What Is S&OP? Sales and Operations Planning for Finance Teams

In most companies, two plans describe the same year and disagree.

One is built by supply chain and commercial teams: units, product families, production schedules, inventory targets. The other is built by finance: revenue, margin, cash, the number that goes to the board. They are constructed from different assumptions, on different calendars, at different levels of detail, by people who rarely sit in the same meeting. When they disagree – and they always disagree – the gap is closed at the end, manually, by someone who reconciles them into a single submission that neither side fully recognises.

Sales and operations planning exists to prevent exactly this. Whether it does depends almost entirely on one thing: whether finance is in the process or downstream of it.

This article is written for the finance side. Most S&OP material is written for supply chain audiences and treats the financial view as an output. That framing is the source of the problem.

What is S&OP?

Sales and operations planning is a monthly, cross-functional process that reconciles demand, supply and financial plans into a single agreed operating plan. It is tactical rather than strategic in horizon, typically covering up to eighteen months, and it works at the product family level rather than at SKU level.

The term and its modern meaning date to the 1980s and are generally attributed to Richard Ling, then a consultant at Oliver Wight. The process it describes has been remarkably stable since: the mechanics have changed, the logic has not.

The canonical process has five steps:

1. Data gathering – collecting actuals, current performance and the starting position
2. Demand planning – building the unconsensed demand forecast
3. Supply planning – testing whether that demand can be met, and at what cost
4. Pre-meeting – resolving conflicts and preparing decisions and scenarios
5. Executive meeting – taking the decisions that cannot be taken below that level, and agreeing one plan

Participants typically span sales, marketing, operations, manufacturing, materials management, engineering and finance. The intent is that these functions work from the same assumptions rather than from their own.

Four terms worth using precisely

Much of the friction between finance and operations in S&OP is vocabulary. Four terms carry most of the weight:

• Unconstrained demand – what the market would buy if supply were unlimited. It is the honest starting point, and the one most often skipped, because commercial teams tend to submit what they believe can be supplied rather than what customers want.
• Consensus demand plan – the demand plan agreed across sales, marketing and finance after the demand review, with its assumptions recorded.
• Constrained plan – what can actually be supplied, given capacity, materials, lead times and inventory policy. The difference between unconstrained demand and the constrained plan is the revenue the business is choosing, or being forced, not to capture – and it is a financial number.
• Gap to plan – the difference between the agreed operating plan and the budget or financial commitment, expressed in units and in money.

That is the textbook description, and it is available in a hundred places. What follows is what it means if you carry the P&L.

The five steps, and what finance should actually be doing in each

The most common failure in S&OP is not that finance is absent. It is that finance attends the executive meeting and nothing before it – which means finance arrives to validate a plan it had no part in constructing, and either rubber-stamps it or rejects it too late for anything to change.

Step	What usually happens	What finance should be doing
1. Data gathering	Supply chain pulls volumes; finance pulls actuals separately	Agree one set of actuals and one product family hierarchy, so both views start from the same base
2. Demand planning	A unit forecast is produced; price and mix assumptions live elsewhere	Own the price, discount and mix assumptions that convert units into revenue – and attach them to the same plan version
3. Supply planning	Feasibility is assessed in units, capacity and lead times	Price the supply options: what each scenario does to cost of goods, inventory, working capital and cash
4. Pre-meeting	Gaps and conflicts are surfaced operationally	Quantify the gap to the financial plan in money, and bring the bridge, not the complaint
5. Executive meeting	A volume plan is approved	Present the volume decision and its financial consequence as one number set, so the decision is made with both visible

The change this table describes is not organisational. It is a data change: the financial assumptions have to live on the same plan, at the same version, as the operational ones. Everything else follows from that, and nothing works without it.

What a monthly S&OP cycle looks like

The five steps run on a monthly calendar that starts when the books close. The pattern below is common across industries; the exact timing varies, but the sequence and the dependency between steps do not.

Timing	Step	Operational output	Finance deliverable
First working days after close	Data gathering	Actuals by family, forecast accuracy, open issues	Closed actuals mapped to the family hierarchy, with price and cost actuals
Week 2	Demand review	Consensus demand plan and assumptions log	Revenue translation of the demand plan and the gap to budget
Week 3	Supply review	Constrained plan, capacity and inventory projections	Cost of goods, inventory value and working capital by scenario
Early week 4	Pre-meeting	Recommendations, open decisions, scenarios	Financial bridge: this cycle against last cycle and against budget, by driver
Week 4	Executive meeting	Decisions and the approved plan	Updated financial outlook as an output of the meeting, not a follow-up to it

The column that matters is the last one. In most organisations it is either empty until week 4, or produced a week after the meeting – at which point the decisions it should have informed have already been taken.

