Skip to main content
55 min readsupply planningwhat is supply planning

What Is Supply Planning?

Supply planning is the process of deciding how, when and from where the quantities in a demand or merchandise plan will be supplied: converting planned receipts into purchase orders, production bookings, raw-material commitments and inventory targets against lead times, capacity and minimums, so stock arrives where and when the plan needs it.

Supply planning is the process of deciding how, when and from where the quantities in a demand or merchandise plan will be supplied: converting planned receipts into purchase orders, production bookings, raw-material commitments and inventory targets against lead times, capacity and minimums, so that stock arrives where and when the plan needs it. The demand plan and the merchandise financial plan say what the business intends to sell and own. The supply plan says what will be ordered from whom, when each commitment has to be made and when the goods will land. It is the layer at which a plan stops being a number in a model and becomes a promise to a mill, a factory and a freight forwarder.

This guide defines supply planning as a discipline. It covers where supply planning sits between the plan and execution; how it differs from demand planning and merchandise planning on one side and from supply chain planning, S&OP and replenishment planning on the other; the process step by step, from netting the receipt plan to the weekly exception review; the inputs it needs and the outputs it produces; how apparel supply planning differs from the consumer packaged goods model that standard supply planning technique assumes; how the work changes across ten retail verticals; a worked example that turns one style group's receipt plan into dated purchase orders and a fabric commitment; the failure modes; and what to look for in software. It is the companion to demand vs merchandise vs supply planning, which draws the boundaries between the three disciplines, and to supply chain software vs merchandise planning software, which separates the systems that hold their decisions.

What supply planning is

Three properties separate a supply plan from the plans that feed it. First, it is denominated in commitments rather than intentions. A merchandise plan can be revised by editing a cell; a supply plan is a set of promises to third parties, each with a date after which changing it costs money. Second, it works in lead times rather than selling periods. The merchandise plan phases receipts by the month they should land; the supply plan reads each of those months backwards through transit, production, approvals and material to find the date each commitment has to be made, which is where the real calendar of the season lives. Third, it is constrained from below. The plans above it are constrained by money and by the customer; the supply plan is constrained by the mill's minimum, the factory's capacity in a given month, the dyehouse's lot size and the sailing schedule, none of which the plan chose.

Those properties make supply planning the discipline that routinely tells the others they are wrong. A minimum order quantity rounds a buy above what the plan asked for. A factory that is full in June moves a delivery into August. A fabric that takes nine weeks to knit and dye moves the point of no return into the previous quarter. Each of these is a fact about supply that changes the money, and the demand vs merchandise vs supply planning guide describes the return arrow as the handoff that is easiest to skip. This guide is about what happens on the supply side of that arrow, in enough detail to do the work.

Where supply planning sits: envelope, commitments, execution

Supply planning is the middle of three layers, and the confusion about it comes from the fact that it shares an object with each neighbour. With the layer above it shares the receipt plan; with the layer below it shares the purchase order.

LayerWhat it decidesDenominated inUsual ownerArtifact
The envelope: demand plan, merchandise financial plan, open-to-buyWhat to sell and own, by period and channel, and what it must returnDollars at retail and cost; units by item and weekMerchandise planner, demand planner, financeMerchandise financial plan, open-to-buy, receipt plan
The commitments: supply planHow, when and from where the planned receipts will be suppliedUnits by vendor and date; meters of fabric; capacity by monthSourcing, production planning or supply planningDated purchase order schedule, material commitments, capacity bookings, delivery schedule
Execution: purchase orders, production, logisticsMaking, moving and receiving what was committedOrders, milestones, shipments, cartonsVendor, factory, forwarder, distribution centerPurchase order record, time-and-action calendar, WIP report, shipping documents, receipt

The receipt plan is the object handed down. The merchandise financial plan sets sales, margin and inventory by period; the open-to-buy identity turns them into the receipts each period can take, and the receipt plan phases those receipts across the season by class, channel and delivery. That is where the envelope stops. The receipt plan says how many units should land in which period, in retail dollars or in units at a planned average cost; it does not say which vendor will make them, when the fabric has to be booked or whether the factory has room. Supply planning answers those questions, and the answers come back as a revised receipt plan, because a minimum or a lead time that disagrees with the original sends it back changed.

