Many of us who have implemented, improved, or managed a demand planning process have struggled to get our counterparts in sales to fully participate in that process. We often hear things like:
"I need to spend my time selling."
"This is taking time away from my customers."
"Our business is different; it's too unpredictable and can't be forecast."
This kind of resistance is unfortunate, as input from sales is a critical component of any demand plan. For one thing, the sales organization's leadership in demand planning is necessary for getting a good forecast; it's critical, of course, that sales represent the forecast in the sales and operations planning (S&OP) process. For another, it's important that sales representatives contribute or update near-term forecast numbers from their customer interactions, and that sales leaders work with marketing on longer-term forecasts at more aggregate levels.
There are many reasons why sales organizations—or at least some individuals—may be reluctant to fully engage in demand planning and S&OP. As a result, there are many ways supply chain organizations (which in many companies have overall responsibility for managing S&OP) can help break through those barriers. My firm surveyed managers at a cross section of companies around the world to get their opinions on what works best to engage sales in the demand planning process. The survey results are shown in Figure 1.
Those are all good methods, and they're discussed in this article. But there are others that can be very effective as well. In fact, this article presents 26 ideas for improving sales' engagement and ownership in the demand planning process. These are summarized in Figure 2. The 26 techniques are based on our experiences with global corporations and on our work helping companies improve their supply chain processes, information systems, and business performance.
The techniques are organized around process, enabling technology, organization, and change management. They reflect a sequence that often works well for improvement programs, and which I recommend for implementing the ideas in this article:
1. Process What process is needed to meet business requirements? What components will be included, and how do we make it easy to understand and follow?
2. Enabling technology. How will we scale the process consistently to handle a complex business situation? What will enabling tools do, and how will people interact with them?
3. Organization. What kind of structure is required to support the process, software tools, and strategy? Who should be doing what, and at what level? Who's accountable, and how is success measured?
4. Change management. How can we combine tangible (for example, metrics and defined accountabilities) and intangible (such as coaching) methods to help the sales staff best manage changing expectations and responsibilities? Although change management must be considered throughout the process, organization, and enabling technology segments, it becomes especially important when you are implementing improvements.
In addition, the themes of making things "easy" for sales (to the extent possible), adding value to the business, and continuous improvement will appear throughout the article.
1. Make things as easy as possible. Let's start by emphasizing that it's important to make it easy for salespeople to participate in and contribute to demand planning efforts. Sales positions are different from operational and analytical jobs that have more structure and a more continuous flow of work volume. Sales professionals also are under constant pressure to deliver sales growth, or likely face dismissal. Use their time sparingly.
2. Spell out exactly what the process is. The demand planning process must be easily understood. It should clearly convey who does what, and by when. Put key dates on an easy-to access, easy-to-read calendar. A salesperson should be able to explain it to someone else in just a minute or two and refer the listener to a simple flowchart, associated RACI (Responsible, Accountable, Consulted, Informed) chart, and timeline. Let me be clear: publishing that information in an e-mail will not do the job.
3. Have sales make incremental changes, not build something new each month. At a minimum, sales should start with a statistical base forecast and historical data, both to be provided by the demand manager. Depending upon the sequence and components of your demand planning process, you may also want to give the sales team other information, such as promotional lift or macroeconomic effects (the impacts of statistical correlations between economic or industry indicators and demand) to validate with customers, as the majority of their inputs come from customer interactions.
4. Define how sales should interface with customers to obtain forecast inputs. This topic is too complex and business-specific to discuss in detail here, but I will say that if this is not specified and applied consistently across the salesforce, you will get a hodgepodge of inconsistent information. Sales management needs to develop a specific customer-interaction strategy, often by channel, market, or customer segment.
5. Only ask for the level of detail and forecast horizon the business really needs. What are you really accomplishing with stock-keeping unit (SKU)/location forecasts for 24 or 36 months out? Sales input is more tactical and therefore is usually just a few months out. Sales executives and your marketing organization should be building upon the base forecast (among other inputs) with higher, S&OP-level forecasts for the medium- and longer-term months of your demand and financial planning horizon.
6. Don't ask sales to forecast every item. Ask your sales organization to provide input on items of relatively high importance and variability; do not waste their time on those that can be forecast using statistical techniques. Use techniques for segmenting demand, forecasting SKUs with different methods based on their volume, profitability, and variability.
Normally, sales should only forecast items with high coefficients of variation (standard deviation of demand divided by mean demand). The definition of "high" is relative to the variability of your overall data set, but one place to start could be to manually forecast or check items with coefficients of variation above 50 percent. This type of analysis is also important when determining which items will be made-to-stock versus made-to-order.
