Monthly S&OP, Slide by Slide

The 4 weeks of the cycle

The deck is not built the night before the meeting. Each week has a forum, and each forum hands part of the deck to the next one.

Week 1
Close and data
Who: Planning and BI
Output: month closed, with sales, inventory, backorder and pipeline. Slides 1 to 5.
Week 2
Demand Review
Who: Demand, Sales and Marketing
Output: consensus forecast, events and risks. Slides 6 to 11.
Week 3
Supply Review
Who: Supply, Production, Procurement and Import
Output: purchase and production plan and capacity gaps. Slides 12 to 16.
Week 4
Pre-S&OP and executive meeting
Who: Finance and leadership
Output: approved plan and recorded decisions. Slides 17 to 19, plus the summary on slide 3.

Whatever does not close in a week goes up as a decision for the executive meeting. That is why the last slide is about decisions, and slide 4 next month checks whether they were carried out.

Slide by slide

The numbers belong to a fictitious company, Example Co., in the November 2026 cycle with the October close. It has five product lines, each built to show a typical supply case:

Line	Behavior	Where it shows up
A High runner	Local production, high and steady volume	Backorder, capacity, accuracy
B Imported	Long lead time, stock in transit	Import pipeline, coverage
C Seasonal	Peak in November and December, with a promotion	Forecast vs forecast, FVA, risks
D Launch	On the market since August, no history	Launches, accuracy, revenue
E Long tail	Being phased out, sales falling	Excess stock, SLOB, decisions

The numbers tie out: the summary on slide 3 matches slides 5, 12 and 14. Click any slide to see it full screen.

01Opening

Cover

Question it answers: Which cycle is this meeting, and what is the last closed month?
What it needs: The cycle month, the closing month and the company or business unit. Here: November cycle, October close.
Where the data comes from: The S&OP calendar for the year, agreed with Finance.
Common mistake: Mixing up the cycle month and the data month. Readers need to know right away that October is the last month with actual sales.

02Opening

How the cycle works

Question it answers: Who decides what, and in which week?
What it needs: The 4 steps, who joins each one and what each meeting hands to the next.
Where the data comes from: The process design: calendar and attendee list for each forum.
Common mistake: Skipping the Supply Review and taking the forecast straight to leadership. They get demand without knowing whether it can be supplied.

03Opening

Executive summary

Question it answers: What changed this month, and what does leadership need to decide today?
What it needs: 4 numbers (sales attainment year to date, accuracy, average coverage and backorder), 3 short messages and the decisions needed.
Where the data comes from: The deck itself. It is the last slide I build, and every number has to match its source slide.
Common mistake: Messages that only describe the chart. "Sales 2% below" does not help; "line D below its launch curve, review the launch forecast" does.

04Opening

Actions from last month

Question it answers: Was what we agreed last cycle actually done?
What it needs: Action, owner, due date and status. A late action that does not close before the executive meeting becomes a decision point.
Where the data comes from: The minutes of last month's executive meeting.
Common mistake: Never coming back to the actions. Without this slide, the same decision gets made again every month.

05Demand

Sales actual vs forecast

Question it answers: Did we sell what we planned this year?
What it needs: Actual and forecast year to date, the gap, the plan still open until December and attainment by line.
Where the data comes from: Invoiced sales from the ERP and the frozen forecast for each month, kept in the planning system or in a snapshot table.
Common mistake: Comparing against the latest forecast, which was already adjusted toward the actual. The yardstick is the forecast that was valid when the month started.
How to calculate: Attainment = actual ÷ forecast. In the template, green from 97%, orange between 90% and 97% and red below 90%.

06Demand

Forecast waterfall

Question it answers: How much did each month's number change from one cycle to the next, and which version was valid when the purchase was placed?
What it needs: One row per cycle, one column per forecast month, the lag 1 and lag 3 diagonals marked, the actual at the bottom and the change from lag 3 to lag 1.
Where the data comes from: A frozen copy of the consensus forecast, saved with its date at the end of every cycle, in the planning system or a history table. Without that copy there is no waterfall.
Common mistake: Overwriting the forecast every month. Without the old versions you cannot measure lag, explain why a purchase went wrong or prove that the adjustment helped.
How to calculate: Lag = month forecast − cycle month: the August cycle forecasting October is lag 2. Change = (lag 1 − lag 3) ÷ lag 3 shows how much the number moved after the purchase was placed. Here December went up 5.7%, driven by the line C promotion and line A.

07Demand

Forecast accuracy and bias

Question it answers: Is the forecast right? And when it misses, does it always miss the same way?
What it needs: Monthly MAPE at lag 1 and lag 3, bias by line and a plain reading: balanced, over-forecasting or under-forecasting.
Where the data comes from: The diagonals of the waterfall on slide 6 (without versions saved every month there is no lag) and actual sales by SKU.
Common mistake: Measuring MAPE on the company total. One SKU above offsets another below and the error disappears. Here it is calculated by SKU and month.
How to calculate: MAPE = Σ|actual − forecast| ÷ Σ actual and Bias = Σ(forecast − actual) ÷ Σ actual. Lag 1 is the forecast made one month before; lag 3, three months before. I read bias between −5% and +5% as balanced.