The purchase order is the object handed down again. Once the supply plan has fixed quantities, vendors and dates, the purchase order is the commitment to a vendor, and from that point the question changes from what should happen to what is happening. The T&A calendar back-schedules the milestones; the WIP report says where each order really stands; the shipment becomes a receipt, and the receipt relieves the on-order line in the open-to-buy. The supply chain software vs merchandise planning software guide maps the systems that hold each of those records; the point for the discipline is that supply planning ends where the order is placed and begins again when the order reports back, because a slipped ex-factory date is a new supply fact the plan has to absorb.

In a large vertically integrated business these three layers are three teams with a monthly S&OP cycle between them. In a mid-market brand they can be two or three people, and the supply plan lives in the sourcing manager's purchase order tracker next to the planner's open-to-buy. The layers are the same either way. What differs is whether the return arrow is a meeting or a conversation, and whether the two documents are reconciled by structure or by somebody remembering.

Demand planning vs merchandise planning vs supply planning

The demand vs merchandise vs supply planning guide argues this boundary in full, and this guide will not re-argue it. The table is the short form, and the one row that matters for everything below is the last one: each discipline's blind spot is another discipline's whole job.

DimensionDemand planningMerchandise planningSupply planning
Question it answersWhat will customers buy, where and when?What will we own, what will it cost, what must it return?Can it be made and landed on the dates we assumed, and what does that commit us to?
Native unitUnits by item and location, by weekDollars first, then options and unitsPurchase orders, meters and components, capacity, lead time, dates
HorizonThe selling period, rolled forwardThe season and the fiscal yearLead time out from each receipt, which puts its first commitment ahead of the other two
Primary artifactDemand plan by item and locationMerchandise financial plan and assortment planDated purchase order, material and capacity schedule by vendor
What it optimizesAvailability against the shape of demandReturn on the inventory the business ownsOn-time, in-full delivery against committed dates and minimums
What it is blind toWhether the plan is affordableWhether the demand assumption was trueWhether the thing being delivered was worth buying

The three meet at one number, the buy quantity, and supply planning is the discipline that turns that number into dates. Demand planning and merchandise planning can agree, to borrow that guide's illustrative figures, that a class needs 25,000 units across the season; neither of them knows that the vendor's minimum makes it 26,000, that the factory can take only part of it in June and the rest in July, or that the fabric for a September delivery has to be booked in March. Those are the facts the supply plan adds, and the sequence in which it adds them is the process described next.

Supply planning vs supply chain planning vs S&OP vs replenishment planning

A search for supply planning returns three neighbours that use the same words for different jobs. The distinctions are about scope and cadence, not technique.

DisciplineScopeHorizon and cadenceUnitWhere it meets supply planning
Supply planningOne business converting the quantities in its own plan into purchase orders, production bookings, material commitments and inventory targetsLead time out from each planned receipt; set pre-season, re-run as production reports inOrders, meters, capacity, datesIt is the discipline
Supply chain planningThe wider set: network design, inventory positioning across nodes, transport and distribution planning, detailed production scheduling, often across many products and many customers' demandStrategic to operational, multi-echelon, continuousNodes, lanes, flows, schedulesSupply planning is the slice a brand does for its own receipts; a supply chain planning system may run it, among other things
S&OP / integrated business planningThe monthly governance cycle that reconciles the demand plan, the supply plan and the financial plan into one agreed plan with a decision forumMonthly, on a rolling horizon that runs well beyond the current seasonVolume and value by family, with gaps to closeThe supply review is one input; the agreed plan is what the supply plan then commits against
Replenishment planningRule-based refill of continuing items against a target stock or rate of sale, per item and locationWeekly or daily, inside the item's lead timeUnits per item-location, triggers, order multiplesSupply planning sets the lead time, minimum and reserved receipts the rule draws on; replenishment spends them

Two of these deserve a sentence more. Integrated business planning is not a planning layer of its own; it is the cadence that forces the financial plan, the merchandise plan and the operations plan to face each other monthly, and the supply plan is the operations side of that conversation. The supply planner arrives with the capacity ceilings, the minimums and the dates; the meeting decides which of them the money will accept. Replenishment planning, covered in core and NOOS replenishment programs, is the part of supply that runs on a rule once the supply plan has set the parameters: the vendor lead time, the order multiple, the minimum and the receipts reserved for the program in the open-to-buy before the seasonal buy is planned. A replenishment rule left to run without a supply plan behind it will trigger against a lead time nobody has checked this year and an MOQ that has changed.

