I have been following the debate about the use of the terms IBP (Integrated Business Planning) and S&OP (Sales and Operations Planning) over the past few months with a lot of interest. It is with much trepidation that I step into this proverbial hornet’s nest after Lora Cecere wrote a blog titled “What Happens in Vegas should not Stay in Vegas!” in which she argues that:
To get started, let’s get beyond the nuance of the debate. This debate is not about the TERM. I REALLY don’t care what term is used or what process is called. I agree with Shakespeare, “A rose by any other name would smell the same….” But, I don’t agree with the conventional views on Integrated Business Planning (IBP) in three areas: focus, emphasis, and readiness.
Basically Lora argues that IBP is simply sophisticated S&OP. I’m not sure I agree. I think this characterization misses some subtleties between the two terms. Let’s start with Lora’s argument:
The number one change management issue with S&OP is and continues to be the role of the budget. If the company wants to maximize opportunity, the budget should be an input into the process, but not constrain the process.
Yes, I agree, but this is looking at the issue from the wrong perspective. What is broken is the budgeting process, which should be a continuous process driven by the operations forecast. And there is a lot of discussion in the Financial Planning and Analysis (FP&A) space about this, starting from an article titled “The Inherent Folly of Cash Forecasting” by Gavin Swindell in BusinessFinance. Gavin states that:
… The Hackett Group, recently published a survey of CFOs, and 70 percent rated cash flow forecasting as their top priority to be worked upon in 2011. Now, maybe I am being a little simplistic, but I find this strange.
Gavin argues that cash forecasting should be a “dependent” forecast based upon the operations plans for demand and supply. In fact he states:
The point is there is no need to forecast cash in a well-run business. It should be calculated. Anyone who is applying serious effort or cost to do so is papering over a business weakness presumably because they feel they have to or because they cannot fix the underlying causes. In many businesses, the cash flow from operations and the inherent working capital will week in week out account for the majority of transactional volume and value that businesses find hard to plan and predict accurately.
I agree. As importantly so do many others, for example in CFO Russ Banham states in a blog titled “Let It Roll” that:
Unilever parted with its annual budget in 2010, with no tears. So did Norton Lilly International. Statoil and American Century Investments have scrapped their budgets; others are expected to follow suit.
Russ is arguing that the annual budgeting process is broken and needs to change to a rolling process, and gives as an example Unilever which has moved to a quarterly budgeting process with a forward view of eight quarters. Russ has a great quote from Statoil:
Statoil, the large Norwegian oil-and-gas producer, decided to abolish the traditional annual budget in 2005. "We still do what the budget unsuccessfully tried to do for us: target-setting, forecasting, and resource allocation," says Bjarte Bogsnes, vice president of performance-management development. "We used to try to force these three purposes into one set of budget numbers, which created serious problems. For example, how can you expect an unbiased sales forecast from a sales manager if that number also will become a target? And how can you expect unbiased cost or investment forecasts from the organization if those forecasts also serve as an application for resources, and everybody sandbags?"
Who am I to disagree when even the McKinsey Quarterly in an article titled “Just-in-time budgeting for a volatile economy” published in the Spring 2009 edition in which the authors, when commenting on the budgeting process, state that:
Managers often spend significant amounts of time on it, only to be dismayed by how little value comes from four to six months’ effort. Under volatile conditions, when economic forecasts change from week to week, developing one reliable budget to coordinate business units and track performance for an entire fiscal year is very difficult. Following the traditional budget process may even be unproductive.
In fact they argue that the budgeting process needs to include the following:
- Scenario planning with trigger events
- Zero-based budgeting
- Rolling forecasts
- Quarterly budgeting
Perhaps even more importantly budgeting as part of FP&A covers a lot more than demand/supply balancing, which is the cornerstone of S&OP. Lora may argue with me that a mature S&OP should include much of what I discuss above, which is true, but the breadth of S&OP does not usually cover workforce management, R&D spend, capital expenditure, indirect procurement, and a whole host of other areas of interest to the budgeting process. Perhaps the most important of these is workforce management because so much of a company’s cost base is salaries for it employees. Deciding how many people should be in Marketing in a certain region should be based upon the forecast for that region. The manner in which the Marketing budget is spent should depend on the make-up of the projected revenue for that region. The same is true for Sales and Admin, Procurement and Warehousing, etc… Bringing all of these different cost elements together based upon the revenue forecast is integrated business planning, whether we call it IBP or not, and has a much wider scope than does S&OP. But, for all the reasons others have given, the budgeting process should be driven by the revenue forecast generated as part of the S&OP process, not the other way around. And of course adjustments to the S&OP plan may need to be made because of financial constraints that force, for example, a delay in hiring in R&D or expenditure on capital equipment, which impact the revenue projections. Lastly supply chain management (SCM) is very manufacturing focused and, as a consequence, so is S&OP. Of course we can talk about SCM in Retail, but the heart and soul of SCM is manufacturing. So what is the equivalent of S&OP for say a Financial institution or a consulting organization? What about bio-pharmaceutical companies which while still having large manufacturing divisions are primarily focused on research and development on one side and sales and marketing on the other side, which is reflected in the fact that their cost of goods sold is often less than 20 percent of revenue and Selling, General and Administration can often be more than 40 percent of revenue? I’m not convinced that S&OP is sufficient to run these organizations optimally. So I don’t think IBP is simply sophisticated S&OP. So who do you agree with: Lora or me?
Building What-If Scenarios based on the following inputs modeled in the Cube to derive Future Gross Margin
1. Revenue Forecast by Customer/Business Unit
2. Bill of Materials
3. Purchase Price Agreements by Hourly Rates by Tester (Consigned vs Non Consigned) by Contract Manufacturer for Back End Operations (Sort, Assembly/Test/Add Mark)
4. Purchase Price Agreements by Technology, Wafer Size, GDPW
5. Projected Test Time from Test Engineers (For Future Projections)
6. Projected Die Size
7. Sourcing Rules and Sourcing Percentages
8. Sales Price by Customer Agreements
9. Overhead by BU from Hyperion Budgeting
All of the above is modeled in Finance Cube, Demand Cube, Supply Cube and lining these Cubes and ability to do What-If Analysis by tweaking any of the attributes in any of the cubes and projecting Gross Margins by Customer, by Item at the lowest level to by Product Family/BU/Region
One Single View of the Enterprise for the CFO to improve and project Gross Margins for future quarters
I have not seen any S&OP tool doing this.
I have analyzed i2, IBMs TM1 and Oracle's Integrated Operation Planning (IOP) Tool and selected Oracle IOP for the Modeling Capabilities.
There are few issues came across with the Modeling and Presentation. Not sure there is a right tool in the market that does the above.
I am so sorry for replying so late to your great comment. I must have missed it earlier.
It is interesting that you seem to have focused on testing as the key capacity constraint. Also am I correct to deduce therefore that your are/were working for a fabless semi? Or was it that testing was the primary constraint and sufficient to use to determine the throughput under differnt mixes?
You should have had a look at RapidResponse :-)
Leave a Reply