08Demand

Statistical vs adjusted (FVA)

Question it answers: Did the team's adjustment improve or hurt the statistical forecast?
What it needs: MAPE of the statistical and final forecast by line, the FVA of each and where the adjustment made things worse.
Where the data comes from: The statistical and final forecasts saved side by side in the same snapshot, in the planning system or the team's table.
Common mistake: Using negative FVA to blame people. It points to the assumption to review: here, the line C promotion and the line E phase-out.
How to calculate: FVA = statistical MAPE − final MAPE, in percentage points. Positive: the adjustment helped. Negative: the statistical forecast alone would have missed less.

09Demand

Forecast vs forecast

Question it answers: What changed in the plan since last cycle, and why?
What it needs: The current cycle against the previous one for the next 6 months and the largest increases and cuts by SKU, each with its reason.
Where the data comes from: Two rows of the waterfall: this cycle's and last cycle's.
Common mistake: Showing the change without the reason. Every big swing needs someone who can explain where it came from.

10Demand

Launches, promotions and phase-outs

Question it answers: Which events change volume in the coming months?
What it needs: SKU, event type, start, volume over the next 6 months and status (approved, under review, proposed).
Where the data comes from: Marketing, for launches and promotions, and the portfolio committee, for phase-outs.
Common mistake: A launch with no analog product. Without the curve of a similar item, the launch forecast is a guess.

11Demand

Risks and opportunities

Question it answers: What could make the forecast miss, up or down?
What it needs: Event, volume, probability, weighted volume and owner, with opportunities and risks side by side.
Where the data comes from: The Demand Review, with Sales, Marketing, Import and Production.
Common mistake: Putting the risk straight into the forecast. It stays outside the number and only goes in when its probability passes the agreed threshold.

12Supply

Inventory coverage (MOH)

Question it answers: How many months of sales does the stock cover, and where is it outside the band?
What it needs: By line and SKU: warehouse, in transit, total, average forecast for the next 3 months, MOH, status, last month's MOH and the action requested.
Where the data comes from: The month-end stock position in the ERP (in SAP, MB52 or MMBE), open purchase orders in transit and the cycle forecast.
Common mistake: Dividing by past sales. Coverage looks forward: today's stock against the forecast for the coming months.
How to calculate: MOH = (warehouse + in transit) ÷ average forecast for the next 3 months. Template band: 1.5 to 4 months. Below is stockout risk; above, excess and write-off risk.

13Supply

Channel inventory (SIT)

Question it answers: Are distributors buying to sell or buying to stock up?
What it needs: Days of channel inventory month by month, with the target band and a reading of what to do with next month's sell-in.
Where the data comes from: The stock and sell-out report distributors send every month.
Common mistake: Looking only at sell-in. Pushing volume into a full channel turns into lower orders or returns in the following months.
How to calculate: Channel days = distributor stock ÷ average sell-out of the last 3 months × 30. Template band: 30 to 45 days.

14Supply

Backorder

Question it answers: How much customer demand is stuck, and where?
What it needs: Units, value, customers and lines affected, the monthly trend and the SKUs with the most backorder.
Where the data comes from: The open order book in the ERP.
Common mistake: Showing only the total. Each SKU needs its cause (capacity, a late import or a low forecast) and the expected date to ship.

15Supply

Import pipeline

Question it answers: Where is each shipment, and when does it become available stock?
What it needs: Order, SKU, quantity, ETA and the step of each shipment: ready at origin, in transit, customs, receiving and available.
Where the data comes from: Import follow-up with the freight forwarder and the customs broker.
Common mistake: Using the ETA as the date stock is available. Between the vessel arriving and the product being released there is still customs and receiving, and that is where delays usually show up.

16Supply

Purchase and production plan

Question it answers: How much to buy and make in the next 3 months, and can the plant handle it?
What it needs: Suggested order by line and month, available hours against planned hours and the utilization of each month.
Where the data comes from: The cycle forecast, projected stock and plant capacity in hours, taken from the MPS.
Common mistake: Closing the plan without checking capacity. Here December goes over 100%, and the choice between building ahead and approving overtime has to come out of this meeting.
How to calculate: Order = max(0; 1.5 × max(fcst N; fcst N+1) + fcst N − projected stock), where 1.5 is the target coverage in months. Utilization = planned hours ÷ available hours.

17Finance

Revenue vs budget

Question it answers: Is the volume sold turning into the budgeted revenue?
What it needs: Revenue and budget year to date, attainment, how much of the full-year budget is already done and the gap by line.
Where the data comes from: Revenue from the accounting close and the annual budget, together with Finance.
Common mistake: Volume and revenue telling different stories with no explanation. When volume is close to plan and revenue is not, the answer is usually mix or price, and the slide has to say which.
How to calculate: Attainment = revenue ÷ budget for the period and Gap = revenue − budget, by line and in total.