The vocabulary also moves by vertical. In grocery and hardline retail, where the assortment is largely continuous, "demand planning" often names the function that owns replenishment and much of what an apparel brand would call merchandise planning, and "supply planning" names the item-location-week supply side of the same engine. In a seasonal business the words shift: the merchandise planner owns the money and the options, and supply planning is the sourcing and production side of a buy that is placed once. The demand vs merchandise vs supply planning guide covers why the same job titles mean different jobs in the two worlds.

The supply planning process, step by step

The process below is the pre-season path for a seasonal business, with the in-season steps that follow from it. It is a loop: steps one to six run before the buy is placed, step seven reserves what can still move after it, and step eight runs weekly for the life of the season and feeds back into step one. The table is the summary; the subsections are the work.

StepThe questionMain inputsOutput
1. Net the receipt planOf what the plan wants to land, how much still has to be bought?Receipt plan, stock on hand, on order, cancel datesNet requirement by period, class and channel
2. Size to minimumsWhat can actually be ordered, given MOQs, case packs and fabric minimums?Net requirement, vendor minimums, pack configurations, consumption, fabric minimumsOrderable quantities, with every round-up priced
3. Offset lead timesBy when must each commitment be made for the receipt to land in its period?Lead time by stage, transit mode, DC processing timeCommitment dates back to the material
4. Book capacityCan the factories take the volume in the months the dates imply?Capacity by factory and month, holiday calendars, existing bookingsCapacity bookings, provisional then confirmed
5. Safety stock and datingHow much buffer does a replenished item carry, and what caps the cover?Rate of sale, lead-time variability, service target, shelf life and date-code policySafety stock parameters; cover ceilings
6. Split deliveriesWhich units land in which drop, and is each drop orderable?Flow plan, floor-set dates, minimums per orderDelivery schedule by drop
7. Reserve the chaseWhich levers stay open after the buy, and until when?Held material, held capacity, remaining open-to-buyChase reserve with last responsible dates
8. Review exceptionsWhat has moved against the plan this week, and what is the cheapest remaining lever?T&A plan dates, WIP actuals, receipt variancesRevised receipt plan written back to the open-to-buy

Step 1: Net the receipt plan against on-hand and on-order

The receipt plan says how many units should land in each period. Some of that requirement is already in the building or already on its way, and the first act of supply planning is to find out how much. Netting is the subtraction of what exists from what the plan wants: planned receipts for the period, less stock on hand above the opening position the plan assumed, less units already on order and due to land in or before the period, equals the net requirement that still has to be bought.

Three things make this harder than it reads. Where the receipt plan is in retail dollars by class and month and the on-order is in units by style-color and delivery date, the two have to be brought to a common grain before one can be subtracted from the other; the buy quantity formula shows the conversion from a dollar plan to a unit buy. The on-order has to be clean: cancelled lines left open, orders past their cancel date and orders re-dated into a different month all hold money against receipts that will not arrive when the plan says, and the on order entry covers why the open-to-buy period is the landing period rather than the order date. And carryover stock has to be read against what the plan assumed, not against zero: a plan that already expected 1,200 units of a core color to carry into the season is not helped by finding 1,200 units, only by finding more. The PO coverage formula is the running check once orders exist: how much of the plan's requirement the open orders on the book will actually deliver, by period.

Step 2: Size the buy to minimums, case packs and fabric minimums

The net requirement is what the plan wants. The orderable quantity is what the supply base will make, and the gap between them is where money moves without anybody deciding to move it. Three floors sit under a buy. The vendor's minimum order quantity, quoted per style, per style-color or per color across a season, and sometimes per size tier. The pack configuration, where goods ship in prepacks of a fixed size ratio or in cases of a fixed count, so that a quantity is only orderable in multiples; in toys and juvenile hard goods the case pack is the buying unit, as case packs and planned depth sets out. And the fabric or material minimum upstream of the garment, where a mill's minimum per color exceeds what one style would consume, and the planning unit becomes the fabric rather than the style, which planning against fabric minimums works through.

The discipline is not to avoid round-ups, since a minimum exists precisely because the vendor will not go below it, but to make every round-up visible and priced before the order is cut. A round-up has four honest responses: consolidate two deliveries or two channels into one order line so the minimum is cleared without extra units; move the extra units to another color or style on the same fabric; accept the extra units and take the money from elsewhere in the plan, deliberately; or cut the option, because an option that cannot clear its floor at a sensible depth is a range decision rather than an order-week emergency. What a supply plan must not do is round up silently and let the open-to-buy discover the difference when the receipts land. The worked example later in this guide shows the first and the third on one style group: a consolidation that clears a factory minimum without extra units, and a fabric round-up accepted and priced, with holding the dyed residual as the alternative the planner declines.

