Operations Planning and Control: What It Means in Practice (Not the Textbook)

Contents

• What operations planning and control actually means
• Planning: deciding what should happen
• Control: making the actual match the plan
• The loop: plan, execute, control, replan
• Worked example: a single production run
• Worked example: a purchasing cycle
• Why spreadsheets break the control half
• How an owned operations system closes the loop
• FAQ
• How OpsMavix can help

What operations planning and control actually means

Split the phrase and it explains itself. Planning is the set of decisions you make before the work happens: how much you expect to sell, how much capacity and stock you have to meet it, what to buy and when, and what order to do things in. Control is what you do during and after: watch what really happens, compare it to the plan, and act on the difference.

The distinction that matters is the direction of time. Planning points forward — it is a set of intentions about a future you cannot see. Control points at the present and the recent past — it is the honest record of what actually occurred, held up against those intentions. A business can be excellent at one and hopeless at the other, and most growing businesses are: they plan carefully in a spreadsheet at the start of the month and then have almost no live idea, mid-month, whether the plan is holding.

That asymmetry is the whole story. Planning is a document. Control is a feedback loop. Documents are easy to produce and easy to fool yourself with; feedback loops need data flowing back in near real time, and that is the part spreadsheets cannot give you.

Planning: deciding what should happen

Operations planning is four connected decisions, each feeding the next. Get the order wrong and the plan is fiction from the start.

Demand planning. How much do you expect to sell, of what, and when? This is the anchor — every other number is derived from it. It does not need to be a sophisticated statistical model; for most growing businesses a disciplined read of recent sales, known seasonality, and confirmed orders beats an elaborate forecast nobody trusts. The failure here is not imprecision, it is staleness: a demand figure set once and never revisited as the market moves. The practical starting point is covered in demand forecasting for a small business.

Capacity planning. Given that demand, can you actually make or move it? Capacity is people-hours, machine-hours, storage, and cash tied up in stock. The common mistake is planning demand in isolation and only discovering the capacity gap when the orders land. Planning means holding demand and capacity in the same view — if the forecast says 1,200 units and the line does 900 a week, that gap is a decision to make now, not a fire to fight later.

Purchasing planning. What has to be bought, in what quantity, to arrive in time — raw materials, components, finished stock for resale. This is where reorder points, minimums, and lead times turn a demand number into purchase orders. Plan it badly and you either stock out mid-run or freeze cash in stock that will not move.

Scheduling. The finest-grained layer: what happens in what order, on which resource, on which day. Which job runs first, which machine, which picker, which delivery slot. Scheduling is where the plan becomes a set of instructions someone can actually follow tomorrow morning.

Do all four and you have a plan. Notice what you do not have yet: any idea whether it is working. That is the other half.

Control: making the actual match the plan

Control is the half the textbooks name and then barely explain, because it is the half that only exists if data flows back to you. It has three moves.

Track the actual. Record what really happened — units produced, orders shipped, stock received, hours spent — as it happens, not reconstructed from memory at month-end. This is the raw material of control. Without a live actual, there is nothing to compare the plan against, and “control” collapses into a monthly post-mortem where every problem is already three weeks old.

Measure the variance. The variance is the gap between planned and actual: planned 1,200, made 1,050 — variance minus 150. Planned to receive the order Tuesday, it landed Friday — variance three days. Budgeted £6 a unit landed cost, it came in at £7.40 — variance £1.40. Variance is the signal. A plan with no variance measurement is a wish; the number only becomes useful the moment you can see how far reality has drifted from it.

Correct. Act on the variance while it is still cheap to act. Pull a job forward, chase a late supplier, move stock between locations, reprice, or — the honest option people forget — change the plan itself, because sometimes the plan was wrong and the actual is telling you so. Correction is the point of the entire exercise. Tracking and measuring that never lead to a decision are just expensive bookkeeping.

The thing to internalise: control is not a report. A report is a photograph of the past. Control is a loop that changes what you do next, and the shorter the delay between “the actual happened” and “you saw the variance”, the more of it you can still fix.

