Is S&OP an expensive band aid?

Dr. MadhavDurbha

S&OPI love talking to customers and prospects! Each of these interactions provide me with an opportunity to meet someone new, learn about their business, their challenges, dreams, and aspirations. The topic that often comes up in these conversations is Sales & Operations Planning (S&OP). While S&OP as a discipline has been around for over two decades, it has been generating great interest recently. As much interest as it has generated, based on my conversations, the results from S&OP efforts have been mixed at best. From time to time, I hear comments such as:

  • “My monthly S&OP process takes 6 weeks to execute”
  • “We have this massive excel sheet into which we load all our S&OP data to generate the reports for review. The process to gather the data is time consuming and by the time we present our S&OP to our leadership, the world has moved on and our plans are no longer valid”
  • “We started S&OP as our COO insisted we do it. It is turning out to be a report to him, rather than a tool to run our business”

In fact, this has been such a recurring theme that I decided to share my point of view in this blog. Let me elaborate on what I believe are the reasons behind this disillusionment.

  1. Disconnected processes: Several of these companies were advised by well-meaning consultants who have written books preaching S&OP best practices. These consultants talk about “above the line” planning and “below the line” planning. “Above the line” planning in their view refers to aggregate level S&OP planning processes such as demand review, supply review and so on covering an 18 to 24 month horizon in many cases. “Below the line” refers to lower levels of planning down to SKU level and into days and weeks (shorter time horizons). An artificial boundary is drawn between the two realms.

In this paradigm, translating S&OP to “below the line” plans requires running a bunch of excel spreadsheets to disaggregate the information. This causes disconnects between the aggregate volume plans and lower level mix plans. If a major disruption happens at execution level, to pass that information back to S&OP level takes quite a lot of time resulting in disconnects between strategies, tactics, and execution.

  1. Islands of information: Over the last two decades organizations made quite a bit of investment into Supply Chain Management (SCM) systems. These systems span across functions such as demand planning, supply planning, inventory planning, etc. Majority of the established vendors in the SCM space have grown into these disparate planning functions through acquisitions. This resulted in a highly heterogeneous underlying technology. Integration of these distinct capabilities became an afterthought. Sharing of information between these modules itself is cumbersome, let alone bringing external data from internal and external systems. The communication between these functions is often batch, waterfall oriented, introducing latency. These vendors approach S&OP as an overlay, creating what essentially is a band aid.
  1. People working within their bubbles: Traditional roles such as demand planner, capacity planner, inventory planner, etc., place people in individual bubbles with information thrown over the fence from one function to another. These planners are making decisions everyday within their functional domains with no visibility to the companywide impact of their decisions, undermining S&OP, which is meant to be a highly collaborative, cross functional process. The effects of this can be underperforming promotions, or slower inventory turns while increasing the risk of stockouts.

The factors listed above have significant impact on productivity of people. Those in the business community that are key stakeholders in the S&OP process spend countless hours providing inputs, massaging spreadsheets and formatting plans to fit the S&OP template. Most of these plans are out of synch by the executive meeting as the business keeps moving at the speed of….. well….. business! They start to see S&OP as an administrative overhead as opposed to a process discipline to run their business. Creating “what-if” scenarios happens in different versions of spreadsheets which reside in people’s inboxes for collaboration and reviews. In short, it is a mess!

Another problem due to the above challenges is as follows. When a company begins its S&OP journey, a process lead is assigned to orchestrate the process. This individual happens to be one with broad and deep knowledge of different planning functions across the organization. In other words, someone with the intellectual curiosity, knowledge of the business and supporting systems, and people skills. However, latency of information and inability to collaborate across the functional bubbles reduce the “S&OP lead” to an “S&OP coordinator”, an administrative function focused on gathering data and getting reports ready, as opposed to helping run the business. When stakeholders lose faith and the role of the S&OP process lead diminishes, the initiative is destined to fail.

So, should companies get rid of S&OP? The short answer is no. It can be very effective, provided organizations can address the challenges highlighted above. Harnessing the power of connecting the people, processes, and data can significantly enhance the value from S&OP. Here are some suggestions:

  1. Harmonize and surface the right data: Having the technological platform that is agnostic to underlying data systems and provide flexible means to extract and harmonize data can be very useful. Rendering this data in the form that can be consumed by the S&OP processes, and doing so in a timely manner is most fundamental to effective S&OP. This ability to harmonize and surface the right data is even more critical in today’s world where increasing amounts of data comes from external sources such as contract manufacturers or channel partners.
  1. Plan concurrently: The ability to concurrently plan all the way from S&OP down to production, distribution, and procurement planning at lower levels with explicit and visible links between financials, volume, and mix will close the gaps between strategies, tactics, and execution. To be able to assess the impacts and tradeoffs in real time, concurrent planning is critical to an “executable” S&OP plan. Effective end to end planning should be like pulling a chain. You pull at one end and you should feel the tug at the other end. In short, get rid of “above the line” and “below the line” thinking when it comes to S&OP.
  1. Empower people: Information is power. Rendering the right information to the S&OP stakeholders at the right time, enabling them to create what-if scenarios on the fly, to collaborate with other stakeholders, and to make decisions in near real time is absolutely critical to empowering people. Without the right tools, people with best of intentions cannot do their jobs well and feel empowered.

