Operations and More Specifically, Operations Research and Data
Operations is often used as an all encompassing term to describe everything that your business does to make a profit but we could also narrowly define it when in strategic discussions. If we look at it from a high-level, strategic management perspective, that of the VALUE CHAIN then operations falls within the Primary activities and is immediately book-ended by the Inbound and Outbound segments as below. If you would like to see the image in the Strategic Management guide that I put together it is available on the strategy page.
Operations is often used as an all encompassing term to describe everything that your business does to make a profit but we could also narrowly define it when in strategic discussions. If we look at it from a high-level, strategic management perspective, that of the VALUE CHAIN then operations falls within the Primary activities and is immediately book-ended by the Inbound and Outbound segments as below. If you would like to see the image in the Strategic Management guide that I put together it is available on the strategy page.
When I refer to operations I tend to use it to refer to the data, costs, and capacity. First, operations, as the encompassing of all profit generating activities, also represents all available data and therefore represents all figures to accurately describe costs. and identify resource constraints. Second, on the segmented VALUE CHAIN level, these segments offer defined points of data sourcing, cost behaviors of a variable or fixed character, and functional-level bottlenecks. From a finance perspective, viewing operations as the data it produces offers a basis for cost management and costing method, in addition to, areas for capital investment. It is always a good idea to find the primary bottleneck and consider whether capital investment in increasing the throughput is beneficial; your profitability is bounded by your throughput capacity.
Operations as data also offers an area of analysis often overlooked by finance in the planning, analysis, budgeting and forecasting work, that of operations research and its use of mathematics to optimize costs and resources in areas of ordering and holding of inventory, labor utilization and learning curve analysis, operational activity organization, and fixed asset capacity planning just to name a few. Many of you will have heard of calculating the Economic Order Quantity, Inventory Carry Costs, and Throughput, each using operational data in an attempt to minimize costs or maximize utility of a given process. Using these methods can assist in modeling scenarios for planning, creating performance metrics, minimizing labor costs, and creating accurate budgets. Further, operations as data, naturally introduces questions about automation and potential integration since once a metric of costs or activity is created, the next question is always "How can we improve this?".
Data, Data Life-Cycle, Automation and Integration:
There is, not only, a vast availability of data for collection, but there is a natural integration of these activities, and, due to the repetitive nature of many of them, the activities are primed for automation. However, Automation is not Integration. Automation is the completion of a process with little to no manual requirements while integration is the 'communication' between two different systems. For example, if you select an Accounts Payable software application and this application enables an automated two way or three way reconciliation of Purchase Orders, Invoices, Goods Received and Payment with minimal manual entries and oversight, then does this application also communicate with your accounting software to record the payment to the cash, accounts payable, and inventory ledgers. This would be automation and integration. It is well understood that you could automate several processes, but if none of these processes integrate, the data must still be transferred between the systems.
Of course, many ERP systems attempt to remedy this, but to what degree is this effective given cost-benefit of the system? For small to mid-size companies this is a very relevant consideration and most must look through software applications for specialty accounting and operations such as leasing, marketing, sales and use taxation, payroll, general and subsidiary ledgers, supply chain and pretty much every area inclusive of equipment monitoring that can record idle times and engine temperature. With all this data and all these applications, understanding data and a bit of IT is not a luxury but a necessity and someone has to ask the questions. Since, at the most basic level, finance and accounting is truly the only area of an organization that legally requires data due to reporting and taxation regulatory compliance, it only seems proper that finance and accounting cultivates a supplemental role on a data team or as data officer. A key question then becomes
Does your finance leadership understand automation, integration, applications and APIs?
There are levels to the way an individual thinks about a process. Certainly there is the flow of the activities required to complete a process and oversee that it is running smoothly, a first level. Again thinking of a process example such as accounts payable activities, whether mostly manual or not, as long as the activities are completed with "acceptable" competence then this is where most finance leaders stop. But the next level of thinking asks questions such as
Operations as data also offers an area of analysis often overlooked by finance in the planning, analysis, budgeting and forecasting work, that of operations research and its use of mathematics to optimize costs and resources in areas of ordering and holding of inventory, labor utilization and learning curve analysis, operational activity organization, and fixed asset capacity planning just to name a few. Many of you will have heard of calculating the Economic Order Quantity, Inventory Carry Costs, and Throughput, each using operational data in an attempt to minimize costs or maximize utility of a given process. Using these methods can assist in modeling scenarios for planning, creating performance metrics, minimizing labor costs, and creating accurate budgets. Further, operations as data, naturally introduces questions about automation and potential integration since once a metric of costs or activity is created, the next question is always "How can we improve this?".