The loop: plan, execute, control, replan

Planning and control are not two departments or two phases. They are one loop that turns continuously:

1. Plan — decide what should happen (demand, capacity, purchasing, schedule).
2. Execute — do the work.
3. Control — track the actual, measure the variance against the plan.
4. Replan — feed the variance back in and adjust the next plan.

Then round again. The quality of an operation is mostly the speed of this loop — how fast a variance in step 3 reaches the plan in step 4. A business running the loop weekly with live data corrects small; a business running it monthly from a reconstructed spreadsheet corrects late and large, usually after a customer or a cash-flow squeeze has already noticed.

This is exactly why “operations control” is worth treating as a system rather than a habit — the concept and the components sit in an operations management system. Below, two concrete versions of the same loop.

Worked example: a single production run

Take a small manufacturer building 1,200 units of a product over a week. Figures are illustrative.

Plan. Demand says 1,200 for the month. Capacity says the line runs ~250 good units a shift, so five shifts. Purchasing confirms components for 1,300 are on hand (a buffer). Schedule: the run occupies Monday to Friday on line 2, with the finishing team starting Wednesday once the first batch clears.

Execute. The run starts Monday.

Control — where it lives or dies. By Tuesday lunchtime the line has produced 380 good units against a planned 500 — a shift-and-a-half in and already 120 behind, because scrap on a new component batch is running higher than assumed. That single fact is the entire game:

• With live tracking, the supervisor sees minus-120 on Tuesday. The correction is cheap: quarantine the bad component lot, pull replacement stock, and either add an overtime shift or tell sales Friday’s promise is now Monday — with two days’ notice, not on the day.
• Without it, the number surfaces on Friday when the run closes short at 1,050. Now the variance is minus-150, the finishing team has sat idle waiting for batches that never came at the planned rate, and the customer finds out the order is short on the day it was due. Same variance, radically more expensive, because the loop ran once a week instead of once a shift.

Nothing about the plan differed between those two outcomes. The only difference was how fast the actual fed back. That is the control half, and it is the half spreadsheets cannot do — a sheet cannot tell you on Tuesday what nobody has typed into it yet.

Worked example: a purchasing cycle

The same loop, in purchasing. A wholesaler reorders a fast line.

Plan. Demand: ~200 units a week. Lead time: 3 weeks. Reorder point and quantity are set so a purchase order fires when stock hits the trigger — the mechanics are in how to calculate a reorder point. The plan raises a PO for 800 units, expected in 3 weeks, to cover the gap plus a buffer.

Execute. The PO goes to the supplier.

Control. Now the plan meets reality, and there are three variances to catch, each needing live feedback:

• Did the supplier acknowledge and commit to the date? If they slip to 4 weeks, that is a variance the day it is known — not the day the stock fails to arrive.
• Did what arrived match what was ordered? 800 ordered, 760 received, 40 back-ordered is a variance that must reconcile against both the PO and the invoice — the discipline behind a purchase-order tracking system. Miss it and you pay for 800, receive 760, and find out at a stock count months later.
• Did the price and landed cost match the plan? A quoted £6 that lands at £7.40 after freight and duty is a margin variance you want to see on receipt, not at year-end.

Run this loop with live PO status and receiving tied together and each variance is a small, timely decision. Run it from a spreadsheet and a diary, and the variances only reveal themselves when the stock is short, the invoice is queried, or the margin comes in thin — always too late to do anything but absorb it.

Why spreadsheets break the control half

Spreadsheets are genuinely good at planning. You can build a demand forecast, a capacity model, a purchasing plan, and a schedule in a spreadsheet this afternoon, and for a small, single-person operation that is a perfectly honest tool. The planning half is a document, and a spreadsheet is a fine document.

The control half is where they break, for reasons that are structural, not a matter of skill:

• No live feedback. A spreadsheet only knows what someone has typed into it. The Tuesday-lunchtime “minus 120” only exists in the sheet if a human stops and enters it — so in practice the actual is always reconstructed later, and control becomes a monthly autopsy instead of a live loop.
• The plan and the actual live in different files. The plan is one tab or workbook; the actual is scattered across an accounts package, a warehouse, someone’s inbox, and a WhatsApp thread. Measuring variance means manually reconciling sources that were never designed to line up — so mostly, nobody does.
• No single version of the truth. Three people with three copies produce three different actuals, and the argument about whose number is right is time that should have gone into correcting the variance. The same disease behind stock discrepancies.
• No alerts. A spreadsheet cannot tap you on the shoulder. It will happily show a line 120 units behind, or a PO two weeks overdue, and say nothing — the variance sits there until a human happens to look, which is usually after it has cost something.

