In this series, we explore how the longer-term demand plan should play a more prominent role in businesses. The first two parts deal with the key elements in setting up the demand planning and forecasting process, and later, we will explore what are the key behavioural elements, i.e. what do we do with the process and the outcomes when we get them. You can read Part 1. here: https://bit.ly/2MpyPzS.
- Assign responsibilities for each layer of planning
“If you don’t like my forecast, then just tell me the number you want, and I’ll do my best to deliver it, but I won’t guarantee it,” – a salesperson’s dilemma.
It doesn’t matter how you look at this problem, if a plan can’t be executed effectively, the plan is not valid. I wish I had a dollar for every time I’ve heard the statement: ‘It was a good plan, it just didn’t get executed properly’. This doesn’t apply exclusively to demand planning by the way, I have heard it being said about strategy, product plans, supply plans, and even between marketing and sales.
A second truism in demand planning is that you will never get the detail right if the aggregate is not right. The process needs to start with higher-level aggregate plans, coupled with feedback loops to continually align the detail back up to the aggregate. Typically, there are four major layers of planning in any organisation (see Figure 5.):
- The annual strategic and business planning cycle.
- The monthly cadence of re-planning and re-optimising the next 24-month rolling horizon.
- The weekly re-planning and tactical execution of the monthly plans.
- The daily execution of the weekly plans.
Figure 5. The four layers of planning.
As illustrated in Figure 5., there should be an annual planning cycle, which (re-)aligns the longer-term strategic plans (usually five-to-10 years), agrees a more detailed business plan for the next three years, and then aligns the annual budget to the longer-range aspirations.
Everyone in a company needs to know about strategy and it is the lead team’s primary area of responsibility to develop the strategic plan and cascade it down to the rest of the business. The monthly cadence, is the routine ‘stop and check’, to make sure projections are still on track. This is also primarily the domain of the lead team, but also integrates middle and junior management in ensuring changes that occur each month are captured, assessed, and used for re-planning the medium-term horizon, i.e. the next rolling 24 months.
This will be discussed in more detail in Key 10 in following articles, but in short, it is the role of the integrated business planning (IBP) process to review the core plans of product, demand, supply, as well as the key input and output plans (asset, people, finance, and inventory), to re-align back to strategy every single month.
The purpose of the weekly cadence is to manage execution of the IBP plans in the shorter term, and because it is run by the next level of management, it is an essential element in freeing up the lead team to stay focused on the longer term. The primary objective of the weekly re-planning meeting and process is to deliver the signed-off IBP plan, or if it can’t be delivered, to communicate this as early as possible. It’s the ‘pulse check’ for how well core plans are being deployed.
Daily activities complete the top-to-bottom alignment of planning with execution. To make this work, there needs to be an open and welcoming culture of ‘bad news early is better than bad news late’. So, each layer of process plays a vital role in ensuring top-to-bottom alignment and communication of the variance to plans. With the right people assigned to the appropriate horizon and clear escalation criteria, the lead team and other senior managers, can work in an environment of ‘silence is approval’, i.e. ‘if I don’t hear from you, I can safely assume the plan is on track’.
- Ownership and accountabilities defined
“Our marketing forecasts are robust and supported by well-defined and quantified assumptions; I just can’t get the sales team to deliver,” marketing director’s dilemma
To get the right level of input and perspective, and to achieve consensus without ‘drowning in the detail’, it is important to define who owns which horizon and to what level of detail. In the example shown in Figure 6., ownership of the higher-level, longer-term forecast is the responsibility of the most senior sales and marketing roles.
Depending on how big the organisation is, there can be a cascade of ownership, defined by an increasing level of detail, as the horizon becomes shorter and closer to ‘the here and now’.
While the bottom level is more execution than prediction, it is nonetheless, important to have mechanisms to ensure the demand plan is being executed effectively. Referring back to Figure 5., this is how changes happening at a daily level, can be continually re-aligned and re-optimised back up to the IBP plan and ultimately back to strategy.
- Spend more than 75% of the time in the monthly management meetings, on the four- to-24-month horizon
“Those people who walk around looking down at their feet, will eventually walk into something solid.”
Increasing time on the medium- to long-term is easier said than done, in that it requires changing well-entrenched behaviour patterns. Combined with ‘Key 5’ below, the structure of the monthly demand review process needs to encourage (and measure) more time spent looking at the future.
Figure 6. Planning ownership.
When I was in industry, we used the term ‘continental drift’ to describe the way marketing and sales would slowly and imperceptibly drift apart as the year progressed to the point where it was almost impossible to bring them back together until the next annual planning cycle, and even then, that was a struggle.