Data, Data Life-Cycle, Automation and Integration:
There is, not only, a vast availability of data for collection, but there is a natural integration of these activities, and, due to the repetitive nature of many of them, the activities are primed for automation. However, Automation is not Integration. Automation is the completion of a process with little to no manual requirements while integration is the 'communication' between two different systems. For example, if you select an Accounts Payable software application and this application enables an automated two way or three way reconciliation of Purchase Orders, Invoices, Goods Received and Payment with minimal manual entries and oversight, then does this application also communicate with your accounting software to record the payment to the cash, accounts payable, and inventory ledgers. This would be automation and integration. It is well understood that you could automate several processes, but if none of these processes integrate, the data must still be transferred between the systems.
Of course, many ERP systems attempt to remedy this, but to what degree is this effective given cost-benefit of the system? For small to mid-size companies this is a very relevant consideration and most must look through software applications for specialty accounting and operations such as leasing, marketing, sales and use taxation, payroll, general and subsidiary ledgers, supply chain and pretty much every area inclusive of equipment monitoring that can record idle times and engine temperature. With all this data and all these applications, understanding data and a bit of IT is not a luxury but a necessity and someone has to ask the questions. Since, at the most basic level, finance and accounting is truly the only area of an organization that legally requires data due to reporting and taxation regulatory compliance, it only seems proper that finance and accounting cultivates a supplemental role on a data team or as data officer. A key question then becomes
Does your finance leadership understand automation, integration, applications and APIs?
There are levels to the way an individual thinks about a process. Certainly there is the flow of the activities required to complete a process and oversee that it is running smoothly, a first level. Again thinking of a process example such as accounts payable activities, whether mostly manual or not, as long as the activities are completed with "acceptable" competence then this is where most finance leaders stop. But the next level of thinking asks questions such as
--What data is actually being collected?
--If there is data not currently collected that would be useful, the can we collect it?
--What IT application is being used for the process?
--Does this IT application capture data for us to access?
--Does this IT application present the opportunity to automate activities that are repetitive?
--Does this application integrate into other applications that are used in processes?
You might wonder why there is such an emphasis on operational data generated from activities.
Always bear in mind that all decisions will flow through finance. Virtually no decision will be made without some level of cost-to-benefit analysis. These analytics are more than a P&L analysis, in fact a three statement model tends to give very naive and generally incorrect assumptions when concerning profitability due to the nature of accrual accounting, but this is a topic for modeling. What data does give us is a definitive relationship between a cost measure relative to an eventual cash inflow and we want to use data, applications, automation and integration to the best of our abilities because it can improve costs and therefore the organization's profitability.
We are looking to create a data-driven organization and gain the benefits that arise from deeper analytics into the operational data and these impacts when translated to financial analysis, reporting, planning, budgeting and performance metrics.
Always bear in mind that all decisions will flow through finance. Virtually no decision will be made without some level of cost-to-benefit analysis. These analytics are more than a P&L analysis, in fact a three statement model tends to give very naive and generally incorrect assumptions when concerning profitability due to the nature of accrual accounting, but this is a topic for modeling. What data does give us is a definitive relationship between a cost measure relative to an eventual cash inflow and we want to use data, applications, automation and integration to the best of our abilities because it can improve costs and therefore the organization's profitability.
We are looking to create a data-driven organization and gain the benefits that arise from deeper analytics into the operational data and these impacts when translated to financial analysis, reporting, planning, budgeting and performance metrics.
There are six V's of data to act as a guide for collection and scalability:
Volume - the amount of data generated
Velocity - how often it is generated
Variety - the types of data generated
Veracity - the reliability of the data generated
Value - the usefulness of the data generated
Volume - the amount of data generated
Velocity - how often it is generated
Variety - the types of data generated
Veracity - the reliability of the data generated
Value - the usefulness of the data generated
There are different mappings for the data life cycle but in general:
Generation - what activity is producing the data
Capture - where is the data coming from
Processing - how is the data being manipulated for usability, reliability, and consistency
Storage - where is the data being stored and how is it accessible
Maintenance - how is the data organized, maintained and governed for use, quality, and privacy
Analyzing - how is the data being used for analysis such as tools, applications, methodology
Communication - how is the data being reported and visualized
Interpretation - what conclusions and insights are being drawn from the data
Generation - what activity is producing the data
Capture - where is the data coming from
Processing - how is the data being manipulated for usability, reliability, and consistency
Storage - where is the data being stored and how is it accessible
Maintenance - how is the data organized, maintained and governed for use, quality, and privacy
Analyzing - how is the data being used for analysis such as tools, applications, methodology
Communication - how is the data being reported and visualized
Interpretation - what conclusions and insights are being drawn from the data
Data will be collected and stored somewhere if you are using an application. POS (Point-of-Sales) systems generally store data, accessible via an API. Other applications such as those used for logistics tracking or heavy equipment monitoring do the same, often using APIs that integrate directly with BI tools such as Power BI. Reiterating an earlier mention, find out what data is stored, where is it stored, it is accessible, would it benefit you to have it in a central location such as a data lake.