Step 3: Offset lead times from the in-store date back to the fabric commit

Every receipt in the plan has a landing period. Behind it is a chain of intervals, each a commitment somebody has to make by a date, and the supply plan finds those dates by subtraction. From the in-store date, subtract the distribution center's processing and allocation time to get the DC receipt date; subtract transit and clearance for the chosen freight mode to get the ex-factory date; subtract production time to get the production start date; subtract the approval loops to get the date the order and the approvals have to begin; subtract material lead time to get the date the fabric, the components or the leather have to be committed. The time-and-action calendar is the artifact that holds this chain for one program, and its arithmetic is the supply plan's arithmetic.

Two consequences follow. The real start of the season is the material commitment, not the purchase order, and a plan that reports commitment from the purchase order date understates how much of the season is fixed by the whole length of the fabric lead time; the fabric minimums guide calls this the two points of no return. And lead time is a set of intervals, each asserted by a different party, rather than one number per vendor. The lead time entry defines it end to end; the supply plan needs it by stage, by vendor and by freight mode, because a slip in any one stage moves the receipt, and because the only way to learn that a vendor's six-week production window is really eight is to hold planned and actual dates for each stage in the same structure season after season. Speed to market in apparel covers which of those intervals are worth compressing and which are not.

Step 4: Book capacity by factory and month

Lead-time offsetting produces production windows: this order needs to be on the line between these dates. Capacity booking asks whether the factory has room in those weeks, and reserves it. Factories plan their lines by month, and a reservation made early and confirmed on time is what secures the month the arithmetic needs. A booking works in two stages: a provisional reservation at or before material commitment, expressed as a volume range by month, and a confirmed booking when the purchase order is placed. The provisional booking is a forecast of the buy, and it is wrong in the direction the line review moves it, which is why the failure modes section below treats capacity booked before the line is locked as its own problem.

Capacity is not one number. It is capacity per factory per month, per product type where lines are specialized, and it is reduced by holiday shutdowns that remove whole weeks from the calendar and by the congestion on either side of them as every customer tries to ship before the closure or immediately after it. A supply plan that books by month and ignores the factory calendar will book a receipt into a week the factory is closed. The planner's instruments here are the vendor master with capacity and calendar by factory, the booking record with provisional and confirmed status, and the aggregate view that shows, for each month, bookings against capacity across every program. How sourcing agents keep production visible describes the order-book discipline an agent uses to hold that view across many brands and factories; a brand needs the same view for its own.

Step 5: Set safety stock and dating rules

Safety stock belongs to the part of the range that replenishes. For a continuing item with a rate of sale and a repeat order behind it, safety stock is the buffer above expected demand over the lead time that protects against demand variability and lead-time variability to a chosen service level, and the safety stock formula gives the arithmetic. The supply plan's job is to set the parameters the rule runs on, the lead time, its variability, the review period and the order multiple, and to revisit them when any of them changes, which they do every season. A seasonal one-shot buy has no safety stock, because there is no reorder for it to protect; its equivalent levers are planned depth, the chase reserve and a credible exit plan, and dressing a one-shot buy in a safety stock formula produces a bigger buy with no more protection.

Dating rules cap the other end. Where product carries a shelf life or a retailer's date-code acceptance policy, the cover a batch can justify is bounded by the weeks it will remain sellable (the period-after-opening mark on the pack is a consumer label, not a planning ceiling), and dating rules and weeks of supply computes that ceiling and shows why it caps safety stock rather than adding to it. The weeks of supply that a receipt creates at the moment it lands is the measure both halves of this step read: the floor the buffer sets and the ceiling the dating sets, with the planned cover between them.

Step 6: Split deliveries and drops

A season's receipt plan lands in several deliveries, and a supply plan has to decide which units arrive in which one. The merchandise plan sets the flow for the money's reasons: stock-to-sales targets by month, floor-set dates, the newness cadence the customer expects. The supply plan tests each drop for orderability and cost: does each delivery's order clear the minimums on its own, or do two deliveries have to be consolidated into one order line and warehoused; does a split delivery change the freight mode or the container utilization; does the factory have capacity in both production windows or only one. A delivery split that the plan wants and the minimums will not allow sends the supply plan back different from the receipt plan it was given, and the worked example below shows the consolidation decision in units and weeks of early stock.

Drops also set the exit. A drop that lands late compresses its own full-price window, and the supply plan's dates decide how much window each drop gets. Reading the delivery schedule against the planned exit date for each drop, rather than only against the in-store date, is what tells the planner whether a slipped delivery is a late drop or a markdown.

