will always be there as a requirement, and sometimes you can be trapped in not using it at all based on the fact you are not 100% confident in the accuracy. The trick here is to know what is inaccurate and developing ways to use it and take into account the weaknesses.

For instance, if you have verified 80% of the data then use that as the base. When using any generated analytics declare the data in two pieces. Verified and unverified data. Then you can make decisions and actions related to the verified and use things like trends to apply to the rest. This will also begin to build the trust with stakeholders. Defining the data, you and they can trust and the data that is yet in that state or usability.

Ensuring accurate data sources, engaging partners, internal stakeholders and customers in the process of data cleansing/data management is the first fundamental step to mature analytics. Creating accountability and performance management around data management can start to build trust in the sources and rectification of the weaknesses. MDM is the basis of an effective organisation and is only now becoming critically important for companies to invest in as they realise the connected value and profit. Far beyond that of its cost/investment. Some companies are even raising it to a C-level employing a Chief of Data as part of the executive.

Businesses that do not focus on data will always bleed profit and create both operational and business management inefficiencies as well as creating poor employee and customer experience. Spend amazing amounts on data rectification, reporting development. Relying on individual IP holders “That Know” the data to be the gatekeepers. Only those mythical team members being able to sort the fact from fiction. The Seers as I refer to them. In fact, these individuals create enormous business risk and can hold the business Intellectual property to ransom. Not that these people are evil but often have just developed into these owners through time.

Beyond data let me explore the processes and benefits of moving through the other layers of maturity.

Planning

Planning is the fundamental processes that all business attempt to achieve and often fail. I will get my historical data at the best lowest level I possibly can. Gather intelligence on potential change influencers and create a plan. From that point on I begin the cycles.

It doesn’t matter what area of business or what industry, plans are required but are usually restricted based on the constraints of transparency and data availability. This is why the first step of data management is so critical. Maturing is to be able to bottom-up plan and roll down all at the same time? Building plans based on historical information, and trends as a standard practice.

The success I have achieved only started with strong planning practices. Understanding the splits required to manage a plan and making sure those splits, aggregation or disaggregation points exist for future management. I always try to develop plans from the lowest level of data available. Transactional information and associated cost been the best starting point. What this means is that I the future I can validate and check compliance at any level to help me move to greater analytical maturity in time. Just creating plans and engaging stakeholders in this method of planning can greatly increase trust and empower greater input into the change components for new plan delivery.

Of course, basic analytical/statistical practices can be used in the process. Trends, seasonality, outliners, bell curves etc. will be used to develop a forward-based plan. What this does allow though is an appropriate and trackable modification. When applying a change to a plan, it can be done at a level that makes sense and can be easily validated and track for variance as the plan in implementation. When assumptions are made, they can be applied as a separate component to a particular plan level or built knowingly to a plan number as a variable adjustment. This creates the power to understand which levers of change had what result. If multiple changes are made without an ability to measure impact, then the ability to KNOW is lost.

Often the journey starts with atop level trend based plan. I would take the total historical result and create a plus-minus variant plan for the future. Plans like this are impossible to manage and are set for failure. Even if met it will be almost impossible to understand why.

This does not mean that stretch cannot be planned. And it does not mean that creating plans purely on trend creating self-fulfilling forecasts are the result. Businesses always need to push hard to move forward. What this does mean though is that the plans are structured in a way that stretches can be applied and managed effectively. Built into the plans in rational ways even if the answer is still unclear on how to get there.

If not it is all but impossible to achieve a result. If you do, it was by luck not plan.

Compliance Management

At the end of the first cycle, I start comparing what my result was versus the plan. Am I on Plan or not? Often, even a month in, variances begin to appear and this is where it all can unravel.

If planned correctly this is where my process of analytics and KNOWS logic can start driving change. But I need to know what are the potential things that can drive a variance. As an example in transport. Rate, KG shipped, region shipped, Volume break compliance, service type, shipping type, cubic volume truck type. The list goes on. Let us review a basic principle. If this is where a variance can occur, then two things need to happen.

1.   We need to plan at this level

2.   We need to measure compliance at this level

To understand the “WHY” then measure the conformity of these to the plan can start filling in the gaps. For instance, If I planned a certain amount of KG volume in a regional environment at a cost expectation what was my compliance. If the actual vs plan shows something wrong, then this can lead to a greater WHY understanding.

I know this sounds simple and in principle it is, but in practice takes a lot of investment and commitment.

Compliance validation though still very much in the first KNOW (Where am I) but does start to identify the second KNOW (Why). Taking compliance one step further by measuring variance at a driver level take analytical maturity to the next phase. Having the drivers measured and understood greatly narrows down the cause and can create the appropriate focus for deep-dive analytics and additional reporting requirements. Even more importantly it starts to drive actions and decisions

In reality, there is yet a long way to go before maturity. It still only begins to empower actions and decisions. Usually, the decisions still require further investigation. Think of it this way also. How long does the compliance process take? Often the cycles of producing the analytics and reporting to answer the compliance questions are laborious an inefficient. By the time they are available the cycle of review is already desperately and overdue. The further time goes the harder it is to understand the why. Just like in criminal investigation the first 48 hours after a crime is the key windows to capturing the criminal or not. The same goes for analytics. It is all just evidence after all.

Meeting on meetings are created to declare conformance only to answer none of the business problems or set direction. By the time some specifics of investigation are identified, and work commenced the cycle has rolled over, and the whole process begins again. Trapping business in a slow creep progression of improvement or bad decision processes. The only way to break these cycles is to ensure that compliance information and even better still driver level compliance information is available quickly. Structured in a way that makes sense and can be tabled for review easily.

When I broke this cycle, it was incredible of the power analytics available me. Variance information at readiness starts to focus on real problems and actions could be put in place with immediacy. Improvement identification implemented and dramatic business step change improvement achieved.

Projection and Simulation

But still there is further to go. I want my focus centred on the third KNOW. (Where am I going?) This can come in a few different forms. The previous level of maturity are predominately reflective processes and give limited forward perspective.

Projection – Based on the available information where will I land. What will be future compliance?

Simulation – What if something was to change

Action Impact – How will my actions affect the result

GAP management – What is the non-compliance in my plans I need to address.

Only when you have tools and analytics to support this activity can right business begin. Taking information from drivers, making decisions against the drivers to affect outcomes and then tracking the compliance of those impacts. Building bridges for gap contingency, building confidence in performance management, empowering the greater stakeholder groups to identify change requirements and enabling cross function involvement in resolution. It is where business begins to fly.

The power of leadership grows rapidly.

When first I began the journey, budgets created by using trend over previous years. I was unable to tell what money was spent and why. What is best practice? Where are the improvement opportunities and what is the urgency of changes required and where? Floundering and lost though at the same time buried in data and reports.

By implementing the above approach and building from the foundation have developed ground-breaking financial transparency and performance management. Being able to drill to individual transaction cost, manage compliance to plans, identify improvement immediately and engage decision to improve. Develop reporting to interact with stakeholders that are trusted and relied upon. Develop long-range strategies based on the insightful data leading to dramatic business performance improvements. Development of plans that are SMART objectives based that motivate people through a clean line of sight to HOW.

Using these processes has led to three-figure million dollar improvements to cost structure in the business and created improvements to both employee and customer experiences. Over a three-year period, we have become the business leaders in information, planning and performance management. A trusted advisors and sort after for innovation and change. Able to implement change and to articulate P&L effect. Trusted and relied upon members of cross-functional groups with a seat at the table for decision-making processes.