18Finance

Inventory at risk (SLOB)

Question it answers: How much stock could turn into a loss, how much already hits the income statement as a provision, and what do we do with each bucket?
What it needs: Units and value at cost by expiry bucket, plus slow-moving stock, the provision for each bucket and a strategy with an owner. Units show the sales effort, value shows the impact on the result: the 7 to 12 month bucket has 26% of the units and only 18% of the value.
Where the data comes from: Stock by lot with its expiry date, the forecast for the next 6 months and the standard cost of each SKU. The provision percentage comes from the Finance policy.
Common mistake: Calculating from the SKU's total stock instead of by lot. A SKU can have healthy coverage and still hold a lot that expires before it sells. Another mistake is showing only the value: without a strategy and an owner, the up-to-3-month bucket becomes a write-off next month.
How to calculate: Units at risk = lot stock − forecast it covers by FEFO until the cut-off date, where the cut-off is the expiry date minus the minimum shelf life the customer accepts (3 months here). Then Value = units × unit cost and Provision = value × bucket % (100%, 50%, 25% and 10% in this example). For lot L2410 of E-502, expiring Apr/27: 1,200 − 710 = 490 units × R$ 27.50 = R$ 13,475, with a provision of R$ 6,738. A lot with long shelf life counts as slow-moving when it exceeds 6 months of coverage. The strategies follow the time left: up to 3 months, discount, donation or write-off; 4 to 6 months, transfer and campaign; 7 to 12 months, hold purchases and production; slow-moving, review the forecast and assess discontinuation.

19Closing

Decisions and escalations

Question it answers: What leaves this meeting approved?
What it needs: Decision needed, options, recommendation, owner and due date.
Where the data comes from: The gaps left over from the Demand Review and the Supply Review.
Common mistake: Bringing a problem with no recommendation. Leadership decides faster when the team arrives with a proposal.

How to adapt it to your industry

The structure of the 19 slides works for any company that plans inventory. What changes is how much weight each slide gets.

Pharma

Shelf life drives SLOB, and the buckets are by expiry. Quality release is part of the lead time, and a launch depends on regulatory approval. Tenders matter more than promotions.

Consumer goods

Sell-in vs sell-out and channel inventory (SIT) become the core of the Demand Review. Promotions and trade marketing are the biggest forecast risks.

Retail

The channel slide becomes in-store stockouts, turnover and coverage by store. Demand is by SKU and store, and replenishment weighs more than production.

Manufacturing

Capacity, setups and raw materials rule. The production plan becomes the main slide, with sequencing and material constraints.

Glossary

S&OP: Sales and Operations Planning: the monthly cycle that brings demand, supply and finance into one plan.
IBP: Integrated Business Planning: S&OP with finance and portfolio built in, over a longer horizon.
MBR: Monthly Business Review: the executive meeting of the month, where the plan is approved.
Demand Review: Meeting where demand, sales and marketing agree on the consensus forecast.
Supply Review: Meeting where supply, production and procurement check whether the forecast fits capacity and stock.
Lag: How many months before the forecast was made. Lag 1: one month before the measured month.
Waterfall: Table with every version of the forecast: one cycle per row and one forecast month per column. It is the record that makes lag, stability and accuracy measurable.
MAPE: Mean absolute percentage error, weighted by volume. Lower is better.
Bias: Whether the forecast keeps missing high (positive) or low (negative).
FVA: Forecast Value Added: how much the team's adjustment improved the statistical forecast.
MOH: Months on Hand: how many months of forecast the stock covers.
SIT: Stock in Trade: inventory sitting in the channel (distributors and customers), measured in days of sell-out.
Backorder: An accepted customer order that has not shipped yet.
SLOB: Slow-moving and Obsolete: stock that is not moving or is close to expiry.
NPI: New Product Introduction: a product launch.
EOL: End of Life: a product being phased out.

Where I did this

Sanofi (2026): Demand Planning in the IBP cycle. Demand Review in Kinaxis, MAPE, Bias and FVA at lag 1 and lag 3, channel inventory (SIT) in the MBR and ODR and SLOB dashboards for the forums.
Hering (2025 to 2026): stockouts, turnover and coverage in the company's own stores, with automated bulletins for store managers and coordinators.
Hypera Pharma (2024 to 2025): the MPS cycle and the monthly presentation of the 3-month production plan and sequencing at the MBR.
Abbott (2022 to 2024): production. KPIs for the technical team and an A3 project that gained capacity, the basis for the plan and utilization slide.

About the data: everything is fictitious, from a generic company. No number, product or slide from any company I worked for appears here.

Another projectStore Stockout Bulletin (interactive dashboard) Another projectPower Automate flows

The 4 weeks of the cycle

Close and data

Demand Review

Supply Review

Pre-S&OP and executive meeting

Slide by slide

Cover

How the cycle works

Executive summary

Actions from last month

Sales actual vs forecast

Forecast waterfall

Forecast accuracy and bias

Statistical vs adjusted (FVA)

Forecast vs forecast

Launches, promotions and phase-outs

Risks and opportunities

Inventory coverage (MOH)

Channel inventory (SIT)

Backorder

Import pipeline

Purchase and production plan

Revenue vs budget

Inventory at risk (SLOB)

Decisions and escalations

How to adapt it to your industry

Pharma

Consumer goods

Retail

Manufacturing

Glossary

Where I did this