S&OP vs Integrated Business Planning

The distinction comes up in every vendor conversation and is worth settling quickly.

Integrated Business Planning is the evolution of S&OP, and the term also originates with Oliver Wight. Three things were added to the original process over time:

• Financial integration – added in the late 1980s, and described by Oliver Wight as one of the critical success factors of effective S&OP
• Product and portfolio management – added in the late 1990s, bringing product development into the same planning cycle rather than running it alongside
• A strategic horizon – where classic S&OP looks out over months, IBP requires the business to look out over at least 24 months, and most organisations using it look out 36 months or more

On the financial point, Oliver Wight’s own description of conventional S&OP is unusually direct: finance is not at the table, and the consequences are “multiple sets of numbers, lack of trust, sandbagging, and bias.” Under IBP, finance has a role throughout the entire process, so that everyone understands the financial implications of the bottom-up plans.

That is the whole distinction, stated by the people who named both processes. It is worth keeping in mind whenever a maturity model or a software category is being discussed: the difference between S&OP and IBP is mostly the difference between having finance in the room and not.

The practical advice is to ignore the label. Nobody’s plan improves because the meeting was renamed. The test is simpler: ask whether the volume plan and the financial plan are the same plan. If they are two documents, the process is S&OP at whatever stage of maturity, regardless of what it is called.

Why the operational plan and the financial plan disagree

Four structural reasons, none of which are anyone’s fault.

Units and value are different languages. A demand plan is expressed in cases, units or tonnes. A P&L is expressed in currency, net of discounts, rebates, returns and freight. The translation between them is a model, and in most companies that model lives in one analyst’s spreadsheet.

The calendars differ. S&OP is monthly and rolling. The budget is annual and fixed. The statutory close is periodic and backward-looking. A rolling operational plan and a fixed financial target diverge by construction, and the divergence is rarely reported as such – it is absorbed.

The granularity differs. S&OP works at product family level. Finance reports by legal entity, cost centre and account. Neither hierarchy maps cleanly onto the other, and the mapping is usually maintained manually.

Ownership differs. The demand plan belongs to commercial, the supply plan to operations, the financial plan to finance. Each is incentivised differently. Sandbagging on one side and stretch targets on the other are not data problems; they are what happens when three plans can coexist without being reconciled in public.

The translation layer: turning volume into value

This is the part that most S&OP material skips, and it is the part finance actually needs. Every operational variable in the plan has a financial consequence. The table below is the bridge — and building it explicitly, as a model rather than as a reconciliation, is what “financially integrated” means in practice.

Operational variable	Financial consequence	Where it usually breaks
Unit volume by family	Gross revenue	Price and mix assumptions are held outside the plan, so volume changes don’t reprice
Promotions and trade spend	Gross-to-net deductions	The volume uplift is planned; the deduction is accrued separately and later
Production plan	Cost of goods sold, absorption, variances	Standard costs are static while the plan moves
Capacity decisions	Capex and depreciation	Capacity is agreed in the S&OP meeting, funded in a different process
Inventory and safety stock	Working capital and cash	Targets are set in units, reported in money, reconciled by nobody
Service level targets	Lost sales, penalties, expedite cost	Only the cost of holding stock is modelled, never the cost of not holding it
Lead times and supply mode	Cash conversion cycle, freight cost	Airfreight-versus-sea decisions are made operationally, felt financially
Launches and phase-outs	Cannibalisation, obsolescence write-offs	Phase-out volumes are planned; the write-off lands as a surprise
Channel and customer mix	Margin mix, terms, receivables	Mix shift is visible in units long before it is visible in margin
FX and input prices	Landed cost, hedging exposure	Held in treasury’s model, not the operating plan

Two observations about this table are worth stating plainly.

First, every row is a place where a decision is taken on one side and paid for on the other. That is the actual cost of a disconnected S&OP: not inaccuracy, but decisions taken without their price attached.

Second, the rows are not equally hard. Volume-to-revenue is straightforward. Inventory-to-cash and service-level-to-lost-margin are the two that most organisations never build, and they are the two that executives ask about most often – usually in the meeting, usually without getting an answer.

What it looks like when operations and finance share one plan

The translation table is easier to believe with a case in front of it.

