Integrated Business Planning: Where the Value Hides
The most expensive planning mistakes happen between functions, not inside them.
Three examples from my experience.
An industrial group of several dozen plants needed five weeks to update its annual plan — with nine months of the year still ahead. The process ran sequentially: production recalculated volumes, maintenance shifted schedules, procurement rebuilt requirements, and only then did finance consolidate the numbers. By the time the changes had made it through the approval cycle, the assumptions had already changed again.
At another plant, loaded railcars sat on the company's own tracks for up to twenty days. The cargo had been accepted, the production schedule had been met, and dispatch was waiting for its planned window. Every function had delivered against its targets, while working capital sat on rails — and no function owned the cost of the delay.
In a construction group, identical equipment sat idle in one subsidiary while another rented the same machines on the market. Both directors acted rationally: one protected capacity for an expected peak, the other solved an immediate need the fastest available way.
I worked through each of these situations from the inside. No one was at fault. Every plan was reasonable, every manager was doing their job.
No one is to blame. Value is lost where no one owns the trade-off.
The usual conclusion from cases like these is that the company needs "integrated business planning" — and the conversation quickly turns to software. My conclusion after two decades inside these management systems is different: there is no single model of integrated business planning that fits every business. The framework can be common. The planning logic cannot. The critical trade-offs differ by business — resource versus market value in mining, production versus maintenance in heavy industry, sales versus manufacturing in make-to-order, projects versus shared resources in construction.
So the question is not whether a business needs integrated business planning. The question is what needs to be integrated around what. To answer it, start with what ultimately constrains your ability to deliver the plan: available resources, production capacity, demand, committed orders, or a portfolio of projects. That constraint determines where your most important trade-offs sit.
Why the classic playbook skips the first question
Management literature has covered this field from several angles, and each provides a useful piece of the answer. S&OP and its more mature form, Integrated Business Planning, explain how to synchronize plans through a recurring cycle, shared demand and supply reviews, and financial reconciliation. Theory of Constraints explains where the system's bottleneck sits and argues that the rest of the system should be managed around it. Supply chain theory explains where the customer order decoupling point lies — where the chain switches from forecast-driven planning to firm orders.
None of them answers the question a chief executive should ask first: what should the planning cycle be built around in my business? Each methodology assumes that answer is already known. In my experience it rarely is. Production knows its capacity, sales knows its pipeline, finance knows its budget, project managers know their commitments — but no one has turned those facts into an explicit management choice about which constraint should drive the planning cycle.
This is not a methodology choice. It is a diagnosis.
Five planning types
Every planning cycle has a starting point — something the business must secure first if it is going to deliver the result. I call it the planning anchor.
The model at a glance.
Resource-led. Primary constraint: resource availability and quality. The plan starts from the available resource. Core conflict: resource versus market value.
Capacity-led. Primary constraint: asset capacity. The plan starts from the available capacity. Core conflict: production versus maintenance.
Demand-led. Primary constraint: demand uncertainty. The plan starts from the sales forecast. Core conflict: category economics versus assortment.
Order-led. Primary constraint: customer commitments. The plan starts from firm orders. Core conflict: sales versus manufacturing.
Project-led. Primary constraint: portfolio commitments. The plan starts from the project portfolio. Core conflict: projects versus shared resources.
These are not formal industry classifications. They are a practical way of identifying what the planning cycle is built around — and therefore where integration matters most. Three contrasting cases show how differently the same discipline works.
Resource-led. Mining starts the plan from the deposit, not from demand: how much material of what quality can be extracted in a given period. Demand does not disappear — it shapes what the resource is worth. Ore quality drives the blend, the blend drives the product mix, the product mix determines the premium the market will pay. Even the boundary between ore and waste rock is an economic decision, not a geological constant. I have seen geologists plan tonnes while commercial teams plan premiums — and the group sell a blended product at an average price, leaving the difference on the table. The value is lost between resource planning and commercial decisions.
Capacity-led and order-led. Metals and most commodity processing run on capacity. In a normal phase of the cycle the market absorbs the output, and the constraint sits in assets, feedstock and the maintenance schedule. I have seen the same failure repeatedly: maintenance pushed toward year-end to protect the current plan collides with peak demand in December, and the group pays twice — first through lost availability, then through emergency work at a multiple of the planned cost. The conflict is between production, maintenance and logistics.
Make-to-order manufacturing runs on commitments instead. The key decision is where to position the customer order decoupling point for each product line: what to build to stock, what to build to order, where to hold semi-finished inventory. That single decision sets the trade-off between working capital and delivery performance — and it is usually made once, at product launch, then never revisited while the market moves on. The trade-off is between commercial commitments and manufacturing reality.
Demand-led. Retail and consumer goods plan from forecast, and planning here is the business itself — forecasting and replenishment are automated better than anywhere else. Yet even in retail, category economics and assortment plans routinely meet in a quarterly meeting rather than in one integrated business planning cycle. The most tooled-up industry faces the same disease; the tension sits between category economics and assortment decisions.
Project-led. Construction, infrastructure and EPC invert the logic completely. The project comes first; schedules, resource requirements and budgets are derived from the portfolio of commitments. There is no statistical demand signal — there is a pipeline of potential projects weighted by probability, alongside firm contracts. And the constraints extend beyond the client: permits, expert reviews and financing tranches stop a site as effectively as a shortage of materials. The problem is competition for shared resources across projects — equipment, crews, engineering capacity and financing.