Step 7: Plan the chase and in-season re-supply

Everything above is committed before the season starts. The chase is what remains open afterwards, and a supply plan that does not reserve it has no in-season levers at all. The in-season chase defines a chase as any action that changes what the business will have to sell after the buy is committed, and sets out which levers stay open and the last week each one works. The supply plan's contribution is to create the levers before the season and date them: greige or undyed fabric held at the mill against a dye order to be placed later; a provisional capacity reservation in a later month that can be released if unused; a component or trim stock that shortens the next order's lead time; and open-to-buy deliberately left unspent to fund any of them.

Each lever has a last responsible date, found by the same back-scheduling as step three but run from the last week a chased unit could still sell at full price. A chase decision taken after that date is a next-season decision. The weekly in-season read that triggers a chase belongs to demand sensing and the reforecast; the supply plan's job is to have made sure that, when the read says chase, there is material, capacity and money to chase with, and a date by which the decision has to be made.

Step 8: Review exceptions

Once orders are placed, the supply plan is reviewed weekly against what production reports. The T&A calendar holds the planned date for each milestone; the WIP report holds the actual or projected date; the review reads the gaps and translates each one into a receipt fact. A fit sample five days late is a production fact. The supply plan turns it into a planning fact: this program's receipt has moved from one week into the next, which puts a floor set at risk, which means the approval loop compresses, the transit mode upgrades or the receipt plan re-phases. The weekly question is which delivery weeks now carry risk they did not carry last week, and what the cheapest remaining lever is for each, and the answer is written back to the receipt plan and therefore to the open-to-buy, because a receipt that has moved months has moved money.

Exceptions also run the other way. Receipt variance, where what arrived differs from what was ordered in quantity, size breakdown or date, is a supply fact the plan has to absorb, and a size run that arrives short in two sizes is a broken run at full price however healthy the unit total reads. The review closes the loop into step one: the on-order changes, the netting changes, and the next period's net requirement changes with it.

Inputs and outputs

A supply plan is only as good as its inputs, and the list is longer than the planning layer above it needs, which is one reason the two can sit in different systems.

Inputs

  • The receipt plan by period, class, channel and delivery, from the merchandise financial plan and the open-to-buy
  • Stock on hand by location, with the opening position the plan assumed, and aging where it matters
  • Open orders with quantities by size, delivery dates, cancel dates and current status
  • The bill of materials and consumption per unit, including cutting loss, for every fabric, leather, component and trim
  • The vendor master: lead time by stage (material, approvals, production, transit) and by freight mode, minimum order quantities as quoted (per style, per color, per size tier, per season), case pack and prepack configurations, capacity by factory and month, and the factory calendar
  • Size curves by class and cluster, and the pack configurations the allocation will need
  • Freight modes, transit times and container or cube constraints where goods ship by sea
  • Dating rules: shelf life and retailer date-code policy
  • The T&A template for each product type: the milestone chain and the owner of each step
  • The planned exit date and full-price window for each drop

Outputs

  • A dated purchase order schedule by vendor, style-color and size, with the order date and the delivery date for each line
  • Material commitments to mills, tanneries and component suppliers, with quantities and commit dates
  • Capacity bookings by factory and month, provisional and confirmed
  • A delivery and drop schedule reconciled to the floor-set calendar
  • Inventory targets and safety stock parameters for replenished items; cover ceilings where dating applies
  • A chase reserve: held material, held capacity and unspent open-to-buy, each with a last responsible date
  • The weekly exception list, with each exception translated into a receipt-week effect
  • A revised receipt plan, written back to the open-to-buy with every round-up and re-dated delivery in it

The last item is the one that makes the rest worth doing. A supply plan that does not write back to the receipt plan is a schedule, not a plan, because the money is still sitting in the original months while the goods are landing in different ones.

How apparel supply planning differs from the CPG model

The textbook supply planning model, time-series forecasting, reorder points and safety stock, grew up in consumer packaged goods, and it fits that world well: an item that sells for years, a long sales history at item and location, a weekly forecast at that grain, a repeating replenishment order, and a safety stock that buffers forecast error and lead-time variability against a service level. Seasonal apparel, and the seasonal part of each vertical below, breaks each assumption in turn, and the breaks are structural rather than matters of degree.