Charles River Laboratories, a global provider of preclinical and clinical laboratory services, faced a demand surge, scarce raw materials and disconnected legacy systems during the COVID-19 pandemic. 

The company implemented CCH® Tagetik Supply Chain Planning and Sales & Operations Planning, with Inulta as implementation partner, integrating more than 14 data sources – ERPs, spreadsheets and manual files – into a single planning environment with scenario modelling across supply chain and finance.

The published results read almost row by row against the table above:

• 84% reduction in backorders, 68% fewer backorder days and 86% fewer items on backorder – the service-level row, and the lost sales behind it
• 75% projected reduction in expiring inventory – the launches-and-phase-outs row, and the write-offs behind it
• 40% savings in shipping costs for select product lines – the supply-mode row

Two years later, the company extended the solution to Integrated Business Planning, explicitly to break down the barriers between finance and operations – which is the S&OP-to-IBP journey described above, taken in the right order. The Charles River case study describes the programme in more detail.

Six signs your S&OP is not financially integrated

A diagnostic, in rough order of severity:

1. The S&OP number and the number submitted to group differ, and both are defensible. Two plans exist; one of them is ceremonial.
2. The financial view is produced after the executive meeting, not inside it. Finance is validating, not deciding.
3. Nobody can say what a 5% demand shift does to cash. The bridge from volume to working capital does not exist.
4. Inventory targets are set in units and reported in money by a different team. The two never meet in one version of the plan.
5. The S&OP pack contains volumes and the board pack contains margin, with no bridge between them that anyone can reproduce.
6. Scenario questions are answered next month. If “what if we lose the biggest customer” takes a cycle to answer, the plan is a report rather than a model – the gap that scenario planning for CFOs exists to close.

Any two of these is normal. Any four means the process is producing agreement without producing a plan.

What to measure – operationally and financially

An S&OP process that is measured only on forecast accuracy will optimise forecast accuracy. The measures below pair each operational indicator with the financial one it eventually becomes.

Area	Operational measures	Financial measures
Demand	Forecast accuracy, forecast bias	Revenue against plan; price, volume and mix variance
Supply	Plan attainment, capacity utilisation	Cost of goods against standard, production variances
Inventory	Days of supply, inventory turns, excess and obsolete stock	Inventory value, working capital days, write-offs
Service	Fill rate or on-time-in-full, backorders	Lost margin, penalties, expedite and freight cost
Plan quality	Plan stability between cycles, decision latency	Gross margin against plan, cash conversion cycle

One measure deserves particular attention from finance: forecast bias. Accuracy tells you how far the forecast was from reality; bias tells you which way it leans. A forecast can be reasonably accurate on average and still be systematically high – and systematic over-forecasting becomes excess inventory and write-offs, while systematic under-forecasting becomes expedite cost and lost sales. For finance, a biased forecast is more expensive than an inaccurate one, because its errors do not cancel out.

Horizon, cadence and granularity

Three mechanical choices determine whether the process can carry financial weight.

Horizon. Classic S&OP runs to eighteen months. IBP pushes to 24–36. The financial argument for the longer horizon is specific: an eighteen-month horizon cannot see a full capex cycle, a product launch ramp, or the second year of a multi-year contract. If S&OP is meant to inform investment decisions, its horizon has to outrun them.

Cadence. Monthly is standard, and it is standard for a good reason: it matches the reporting rhythm, which means the plan and the actuals meet regularly and the comparison is cheap. Event-driven cycles are a useful addition during volatility, not a replacement.

Granularity. Plan at family level, execute at SKU level. Attempting S&OP at SKU level produces a process that cannot be completed inside a month and decisions nobody can see. The financial corollary is that the family hierarchy must map to the reporting hierarchy – if it does not, every cycle begins with a mapping exercise, and that exercise is where the two plans start to drift.

The budget relationship. The rolling plan and the annual budget will diverge; the question is whether the divergence is visible. Mature processes report the gap to budget as a standing output of the S&OP cycle rather than discovering it at quarter end. Some organisations go further and let the rolling forecast replace the budget as the basis for management decisions, keeping the budget only as the commitment it was approved as. A French automotive manufacturer that deployed rolling forecasts across its vehicle lines and regions, bringing operational and financial data into one planning process, reports more than €20M in annual cost savings from the programme.

How S&OP changes by industry

The process is the same everywhere. What changes is which rows of the translation table carry the weight.