Senior marketing and sales managers should, therefore, routinely spend one-to-two hours a month reviewing the 24-month demand plan. The key elements and flow of this meeting are shown in Figure 7. To support this, companies often measure where time is spent. The key parameters can be easily jotted down just using pen and paper, or set up in a simple a spreadsheet to measure:
- Data accuracy / integrity discussion – this should be zero.
- Process integrity discussions – this should be zero.
- Assumption performance and key metrics – ideally this should be about 15% of the time.
- The short term, i.e. next three months – ideally this should be no more than 10% of the time.
- The next 12 months and this financial year – ideally about 35% of the time.
- Months 13 to 24 – ideally about 40% of the time.
|1.||Start with a review of progress to strategy or critic success factor – this sets the longer-term perspective up front.|
|2.||Assess changes to key strategic assumptions (also known as key performance drivers) – this is one of the most critical areas in managing strategically and will be discussed in more detail later under Key 6 in Part 2.|
|3.||Analyse performance of key operational metric (forecast accuracy and bias) – at least a little time needs to be spent on history to make sure we are continually getting better at planning and execution.|
|4.||Review projections in aggregate – an important element is to not “get lost in the detail”, so several families / aggregates / categories, are the best way to view the new demand plans, knowing that the detail is being managed by the demand management and planning team.|
|5.||Identify upsides and downsides to the projections, and most importantly, agreeing and documenting the contingency actions that will be deployed.|
|6.||Model possible scenarios for major events not covered in point 5 above.|
Figure 7. Demand planning review structure.
Figure 8. is a representation of where time should be spent in the monthly cycle (IBP), but as shown in Figure 5., to facilitate this, an organisation also needs a weekly re-planning process, which we call integrated tactical planning (ITP), whose charter is to deploy the monthly plan through a series of weekly re-balancing iterations. This is to focus where time is spent on the shorter term.
Figure 8. Demand planning time allocation.
- Consensus demand planning
“When I get everyone together to debate the demand plan, all I get is what I want, and then it turns out to be wrong. There’s’ got to be a better way to forecast,” sales director’s dilemma.
When the term ‘consensus forecasting/demand planning’ is used, most people think that it is a group of people sitting around a boardroom table coming to a consensus on the plan. This is not the definition of ‘consensus’, and doing it this way only leads to ‘group think’.
Similar techniques to consensus forecasting, such as the Delphi technique and ‘panel forecasting’ have varying degrees of success, but to truly arrive at a consensus, the process needs to have three essential criteria:
- The individual forecasts must be done in isolation and derived from those forecasters’ perspectives.
- Key drivers or underlying assumptions have to be documented to describe the thinking behind how the numbers and plans were generated.
- The consensus is reached by rigorous challenge and debate about the different sets of assumptions and perspective, not the actual numbers themselves.
- This has been demonstrated to deliver good forecast outcomes, as well as minimising the group think effect. Typically, the three primary perspectives in most businesses are sales, marketing, and statistical forecasts, but could also include many other inputs, as shown in Figure 9.
- Each of these inputs needs to be quantified and time-phased, so there can be an ‘apples-to- apples’ comparison.
- A good starting point is to set up different views for sales and marketing, and add statistical forecasting projections to anchor the plans. Such as:
- – Sales forecast by ‘A’ SKU, by customer, for the first six months – the bottom-up forecast.
- – Marketing forecast by category for months one to 24 – the top-down forecast.
- – Document assumptions supporting each forecast.
- – Gaps to the bottom-up and top-down forecast, and the ‘step’ at month six, then become the topic of dialogue and debate.
- The important point is that we want to debate the ‘thinking’ behind the numbers, not the numbers themselves.
Figure 9. Demand planning inputs.
We need to set up the demand planning and forecasting environment to ensure the behavioural and process characteristics are understood and well embedded.
The critical elements are to ensure there is an ‘anchor’ as a starting point, ownership is defined for both horizon and level of details, and multiple perspectives are not only allowed for, but encouraged, before coming to a consensus on best possible demand plans based on the knowledge we have at hand today. As a last word in this section, I quote Brad Pitt in the movie Moneyball, “… it is a process, a process, a process ….”
In the next articles we will explore what to do with the demand plan once we have the process embedded in the business.
This article will continue in the November-December issue of MHD Supply Chain Solutions magazine.
Rod Hozack is a partner at Oliver Wight. For more information email email@example.com or visit www.oliverwight.com.