DimensionTextbook CPG supply planningSeasonal apparel supply planning
Item lifeYears; the same SKU on the shelfA season; the style-color exits whether it worked or not
HistoryLong, at item and locationNone for a new style; analogues and class history only
Forecast grainItem-location-week, statisticallyClass and style-color for the buy; size by curve
Order patternRepeating replenishment orders inside the lead timeOne buy, sometimes two drops, placed before any sales exist
BufferSafety stock against forecast error and lead-time variabilityPlanned depth, a chase reserve and an exit plan; no reorder to protect
Binding constraintLine capacity, shelf life, distribution networkFabric minimums and lead time, factory calendar, size-run integrity
Point of no returnThe replenishment order, every cycleThe fabric commitment, months before the garment order

The one-shot buy is the first break. Because the buy is placed once, before the item has sold, the supply plan cannot correct itself through the next replenishment cycle; its errors land on the floor as depth. That moves the planning effort upstream, into getting the first commitment right and reserving the chase, rather than into tuning a reorder rule.

Newness without history is the second. A new style-color has no sales of its own to forecast from, so the quantity comes from the plan, the class history and an analogue style, and the supply plan is sizing commitments against a judgement rather than a time series. The parts of the range that do repeat, core, carryover, size curves by class, behave like the textbook assumes and should be planned that way; demand forecasting for fashion covers which parts those are.

Size-level sparsity is the third. The factory needs a quantity by size, and demand at the size level for one style in one door is too thin to forecast item by item, so the run is bought as a curve applied to the style-color total, and the supply plan's job is to keep that curve intact through the order, the factory's cutting plan and the shipment. A run that is flattened to a default ratio or cut short in two sizes is broken at full price while the unit count reads healthy. Size and pack optimization covers the method; size curve covers the term.

Long fabric lead times are the fourth. The textbook model has one commitment, the order. Apparel has two: the fabric, committed to the mill weeks or months before the garment, and the garment, committed to the factory. Exposure begins at the first, and a plan that measures commitment from the purchase order understates how much of the season is already fixed by the whole fabric lead time. Planning against fabric minimums describes committing the fabric and deferring the cut as the way to split one irreversible decision into two.

The factory calendar is the fifth. In a replenishment world the supplier's calendar is a constraint on the reorder cycle; in a seasonal world it decides which month a receipt can land in. Holiday shutdowns remove weeks from the calendar, the weeks on either side of them are congested as every customer ships before or after, and a factory that is full in the month the lead-time arithmetic points to pushes the receipt into the next one. The supply plan has to hold the factory calendar as an input, not discover it as an exception.

Supply planning by vertical

The eight steps are the same everywhere. What changes is the grain the commitment is made at, the constraint that binds and the calendar the receipts run on, and a supply plan that is built in another vertical's vocabulary will miss the constraint that matters. Each section below is short on purpose; each vertical has its own planning guide.

Apparel

Apparel commits at style-color, cuts at style-color-size and sells through DTC, wholesale and stores on different clocks. The binding constraints are the fabric minimum per color and the fabric lead time, which put the first point of no return at the mill rather than the factory, and the size curve, which has to survive the order, the cutting plan and the shipment intact. Wholesale bookings confirm part of the quantity before the purchase order, so the supply plan nets the booked units from the buy and treats the at-once stock as a decision. Deliveries are planned as drops against floor-set dates, and the chase is decided before the season by holding greige against a later dye order. The worked example below is an apparel example for these reasons. See planning against fabric minimums, the time-and-action calendar and the apparel brands page.

Footwear

Footwear commits pairs by size within width, and the supply plan's first job is size-run completeness: a model-color is bought as a run, and a factory that ships the run short in two sizes has broken it at full price however complete the pair count reads, so the order, the factory confirmation and the shipment all have to carry the run at the size-within-width grain. The second constraint is tooling. Lasts and outsole molds are committed before the first purchase order, mold capacity limits how many pairs of a given sole can be produced in a month, and the mold is amortized across a model year, which is why footwear plans on the model year rather than the season alone and why carryover models continue across it. Dealer prebooks size the production buy months ahead, and at-once stock is a deliberate hold rather than a residual. Factory minimums apply per model-color, per material or per size tier. See assortment planning for footwear brands, planning a model-year changeover and the footwear brands page.

Accessories and bags

Accessories commit at colorway and hardware finish rather than size run, so the supply plan's unit is the colorway and the material behind it. The binding minimums sit at the tannery, where a leather is dyed to a minimum per color, and at the component supplier, where hardware finishes are plated in lots, and both are discovered in the order confirmation unless the plan carries them. An evergreen core replenishes on a rule beneath a seasonal layer bought once, so the supply plan has to run two patterns on one category: lead time, minimum and reserved receipts for the core, and a one-shot buy with a chase reserve for the hero color. Carryover is the norm for core shapes, and the model year matters less than the material commitment. See planning accessories lines and the accessories brands page.