Pharma and life sciences. Demand is separated from the patient by distributors and wholesalers, so the plan has to distinguish sell-in – shipments to the channel – from sell-out – what the channel actually dispenses – and track distributor stock coverage between the two. Shelf life turns inventory decisions into write-off decisions, tender-driven markets make demand lumpy, and regulatory approval timing moves launch dates in ways the commercial plan cannot control. The finance rows that dominate are obsolescence, gross-to-net and working capital. More on this under finance transformation for healthcare and pharma.

Manufacturing. Capacity, changeovers and component lead times constrain supply, and the production plan drives cost of goods through absorption and variances. The rows that dominate are production cost, capacity and capex. The manufacturing application is covered under manufacturing finance transformation.

Automotive. A sharper version of manufacturing: long component lead times, model-year and vehicle-line lifecycles, platform sharing across brands and plants, and capacity decisions that commit capital years ahead. Horizon matters more here than anywhere, because an eighteen-month S&OP cannot see a vehicle programme. See automotive finance transformation.

Consumer goods and retail. Promotions, seasonality and channel mix dominate. The volume uplift from a promotion and the gross-to-net deduction that pays for it are planned by different people at different times, and omnichannel fulfilment means the same unit carries a different cost depending on how it reaches the customer. The rows that matter most are promotions, channel mix and inventory. See retail finance transformation.

Maturity: where most organisations actually are

The widely used five-stage model runs from no S&OP process, through reactive, standard and advanced, to proactive – assessed across meetings, organisation, measurement, IT and plan integration. A six-level variant runs from undeveloped to proactive across process effectiveness, efficiency, people and IT.

The honest position: most organisations that believe they run S&OP are at stage two or three. There is a monthly meeting, there is a demand plan, the meeting has an agenda, and the numbers are broadly trusted operationally. The jump that changes outcomes is not to a more sophisticated forecast. It is the jump from a plan that is agreed in units to a plan that is agreed in units and money at the same time, which is precisely the dimension the maturity models call plan integration and precisely the one that is hardest to buy.

The barriers are also well documented and unsurprising: siloed culture, inadequate systems, and the difficulty of reaching cross-departmental decisions. Two of those three are not technology problems, which is the right context for what follows.

Where AI helps in S&OP – and the trap

AI in planning is now a crowded claim, so it is worth separating what is working from what is being sold.

What genuinely works today:

• Statistical and machine-learning baseline forecasting. Producing a better unconstrained baseline from history, seasonality and correlated signals is a solved problem in the sense that the techniques are proven and widely deployed.
• Demand sensing. Using short-horizon signals to adjust near-term demand faster than a monthly cycle allows.
• Exception management. Surfacing the plan lines that need human attention, rather than asking planners to review everything.
• Bias detection. Identifying which products, regions or forecasters consistently lean high or low — the measure finance cares about most, and the one that is hardest to see by inspection.
• Scenario generation. Producing and comparing many more scenarios than a team would build by hand, which is what makes the executive meeting a decision forum rather than a review.
• Anomaly detection. Catching the data quality failures that silently corrupt a plan.

CCH® Tagetik’s supply chain planning capability sits in this territory, describing AI-driven predictive forecasting in demand planning, AI-based predictive intelligence for building probabilistic plans using risk-adjusted predictions, and the modelling of exceptional events across demand and supply.

What is not working yet, and the evidence here is worth quoting because it cuts against the marketing: BCG’s 2026 assessment of supply chain planning concludes that what separates leaders from laggards is not the technology itself but how planners apply advanced capabilities in evaluating trade-offs and responding under pressure. The most valuable AI applications today are foundational – forecasting, exception management, data interpretation – rather than autonomous planning. Organisations attempting to leapfrog planning maturity by means of AI alone, BCG finds, tend to struggle.

The finance-specific trap

There is a more specific version of that warning, and it is the one this article exists to make.

A gain in forecast accuracy does not become money until something translates it.

Improved accuracy is realised physically: as less safety stock for the same service level, fewer expedites, fewer write-offs, fewer lost sales. Each of those is a financial outcome, but none of them is a financial number until the plan contains the model that converts inventory into working capital, service level into lost margin, and expedite decisions into cost. Without that translation layer, a three-point improvement in forecast accuracy produces a better forecast and an unchanged P&L – or, more precisely, an improved P&L that nobody can attribute, which is functionally the same thing when the investment comes up for review.

This is why the sequencing matters. An organisation that buys a forecasting engine before it builds the volume-to-value bridge has bought a better input to a process that still cannot price its own decisions. The reverse order – bridge first, engine second – makes the engine’s benefit measurable on the day it is switched on.