Any data generated may have to be cleaned and then made accessible for reporting, modeling and analysis. Cleaning data usually happens through an API development or some developer application so all data that comes from the same source will all be formatted in the same way, ready to be accessed. These activities might include a robust time-intelligence table, corresponding dimension and fact tables and views or simply cleaned formats of data addressing null values, dates, aggregations or any other manipulations that the downstream users would repetitively have to correct. A sit down, led by the primary users of the data, the finance team and leadership, with IT or data engineering on what is required at a high-level can only save time and money downstream.
Does your finance leadership know how to communicate this with data engineers or IT?
Another observation that you might notice, after the high-level Primary and Support flow categories and the data produced, is that there are bullet points under each. These describe just some strategic considerations that coincide within the activities. Specific decisions that will require the data and analysis mentioned earlier. Many of these should be familiar to you.
When laying out the future path for the organization, the journey to the vision, the strategic formulation must be developed and operations data relative to financials are the influencing factor for all levels of the organization, business level, functional level, and unit levels.
The options for strategy require digging into the data and translating this data to financial measures so that a comparison can be done. Why? Because decisions should increase profitability. Who but your finance and accounting people can understand how this data is translated and how to assess profitability and comparing options? But the biggest question falls back to the depth of ability of the financial leadership within all areas of finance, accounting, operational research, data science, and business intelligence.
Does your finance leadership understand operational analytics and statistics beyond that of the standard accounting ratios?
--This gets you to probabilities for risk, resource planning, and scenario analysis.
...What data is needed and Which analytical methods work better and why?
--This gets you to the Contribution Margin, Cost-Volume-Profit and Break-Even Analysis, After-Tax Cash Flows for Budgeting and Profitability Decisions.
...How Operations Research models such as linear programming make for optimal resource utilization and cost structures?
--This gets you to modeling and visualization of the usage and needs of scare resources.
Certainly, the standard fastidious accountant is necessary as it is the 'bread and butter' of accounting and reporting for the finance leader, but very few have a breadth of knowledge that extends into the analytical and data science realms that can change an organization to a data-driven organization.
Any data generated may have to be cleaned and then made accessible for reporting, modeling and analysis. Cleaning data usually happens through an API development or some developer application so all data that comes from the same source will all be formatted in the same way, ready to be accessed. These activities might include a robust time-intelligence table, corresponding dimension and fact tables and views or simply cleaned formats of data addressing null values, dates, aggregations or any other manipulations that the downstream users would repetitively have to correct. A sit down, led by the primary users of the data, the finance team and leadership, with IT or data engineering on what is required at a high-level can only save time and money downstream.
Does your finance leadership know how to communicate this with data engineers or IT?
Another observation that you might notice, after the high-level Primary and Support flow categories and the data produced, is that there are bullet points under each. These describe just some strategic considerations that coincide within the activities. Specific decisions that will require the data and analysis mentioned earlier. Many of these should be familiar to you.
When laying out the future path for the organization, the journey to the vision, the strategic formulation must be developed and operations data relative to financials are the influencing factor for all levels of the organization, business level, functional level, and unit levels.
The options for strategy require digging into the data and translating this data to financial measures so that a comparison can be done. Why? Because decisions should increase profitability. Who but your finance and accounting people can understand how this data is translated and how to assess profitability and comparing options? But the biggest question falls back to the depth of ability of the financial leadership within all areas of finance, accounting, operational research, data science, and business intelligence.
Does your finance leadership understand operational analytics and statistics beyond that of the standard accounting ratios?
--This gets you to probabilities for risk, resource planning, and scenario analysis.
...What data is needed and Which analytical methods work better and why?
--This gets you to the Contribution Margin, Cost-Volume-Profit and Break-Even Analysis, After-Tax Cash Flows for Budgeting and Profitability Decisions.
...How Operations Research models such as linear programming make for optimal resource utilization and cost structures?
--This gets you to modeling and visualization of the usage and needs of scare resources.
Certainly, the standard fastidious accountant is necessary as it is the 'bread and butter' of accounting and reporting for the finance leader, but very few have a breadth of knowledge that extends into the analytical and data science realms that can change an organization to a data-driven organization.
Related Video Talks:
|
|
Vertical Divider
|
|