Home and furniture

Home and furniture commit SKUs with long lead times, shipped by ocean, and the container is the real order unit: the cube sets the quantity, so a slow finish can be bought to fill a container rather than to meet its own demand, and the supply plan decides which SKUs and finishes fill each container in each month against the receipt plan. Options and finishes multiply the SKU count without multiplying the floor sample, so the floor set is a decision about which options are shown and the supply plan has to hold stocked, made-to-order and dealer-prebook quantities apart. Dealer and trade orders written at market weeks are netted against the buy, and landed cost moves with freight on bulky goods, so the cost the margin was planned on has to be re-read when the container is booked. A container already on the water cannot be changed, so the chase is the next container or a cancellation. See merchandise planning for home and furniture brands, OTB planning for home goods and the home and furniture brands page.

Outdoor

Outdoor plans on the model year, and the wholesale channel prebooks the line at a sales meeting months before delivery, so the production buy is sized from the dealer prebook and the at-once stock held for fill-in is a decision the supply plan makes explicitly. Mixed soft and hard goods run on different lead times: technical fabrics and membranes carry mill lead times and minimums that put the first commitment early, while hardgoods components carry their own supplier calendars, and one model year's line commits both. Carryover models continue across model years, so the supply plan holds the changeover date as the point after which the outgoing model is not re-ordered. Counter-seasonal categories mean the factory calendar is read twice a year. See merchandise planning for outdoor brands, planning a model-year changeover and the outdoor brands page.

Sporting goods

Sporting goods run several seasons at once, one per sport, each with its own start, peak and end, so the supply plan is phased by sport rather than by one retail calendar and a late season start shifts a sport's receipts without shifting the others. Prebook from team and institutional accounts is committed demand with dates, netted from the buy; hardgoods carry sizes of their own, racquets, bats and boots, beside softgoods with apparel sizing; and consumables replenish on a min/max rule with the supplier lead time setting the distribution center's reorder point and the DC-to-door lead time setting each door's. The chase has to land inside a sport's window that does not move. See merchandise planning for sporting goods brands and the sporting goods brands page.

Beauty and wellness

Beauty commits shade ranges as a set, because a missing shade reads to the customer as the product being unavailable, so the supply plan has to keep every shade in the ladder in supply and the slow ends of the ladder are where it breaks. The lead times are in components and filling: pumps, jars, caps and cartons from component suppliers, bulk formula from the filler, and filling runs in batch sizes with their own minimums, and the longest of these sets the commitment date. Shelf life and retailer date-code policy cap the cover a batch can justify, so safety stock on a slow shade is bounded from above by dating rather than set from below by variability; the period-after-opening mark on the pack is a consumer label and is not the figure the ceiling is computed from. Retailer gondola resets and promotional calendars set when a range can change on shelf, and where the retailer controls replenishment the brand's levers are its fill to the retailer's DC and the next batch. See merchandise planning for health and beauty brands, dating rules and weeks of supply and the health and beauty brands page.

Toys and games

Toys concentrate demand in the gifting peak, and the factory lever closes before it: production for the peak is committed months ahead, so the supply plan commits the peak's volume as a single decision with a domestic holdback as the only in-season lever. Licensed product runs in windows with a start and an end the brand does not control, so the supply plan dates the last production run from the window end minus the lead time and plans the run-down into it. The case pack is the buying unit and its configuration is fixed long before the assortment conversation, and safety testing adds a lead time between production and shipment that a reworked item has to pass again. Retailer commitments are firm sell-in with dates: the buy is sized against them rather than against a forecast, and they are not on order. See merchandise planning for toy and game brands, planning a licensed product window, case packs and planned depth and the toys and games brands page.

Baby and juvenile

Baby and juvenile hard goods see part of their demand months ahead through the registry, which gives the supply plan a forward read no other vertical has, and part of it as steady replenishment of consumables as the child grows. Safety testing and certification are a lead time of their own: a certified configuration cannot be substituted, a configuration change triggers re-testing before the item can ship, and recalls are lot-level events, so the supply plan carries the production lot through the order and the receipt. Case packs set the order multiple, and long hard-goods lead times put the commitment early against a model-year changeover that moves registrants to the new model once it is listed. See merchandise planning for baby and juvenile brands, planning with registry demand and the baby and juvenile brands page.