The net effect: businesses on spreadsheets plan well and control blind. They set good intentions at the start of the month and then run half the loop, discovering variances only when reality forces the issue. That is not a discipline problem you can train away. It is the tool being incapable of the second half of the job.

How an owned operations system closes the loop

Closing the loop means one thing: the actual flows back fast enough, and next to the plan, that variance becomes visible while it is still cheap to fix. That is what an owned operations system is for — not replacing the planning you already do well, but doing the control half the spreadsheet structurally cannot.

In practice that means:

• One live record of the actual. Production output, stock movements, orders, and receipts recorded as they happen, in one place, so the actual is never reconstructed from memory.
• The plan and the actual in the same view. Planned versus made, ordered versus received, budgeted versus landed — variance calculated for you, not assembled by hand across four files.
• Alerts on the variance, not just the data. The system tells you the run is 120 behind on Tuesday and the PO is two weeks overdue — pushing the exception to you instead of waiting to be found.
• A short loop. Because the actual is live and the variance is automatic, replanning happens on the timescale of the operation — a shift, a day — not the accounting calendar.

This is the point where “operations planning and control” stops being a phrase in a textbook and becomes a thing your business actually does, continuously, without heroics. It is the same argument made in full in an operations control system — an owned system, right-sized for a business too messy for spreadsheets but not ready for a full ERP, that closes the loop instead of leaving the control half to chance.

FAQ

What is the difference between operations planning and operations control?

Planning decides what should happen before the work starts — expected demand, available capacity, what to purchase, and the schedule. Control makes the actual match that plan while and after the work happens — tracking what really occurred, measuring the variance against the plan, and correcting. Planning points forward and produces a document; control points at the present and produces decisions. They are two halves of one continuous loop, not two separate activities.

What are the main steps of operations planning and control?

Four, running as a loop: plan (decide demand, capacity, purchasing, schedule), execute (do the work), control (track the actual and measure the variance against the plan), and replan (feed the variance back into the next plan). The quality of an operation is largely the speed of this loop — how quickly a variance you spot in the control step reaches the next plan.

Why do spreadsheets struggle with operations control?

Spreadsheets are fine for the planning half — building a forecast, a capacity model, a schedule. They break on the control half because they only know what someone types in, so the actual is always reconstructed late; the plan and the actual live in separate files that must be reconciled by hand; multiple copies produce conflicting versions of the truth; and a spreadsheet cannot alert you to a variance. The result is planning well and controlling blind.

Do I need special software for operations planning and control?

Not to plan — a disciplined spreadsheet covers the planning half for a small operation. You need a system for the control half, once your loop is too fast or too shared for a spreadsheet to keep up: when variances keep surfacing too late to fix, when several people touch the same numbers, or when the plan and the actual live in too many places to reconcile by hand. At that point an owned operations system closes the loop that the spreadsheet leaves half-open.

How does operations planning and control relate to production tracking?

Production tracking is the control half applied to making things — the live actual (units produced, scrap, time) fed back against the production plan so variances show up per shift, not per month. It is one concrete instance of the wider planning-and-control loop, and the part that turns a production schedule from a hopeful document into something you can steer while the run is still going.

How OpsMavix can help

The planning half of this loop you can probably do already — a forecast, a capacity view, a purchasing plan, a schedule, all in a spreadsheet, and for a small operation that is honest and fine. What OpsMavix builds is the control half the spreadsheet structurally cannot: owned production and operations tracking that records the actual as it happens, holds it next to the plan, calculates the variance for you, and alerts you while the gap is still cheap to close.

If your problem is that you plan carefully at the start of the month and then fly blind until the numbers come back too late to act on, that is not a discipline failure — it is a spreadsheet running only half the loop. Closing it is the whole job.

Sources

• Operations management — Wikipedia
• Production planning — Wikipedia
• Production control — Wikipedia
• Feedback — Wikipedia