7. Let sales know that the process is open to change. No business process should ever be "sacred" or set forever. As part of a cross-functional team, sales should be able to suggest changes in the process. Expectations regarding process evolution should be well communicated. Everyone should understand that there will be some give and take, and that any process changes have to be grounded in business requirements.
Every company needs to forecast demand, plan operations and supply, and integrate this with financial planning, yet plenty of companies have yet to implement some kind of forecasting software to support their demand planning process. Among other things, these applications are good at collecting and aggregating forecast information, showing it in whatever view you want, and integrating that information with supply and financial planning. The systems are also good at statistical forecasting based on historical actuals. This article assumes that you have a forecasting tool and need to get better process compliance and ownership from sales. Toward that end, I'll suggest the following:
8. Match the software's configuration to your process. That may sound obvious, but you'd be surprised how often companies have gaps here. Make sure the foundational elements of your process are configured consistently in the system. For example, if your process is based on an 18-month rolling forecast, then show 18 months in the system and 18 months on the demand plan that you bring to S&OP sessions. For sales input, make sure views of the forecast map to the way you actually collect and manage the relevant data (for example, by customer and/or by product family and item).
9. Give every salesperson a view that specifically pertains to him or her. You can give others (especially managers) roll-up views or other information, but it's important to make it quick and easy for salespeople to find, view, and update their own "slice" of the forecast. Online access to their own views is certainly more efficient than exporting and e-mailing spreadsheets back and forth.
10. Explore using software people are already familiar with. Today's customer relationship management (CRM) tools incorporate more forecasting capabilities than in the past. It's worth investigating whether salespeople could manage their inputs with a CRM application they're already using, and whether your forecasting/supply chain planning system could interface with that.
11. Make historical data easily accessible. Salespeople need historical data when they're discussing future sales with customers. Make sure that historical actual sales as well as historical forecasts are automatically integrated into the software so that information will always be instantly available.
12. Highlight exceptions. Sales professionals should be made aware of exceptions or problems requiring follow-up that pertain directly to them. Rules can be put in place that automatically highlight exceptions, or a demand manager could manually indicate something that a salesperson needs to check or verify.
13. Have the software compute key performance indicators (KPIs). Salespeople understandably have a keen interest in their personal metrics, so it's helpful if the forecasting software calculates the forecasting KPIs to make it easy for them to see where they stand. Forecast-accuracy metrics are often not included in sales KPIs, but if your company is serious about demand planning, then they should be.
14. Use e-mail "triggers." Most demand planning systems today have incorporated some elements of automated workflow. This means salespeople can receive reminder e-mails when due dates are coming up, and supervisors can get e-mails when due dates have been missed!
15. Be mobile-friendly. Salespeople need a means of quickly and conveniently updating forecasts to reflect something they've heard from a customer. But sales representatives are frequently on the road, perhaps with intermittent connectivity and/or slow connections (such as at a hotel). Consider taking advantage of the increasingly popular applications that can be used on tablets and even smartphones.
16. Be clear about who is accountable for what. Who's accountable for the forecast—sales? Finance? The demand management function? I've seen companies handle it several ways. The issue of accountability also raises the question of who gets the final say if there is a disagreement while reaching the consensus demand plan. I recommend that demand management be accountable for the process, and that sales have the final say on the consensus demand plan. That means sales will also be accountable for forecast accuracy. If you want to keep someone or some function engaged, make them accountable for results.
17. Create a capable demand planning organization. Some companies still don't have demand planning roles, or have them at the wrong levels. Gaps here will create a huge challenge regarding everything we're talking about in this article. Most large and mid-market companies have found it beneficial to create a demand management function and the role of demand manager/planner within the supply chain organization. The demand manager/planner is charged with shepherding and owning the demand planning process. This requires things like knowing the market(s) and sales history, correlating market trends and economic factors with demand, collecting inputs and bringing them together, having a point of view, liaising and challenging at all levels as necessary, generating a statistical forecast (including knowing which models to use and how to maintain them), running consensus meetings, and integrating the consensus demand plan with both the business plan and supply chain execution.
Larger companies typically have several demand managers who specialize in markets or business units. Demand managers are quantitative, but they are also collaborative leaders. They understand the entire supply chain and act to ensure integration. I've seen some companies mistakenly put junior people in these roles. The junior people are in no way able to challenge a regional sales manager on his or her numbers, nor are they able to challenge a marketing director on the lift a promotion will generate based on what happened last time. It often makes more sense to have someone in a junior role (a demand analyst, for example) perform the more data-intensive tasks. As these employees gain more experience and collaborative acumen, some will progress to a demand manager or higher position, where they will interface with senior-level people around the organization.
