Demand planning is integral to almost every planning process. Demand is the driver of the business and everyone else’s accuracy is dependent on how well the demand is planned. Despite how important demand planning is, it is often done in isolation and using antiquated tools and process. So, what is the optimal way to plan demand within an organization? I dissect the best demand planning processes I have seen and identify why I think they are so successful.
An integral part of demand planning is how long you are looking out. This one may seem obvious, but there should be a greater focus on the immediate term. Most companies will do a weekly level forecast for the coming 3 months (12 weeks) and will have a monthly level estimate for the following 9 months. The short-term estimate could vary and is based on how long it takes to prepare your product for sale. If your company requires 3 months lead time to affect inventory change, you may need to extend the short-term planning window. The 3-month view should be at a minimum of 95% accurate.
This short-term forecast drives the operations of the company. This is how you can make sure you have enough product in stock, or cutback if manufacturing is outpacing demand. Managing inventory is a delicate balance, and an accurate forecast is the strongest tool you have to control it.
The following 9-month forecast (long-term) is generally more of an advanced look at how things are looking for the out periods. The accuracy is not quite as important, but it will still drive significant financial decisions for the company. Generally, this is meant to capture last year’s seasonality plus any changes that have occurred in the business since then. If there is significant change coming, then this forecast is more important, but that should generally be handled on a case by case basis. The long-term forecast can also vary in length. For instance, non-profits may need to forecast out several years to ensure the appropriate grants are in place while an airplane manufacturer may need to look out decades since manufacturing is a very slow process. Having a 12-month outlook is about average, but it isn’t the only option.
The frequency of updating these plans goes almost hand in hand with the level you are planning at. If you have a weekly demand forecast. You need to be updating it weekly. I do believe that it is important to keep it ‘live’ though. What I mean is that if you know a change is coming, don’t wait until the next forecast to note it. If I learn of a big new order that came through, I should add it to my forecast that day. Depending on connection to other tools, this could be the difference between fulfilling the order on time or running out of stock.
Who should be planning the demand? Unfortunately, this varies by organization. For a consumer product company, it would make sense to have a team dedicated to this job, since margins are generally small, and inventory requires constant attention. If you are at a professional services company, this is something that could likely be done by practice managers. Hospitality companies should probably also have managers maintaining these numbers. Overall, the question comes down to the importance of the demand forecast, and whether it is time consuming enough to require specific resources to manage.
How does the demand planning fit into the rest of the organization’s systems? There are two sides to this. Data coming in, and data going out.
Many systems will already give you a starting point for what is coming in. Salesforce is a big one for most companies, but there are always other sources. You could be reliant on an order system which has all future orders tracked, or an external data source like Nielson to drive your forecast. I believe it is important to connect these systems together and have a process that incorporates them together.
Manually consolidating data systems is a time drain for an organization, but it never seems like one. You will likely hear from people that pulling data is just a fifteen-minute process. They might be right. Even so, if 20 people complete your planning process, that adds up to 260 hours a year. That is about a month and a half worth of time per system. Setup a process to load all data to a consistent planning tool. It will set you up for success in the long run.
The next question is where data should go after the forecast is created. There are several tools out there that allow you to create a comprehensive demand plan in the same tool as you create your financial forecast (Workday’s Adaptive Insights, Oracle’s PBCS, IBM’s Planning Analytics). Since these processes are often closely connected, it is likely worth getting of these tools to manage both processes. However, there will always be a need to use this data in other systems. You may need to export your data directly to an inventory management tool, and ERP system, or any number of other different tools.
Generally, the easiest approach is to export the data to a consistent database for access. Often, this is something like a data warehouse, where other systems can go grab it, but be wary of just sending data out there. A process always needs to be defined. If you are updating your demand forecast weekly, make sure the people getting the data understand the intended state of your data if you export to a warehouse nightly. Numbers that haven’t been updated are at best useless, but often cause problems for organizations.
Wondering how your finance team can adapt to ever-changing demands? Are there more efficient ways to budget? What exactly are dynamic planning strategies?
We will walk through how organizations can most effectively adapt to change through preparation, management and reinforcement, plus the critical transformational role FP&A teams have within an organization to encourage faster, smarter decisions! WATCH THE WEBINAR.