All in all, if you want to ensure your S&OP is not an expensive band aid creating work for the sake of work while aggravating disjointed planning processes, addressing the above as you establish your processes is absolutely critical. Would love to hear your thoughts!

Dr. MadhavDurbha

As Vice President of Industry Strategy at Kinaxis, Madhav serves as a trusted advisor for our customers through sales and implementation, ensuring success. He also engages with our strategic customers and key industry leaders to drive thought leadership and innovation. Madhav joined Kinaxis in the summer of 2016, bringing many years of experience in Supply Chain Management across various industries. Madhav started his professional career at i2 (which was later acquired by JDA). During his 17+ year tenure at i2/JDA, Madhav played numerous roles in Customer Support, Consulting, Presales, and Product Management. During his illustrious career, he was instrumental in helping enable numerous large scale transformational supply chain opportunities. He is very passionate about Supply Chain Management and the role it plays in making the world a better place. He shares this passion with others through his engagements and writings. Madhav has a Ph.D. in Chemical Engineering from University of Florida and a B.Tech. in Chemical Engineering from Indian Institute of Technology (IIT), Madras.

More blog posts by Dr. Madhav Durbha

Discussions

  1. An excellent critique of the S&OP “state of the art” and solutions 1 and 3 are spot on. Not sure about 2 though – if concurrent planning means disaggregating the S&OP plans into sku level execution plans then thats a mistake because the former are forecast driven (and therefore very inaccurate); execution should be demand driven, even in fast moving ex-stock supply chains.

  2. #2 is open to clarification. The disaggregation of the Production Plan (at the product family level) output of the S&OP process is accomplished by Master Production Scheduling

    I believe that the single most important element missing that results in the failure of the effort is the lack of education and training of all personnel. There will be varying levels of education/training required but all should receive some level. This will serve to achieve buy in and a complete, team wide understanding and hopefully acceptance of the “new” approach and culture change.

  3. Thank you Simon. I am not advocating going down to sku level where forecast accuracy is in question, i.e., in longer S&OP horizons. However, I am seeing this not having to go down sku level in S&OP context is being applied in a broad brush manner. When we get to near term (say 0 to 3months, depending on the nature of the industry and leadtimes), there is no excuse to claiming that S&OP as a process can run independent of rest of the planning. They need to be synchronized. The technology for concurrent planning to holistically address east-west planning (distribution, manufacturing, procurement, upstream & downstream partner processes) and north-south planning (S&OP down to individual production lines) is here and now. That is my point about concurrent planning and #2. Hope this explains my thinking.

  4. Ben – Totally agree on the need for training. S&OP being a very cross functional process does require some good amount of change management and education around the process.

    As for #2, S&OP process output is sent to MPS for disaggregation. In many organizations that I have seen, this process of communication of S&OP plan and disaggregation to MPS happens by communication through excel spreadsheet and can take several days to manually handle the disaggregation, not to mention the human errors in massaging and reconciling the data. Meanwhile quite a few things change at the least in the execution and tactical levels which makes the MPS and S&OP fall out of sync. Concurrent planning addresses this by leveraging a single data model, running the calculations in memory keeping planning at all levels in synch, and in realtime propagating demand and supply side changes. It is an entirely revolutionary way of rethinking S&OP and its interplay with planning capabilities. The technology is here and now. Hope this explains my thinking.

  5. Hi Madhav,
    Can’t agree with you re use of forecasts, even short term, for driving execution. Even with 80% forecast mix accuracy (ie. world class) the majority of sku’s (ie. those of low-medium volume and high-low variability demand) will be at <60% accurate – with inevitable implications for the inaccuracy of the MPS and inventory balance/service that leads to back order saving schedule interventions and development of unplanned queue lead-time, use of unplanned capacity and excessive inventory (Q=VUT – Factory Physics).

    Absolutely use forecasts for planning capacity etc and so long as this is done reasonably well (not difficult) you'll have the right capacity in place. But allow supply to flow in line with demand using 'pull' – at the sku level the demand v forecast will be all over the place but it doesn't matter because execution is kept stable in terms of optimised sequence and cycle with minimal development of unplanned buffer – time, capacity and inventory. Such Demand Driven Supply Chains consistently deliver their planned service levels with up to 50% less average inventory and lower costs / shorter lead-times versus those using forecast push MPS/MRP.

Leave a Reply