18. Size the demand management function to ensure success. Determining the appropriate size of business functions is both an art and a science. Companies may size their demand management functions based on factors like market maturity, complexity, volatility, international exposure, size of the interfacing commercial organization, complexity of forecasting models and the effort required to maintain them, the use (or not) of demand planning software, number of demand consensus meetings to run, and number of SKUs under management. Keep in mind, too, that for demand managers to effectively work with sales organizations and perform their jobs, they must have enough time to complete their analyses, to build consensus, and to tie gaps in metrics to business impacts.
In the end, you'll have to rely to some extent on judgment and experience when determining how much human input is required to support your process. If you're unsure how large a staff you'll need, then make your best estimate and leave a path for correcting the amount of allocated resources as you learn more.
19. Let demand managers maintain an independent view. Some companies have demand analyst-type roles within their sales organizations, and while they at least have someone analyzing and aggregating demand, these typically junior people usually aren't able to stay independent or have much influence. That is why I believe the demand management function generally should not report to a commercial organization. I usually recommend a reporting relationship outside of sales—perhaps one that's part of an integrated supply chain team reporting to a supply chain director or vice president. This way, demand managers can challenge a senior sales, marketing, or other leader without a direct reporting relationship that compromises their independence.
20. Explain the "big picture." Resistance to changing roles, responsibilities, and expectations often stems from a lack of knowledge; your sales personnel may simply not understand how important their involvement in demand planning is to the company's overall success. Explain the whole demand planning process and what other functions are contributing. Tell them how the demand plan drives ALL downstream planning and, in a make-to-stock environment, drives execution (cost) as well. Explain that the forecast is what operations uses to actually meet the promises that the sales organization makes, keeping the company competitive in the marketplace with respect to innovation, cost, quality, and lead time. Consider also showing a simple, high-level chart of how it all fits together. (In fact, I believe every company should have a standard "Supply Chain Management for Sales" course.) Unfortunately, some won't understand or care.
21. Communicate in terms of WIFM ("What's in it for me"). Appealing to self-interest is the universal language of change management. People will not accept change if they think it will make it harder for them to succeed. Saying—and meaning—things like, "We'll help you keep your promises," "You'll have enough of the right inventory," and "You'll be cost-competitive" will hopefully convince them to participate. Some (but probably not all) will be able to connect this with their own interests in more sales, advancement, and commissions.
22. Include sales in the cross-functional process-design team. This is another recommendation that may seem obvious, yet some companies fail to follow it. Sales professionals need to understand that their ideas and concerns are valued, and that the process can change with their input and as business needs change. Assign them improvement projects, and they'll feel even more invested in the demand planning process.
23. "Triangulate" training and coaching sessions. This refers to delivering information with the same purpose in different ways. People often respond better to one approach compared to another. For example, sales leaders could deliver general training sessions (architected by a core team) to a large audience. But that shouldn't be the only way you provide information. Set up one-on-one coaching sessions, and enlist the support of a demand planning "ambassador" to work with peers in sales. This is analogous to a situation many of us have encountered at home: If I tell my teenager something, he pays little attention, but if his friend tells him something, he may listen.
24. Get support at the top. Like any good manager, sales leaders are protective of their people's time and energy, so you'll need to demonstrate the value of participating in demand planning if you want sales leadership to follow up with their teams on process compliance, improvement opportunities, and, of course, the metrics. What's important to the department leader and what she or he asks about gets the attention of direct reports. Without this show of interest, people will often think, "If my boss doesn't care about it, why should I?"
25. Tie forecast metrics to operating and financial metrics. Everyone measures things like forecast error, forecast bias, and so forth, but not everyone is relating these metrics to on-time in full (OTIF), stockouts, and inventory levels that, in turn, tie to sales, cost, and cash flow. Salespeople don't like these metrics. To get them to accept and support them, it's important to stress that the metrics are not personal; they are only a tool for improving the demand planning process and the effectiveness of the business.
Many demand managers want to measure the accuracy of various components of the demand plan, such as statistical, sales input, and others. This may be good for some organizations, but I'd suggest that you gauge your organization's capacity to absorb several variations of the same metric. If they have many other things on their minds, it may just confuse them. Keep it simple and manage the results.
26. Make it a formal part of the job. Add forecast duties to job descriptions and add metrics to performance evaluations. It may take some time before you are able to implement this, but it should be incorporated in the future. If your team doesn't have the authority to make changes like these, then you may not have the right people, organizational level, or commitments for changes to be sustainable.
Give it a try
Improving the demand planning process and embedding the culture of forecasting within sales requires continuous effort. Tying the process itself to results is key; otherwise much ground could be lost if there is a leadership or organizational change. Perhaps not all of these recommendations will work in all business environments, but they are worth trying. Hopefully you will find some takeaways that will be helpful in your organization. Like most efforts to make operational improvements, success will depend on a blend of people, process, and technology.