Why do companies use budgets? For the same reason that project managers have plans – to have a baseline against which to compare business performance. We all know that. But there are some things about budgets and other planning practices that we need to (re)think through now that we are in the Covid era.
We are in a reality where everything in the business environment changes so fast and so often that old premises behind the practices of budgeting and planning in general no longer apply.
Budgets are a very old practice. A budget, planning techniques like MRP, monthly reviews, quarterly reviews, annual reviews – they all have something in common: the premise that reality can be represented in models with fixed parameters. That assumption implies that reality doesn’t change enough that fixed parameters are still adequate to manage against.
Think about this for a second: a budget is a forecast of how things will happen in the future, established at a point in time. At that point in time, you use many fixed parameters: unit costs, average demand, lead times, the price of materials or parts, not to mention forecasting demand for the next 12 or 18 months.
The fact is that all of these parameters entail enormous variability today. The most common planning and budgeting practices today, are based on methods that were created many decades ago at a time when the world had a certain established order and the structures of business – supply chains, Enterprises, business law, competitive landscape, the way you become competitive – they were all relatively stable, i.e., part of that world order.
For the past 10 years, the world has been changing rapidly, away from that 20th century state of things. With Covid, that change has accelerated at a breakneck speed. Fortunately, the methods and the tools that have emerged to cope with a highly variable environment have also evolved. We are at a crossroads now, where we can and we must walk away from established practices from four decades ago and adopt new ways of managing businesses, aligned with today’s reality.
Believe me, you may not do it but your competitors – old and especially new – are doing it.
Think about this one simple example: the time it takes to perform an operation in manufacturing. What do we use this parameter for? We use it to schedule operations. But the time it takes to perform an operation actually varies quite a bit over time. So, what do planners do? They probably use statistics to come up with one value for planning purposes: that could be an average, a median, a statistical trend – but ultimately they come up with one value to be used all the time.
Now: operation run time will be used for scheduling. But scheduling itself is a practice based on linear methods. Instead of using linear methods, we can use something else: an AI model. Interestingly though, that AI model doesn’t need fixed parameters. The same way it learns from past reality to develop a schedule, it can use that same knowledge to predict what the operating time is going to be.
Imagine that: a planning approach that doesn’t need fixed parameters and doesn’t use linear methods. This is the planning method for the Covid era: use AI technology for anything that has to do with planning.
But continue the thought process: AI doesn’t need fixed parameters but it also doesn’t produce fixed predictions. An AI model can predict what’s going to happen on an ongoing basis and thereby change that prediction as reality changes.
Now let’s go back to budgets. Let’s imagine that there are no budgets, there are no baselines and there are no fixed forecasts. Let’s accept that all the characteristics of reality that make a Business viable and successful are moving constantly and we want to manage on that basis. What do we do, then?
We adopt a very simple principle: you don’t have baselines or fixed plans but you have business targets and you have business constraints.
An example of business targets can be profit or profit margin. An example of a constraint can be working capital limit or available machine hours.
What you do, then, is this: you track your business targets on a non-going basis; you use AI to predict what the performance is going to be against those targets; and you set up the AI model to consider business constraints – resources, space, cash flow, working capital, etc.
You don’t manage against an artificial baseline constructed in the belief that you can predict the future or fix the future. Instead, you manage against business targets and you use AI to not only measure current performance but predict what that performance will be like in the next several months. You do this on a non-going basis. In other words, managing the business becomes a real-time activity and not a periodic activity.
Everything that you do in planning and managing operations – improving them, deciding on resources, deciding on level of investment needed – everything is done in Now Mode. In other words, you are constantly managing on the basis of what is happening now, what has happened in the recent past and what is predicted to happen using AI models. You do this continuously.
If you are running a classic 20th-century enterprise, you may think this is impossible. You may be used to an environment where in order to monitor business performance you need information and it takes a lot of work to bring together that information. However, in today’s world, this limitation should not exist. By using a good big data analytics platform, you will equip yourself with real time visualizations so that the information is available for you to look at, at the touch of a button.
You also need to transform your organization and adopt an Agile Model. In Agile, the management load is distributed by all employees. Multi functional teams of employees are highly capable to manage their work, track performance and make decisions to solve problems.
Agile organizations are collaborative, fast and adaptable to changes in the business environment and in technology.
When you’re managing an Agile organization, you don’t need so much middle management and senior management is not encumbered by operational problems. Senior Management – the people often concerned with budgets and results – can monitor performance and predictions in real time; they can also leverage a collaborative environment to introduce changes in direction quickly.
An Agile organization, a business managed in real time, using big data analytics and AI models – that’s today’s winning Enterprise.
The very good news is that:
- AI has been democratized. One can acquire Machine Learning API’s in Open Source or established platforms like IBM or AWS.
- Creating big data platforms is actually relatively inexpensive.
- Producing visualization dashboards from big data platforms is also inexpensive and quick.
So: rethink everything and adapt to the high volatility of the Covid era. Stop thinking of data as the ‘province’ of classical IT systems. Create an environment with Agile teams and manage in Now Mode.