Automate the reconciliation before you automate the forecast. It is the less impressive project and the one that changes the meeting.

Governance

Two disciplines carry over directly from finance and are usually missing on the planning side: version provenance – knowing which scenario was generated from which assumptions and which model version, months later, when the decision is reviewed – and auditability of AI-assisted numbers. If a plan informs guidance, capex or covenant discussions, the provenance of its numbers is not a nice-to-have. The same governance a finance team applies to the close applies here, and platforms that run planning on the same governed data model as consolidation and reporting start from a considerable advantage.

The direction the market is moving makes this more pressing, not less. The next step – agentic AI in finance, where software does not only flag an exception but prepares the response – moves AI from producing inputs to proposing decisions. In an S&OP context, that is only safe where every proposal carries its assumptions, its model version and a named approver.

How to start

A sequence that works, in order:

1. One data model. Agree the product family hierarchy and its mapping to the reporting hierarchy. This is the unglamorous prerequisite that every subsequent step depends on.
2. One set of actuals. Both sides start from the same base or nothing downstream reconciles.
3. Build the bridge. Volume to revenue, production to cost of goods, inventory to working capital. Start with the three rows of the translation table that your executives ask about most.
4. Move finance upstream. Into steps two and three, not just step five – and give the finance deliverable in each week of the cycle a named owner.
5. Measure bias, not just accuracy. It is the forecast property that costs money.
6. Then improve the forecast. Once the bridge exists, forecast improvement has somewhere to land.

Organisations that do this in reverse – engine, then process, then data model – are the ones BCG describes as attempting to leapfrog maturity, and the outcome is well documented.

Where this fits

The platform capability that connects demand, supply and production planning sits under supply chain management, and the financial planning it feeds under budgeting, planning and forecasting. Both run on the same data model as consolidation and reporting, which is what allows the operational plan and the financial plan to be one plan.

Frequently asked questions about S&OP

What is S&OP in simple terms? 

Sales and operations planning is a monthly cross-functional process that brings the demand plan, the supply plan and the financial plan into one agreed operating plan, typically over an eighteen-month horizon at product family level. Its purpose is to ensure that commercial, operational and financial decisions are made against the same set of numbers.

What is the difference between S&OP and Integrated Business Planning? 

IBP is the evolution of S&OP, and the term originates with the same source. IBP adds full financial integration, product and portfolio management, and a longer horizon of 24 to 36 months. The most practical distinction is that in conventional S&OP finance is often not part of the process, which produces multiple sets of numbers; in IBP finance participates throughout.

What is the difference between unconstrained demand and the constrained plan? 

Unconstrained demand is what the market would buy if supply were unlimited. The constrained plan is what can actually be supplied, given capacity, materials, lead times and inventory policy. The difference between the two is revenue the business is choosing, or being forced, not to capture, and it should be visible in the S&OP process in both units and money.

What is the S&OP planning horizon? 

Classic S&OP typically covers up to eighteen months on a rolling monthly basis. Integrated Business Planning extends this to at least 24 months, with many organisations planning over 36 months or more, because the longer horizon is needed for capacity, investment and portfolio decisions.

Who should own S&OP – supply chain or finance? 

Ownership of the process usually sits with operations or a dedicated planning function, and that is workable. What matters more than ownership is participation: finance needs to be present from demand planning onward rather than receiving the plan afterwards, because the assumptions that convert volume into value are finance’s assumptions.

Which KPIs should an S&OP process track? 

Operational measures such as forecast accuracy and bias, plan attainment, days of supply, service level and backorders, paired with the financial measures they become: revenue and gross margin against plan, inventory value and working capital days, write-offs, and expedite and freight cost. Forecast bias often matters more to finance than accuracy, because systematic over-forecasting turns into excess inventory and systematic under-forecasting into lost sales and expedite cost.

How does S&OP relate to the budget and the rolling forecast? 

S&OP produces a rolling operational plan; the budget is a fixed annual commitment. The two will diverge, and mature processes report that divergence as a standing output of each cycle rather than discovering it at quarter end. Where a rolling forecast exists, S&OP is the natural engine for it, because it is already monthly and already cross-functional.

Can AI replace the S&OP process? 

No. AI is effective today at forecasting, demand sensing, bias and exception detection, scenario generation and anomaly detection – all inputs to the process. Current evidence indicates that what distinguishes high-performing planning organisations is how people apply these capabilities to trade-off decisions, not the presence of the technology. The decisions S&OP exists to make are cross-functional commitments, and those still require the meeting.