The anchor can move
A mine may spend decades planning around the resource. As the deposit matures, the economic constraint moves downstream — toward processing capacity, logistics or product quality. A producer used to selling everything it makes enters a period of oversupply and discovers that demand, not capacity, is the binding constraint. A manufacturer sees orders collapse and inventories grow, while the planning conversation stays focused on production loading, because that is how the company has always planned.
The planning system continues to work. It is just solving the wrong problem.
This is the most expensive moment: the business has moved to a different constraint, but the planning model has not moved with it.
One company, several starting points
The starting point usually sits below the company level — at the level of a product line or business segment. Inside one machine-building plant, catalogue products are forecast-led while engineered-to-order products are order-led. In a full-cycle mining and metals group, the balance between resource and capacity shifts within the year. An industrial holding with a major capital program runs a capacity-led operating plan and a project-led investment portfolio at the same time — competing for the same equipment, people and financing.
The organizational consequence: the group does not need one planning logic. It needs one management framework capable of connecting different planning logics. The standard mistake in large programs is to read "one IBP" as "one planning model for the entire group" — consistency at the expense of economic relevance. Share the framework, the data principles and the decision rules. Let the planning logic differ where the economics differ.
What has to happen in every planning cycle
Not a system, not a report — a repeating management cycle:
Change → Recalculate → Surface conflicts → Decide → Execute → Learn
A change occurs in the market, the resource base, capacity or the project portfolio. Functions recalculate their plans, and conflicts become visible where the plans meet — which is healthy, because a visible conflict means the plans are actually connected. The conflict is resolved by agreed decision rules; without those rules, every difficult trade-off eventually lands on the CEO's desk. The decision is translated into operating plans, actual performance feeds into the next cycle, and the decision rules themselves are refined as the organization learns.
That is the mechanism. Technology matters, but it comes later.
Speed matters more than the calendar
Return to the five weeks from the first example. The problem is not that five weeks is a long time. The problem is the relationship between two speeds: how fast the company can change its plan, and how fast its key assumptions are changing. When a company recalculates its plan more slowly than its critical inputs change, planning stops being a management tool and becomes historical reporting.
There is no universal benchmark — the relevant metric is the ratio between the planning cycle and the speed of change in the business. In my experience, a recalculation cycle of more than two weeks in a large, volatile environment is a warning sign. The annual budget is the extreme version of the same problem: frozen for a year, by March it describes a company that no longer exists.
Where the value is lost
"Cross-functional problems cost money" means little without arithmetic. The economics of local optimization show up in a short list of places: excess inventory, idle capacity, emergency purchases and rentals, lost margin, underused assets, broken commitments. Each can be measured from data the company already has.
The railcars from the second example are a simple illustration. Twenty days of idle time, multiplied by the cost of capital and the number of cars, is a measurable loss — invisible only because it sat between functions. The equipment example works the same way: external rental cost versus the cost of underutilizing the owned fleet. Two numbers existed; they lived in different companies. The same pattern appears in cash. One subsidiary holds surplus cash on deposit while another borrows for working capital at the same time; the spread between the two rates, multiplied by the offsetting volumes, is a direct loss no individual P&L shows. Each finance director makes a rational local decision. The group still loses money.
What is missing is not data. It is the habit of measuring economics across organizational boundaries. In my experience, this is exactly where the value is lost — and where integrated business planning earns its keep: not by producing another report, but by making these trade-offs visible and manageable.
Who decides what matters more?
Every cross-functional conflict eventually comes down to one question: what matters more — and who has the authority to decide? Production wants maximum utilization. Maintenance wants a shutdown. Sales wants the customer deadline. Finance wants lower capital expenditure. Logistics wants a stable flow. Everyone is right within their own function.
The missing element is rarely a process owner — most large companies have documented processes everywhere. The missing element is decision rights. Every critical trade-off needs more than a process description: it needs a person with the mandate to choose what to sacrifice, and accountability for the economics of that choice. Without that mandate, the whole group eventually gets planned by one person — the chief executive, in the time left over from everything else.
The logic of the whole model fits in one line: find the constraint, expose the trade-off, assign the decision rights, and build the planning cycle around them.
Technology comes after the rules
This deserves its own article, so briefly. When decision rules are undefined, digital tools accelerate the argument. When data is unreliable, they scale the error. When the process works, technology turns a manual planning cycle into a managed system. An ERP does not eliminate conflicts between plans — it lets them surface faster and at greater scale. I watched a company implement a sophisticated scheduling system: the schedule became impeccable, and deadlines kept slipping, because the resources behind the schedule were still allocated by voice at the morning meeting.
The same sequence applies to AI. Decision rules first. Trusted data second. Technology third. AI where it creates additional leverage. Technology does not replace the management process — it makes a good one scalable.
Five questions instead of a summary
1. How long does it take to recalculate your plan — and how often do the key assumptions change within that period? If the business changes faster than the planning cycle, the plan is already behind reality.
2. Does the annual budget survive beyond March — or does "rebalancing within approved parameters" take over? If the budget cannot move, decisions are already being made outside it. The organization simply does not call it a new plan.
3. Do shortages and surpluses of the same resource — equipment, people, cash — coexist across business units? If yes, you do not have a group-wide resource balance. You have a collection of local safety cushions.
4. Can you name what drives the planning cycle in each major business segment?
5. Can you name the person who has the authority to resolve the most important cross-functional trade-offs? Not a committee. A decision-maker.
If the answer to any of these questions is "I don't know," you do not have a planning automation problem.
You have a decision-making problem.
This article opens a series on integrated business planning. Next: what a holding company can learn from Gosplan, post-war France and modern China — and what it should never replicate. And separately: why construction holdings need IBP connected to enterprise project portfolio management.