Jewelry and watches

Jewelry and watches commit on a precious-metal cost base that moves between the commitment and the sale, so the supply plan fixes the metal, whether bought, hedged or priced at order, at the point of commitment and re-reads the planned margin against the metal price when the piece is refilled; planning margin on a moving cost base covers the mechanics. Depth is a piece or two per style per door, pieces are refilled from the bench or made to order rather than bought in runs, and memo and consignment make rotation between doors the main in-season lever. The model year sets the changeover, with carryover as the default rather than the exception, and watch movements and stones carry component lead times of their own. See merchandise planning for jewelry and watch brands and the jewelry and watch brands page.

The pattern across the ten is consistent. The vertical sets the grain of the commitment and the constraint that binds it, and a supply plan that carries the wrong grain, pairs without widths, shades without the ladder, SKUs without the container, will be right on paper and wrong at the dock.

An illustrative worked example: from receipt plan to dated orders and a fabric commitment

The numbers below are illustrative round figures chosen to show the arithmetic, not benchmarks, and every date and quantity is a planning assumption for one made-up style group. The group is a women's knit top: one style in three colorways, black, ivory and navy, cut from one cotton-modal jersey knitted from a single greige and dyed to the three colors, sold through DTC and wholesale. The merchandise plan's receipt plan for the group, in units landing at the distribution center, is two deliveries for an autumn season.

DeliveryIn-store datePlanned receipt (units)
Delivery 1Monday 30 August 20276,000
Delivery 2Monday 25 October 20274,000
Season10,000

The supply facts, all illustrative: the factory's minimum is 1,200 units per color per purchase order; the mill's greige minimum is 10,000 meters per knit, and its dyed minimum is 3,000 meters per color across the season, dyed in up to two lots; consumption is 1.25 meters per unit including cutting loss; the landed cost is $11.00 per unit in every color; the factory holds 5,000 units a month of capacity for the brand. Lead times by stage: greige knitting 5 weeks; dyeing and finishing 4 weeks; fabric inspection and relaxation at the factory 1 week; production 6 weeks; ocean transit and clearance 5 weeks; DC processing and allocation to stores 2 weeks. That chain is 23 weeks from greige commitment to in-store.

Step 1: Netting

The plan assumed 1,200 units of carryover on hand at the start of the season, held across the three colors in the group's mix; the count shows 1,800, so 600 units of the planned receipt are already in the building. A replenishment order for 400 units in the same color mix, placed last season, is on order and lands in July, before Delivery 1. Nothing is on hand or on order against Delivery 2.

Delivery 1Delivery 2Season
Planned receipt6,0004,00010,000
Less on hand above the plan's opening position6000600
Less on order landing in or before the period4000400
Net requirement5,0004,0009,000

Step 2: Sizing to minimums

The assortment plan's color mix for the group is 50% black, 30% ivory and 20% navy, so the gross Delivery 1 plan of 6,000 units is black 3,000, ivory 1,800 and navy 1,200, and the gross Delivery 2 plan of 4,000 is black 2,000, ivory 1,200 and navy 800. Netting runs by color, not on the total: the on-hand surplus and the on-order carry the same mix, so Delivery 1 loses 500 black (300 on hand, 200 on order), 300 ivory (180 and 120) and 200 navy (120 and 80), which gives the split below. Two lines fall under the factory minimum of 1,200 units: navy in Delivery 1 at 1,000 and navy in Delivery 2 at 800.

ColorDelivery 1 netDelivery 2 netSeason net
Black2,5002,0004,500
Ivory1,5001,2002,700
Navy1,0008001,800
Total5,0004,0009,000

The first response is consolidation rather than a round-up: navy's two deliveries become one order line of 1,800 units in Delivery 1, which clears the factory minimum without buying a unit the plan did not ask for. The cost is time, not units: the 800 navy units planned for October land in August and sit for eight weeks.

The fabric then produces the round-up consolidation cannot avoid. Black needs 4,500 × 1.25 = 5,625 meters and ivory 2,700 × 1.25 = 3,375 meters, both above the 3,000-meter dyed minimum. Navy needs 1,800 × 1.25 = 2,250 meters, below it. The planner has two honest choices: buy 3,000 meters and hold 750 meters of dyed navy as residual fabric with no garment behind it, or raise the navy buy to 3,000 ÷ 1.25 = 2,400 units, 600 units above the plan. In this example the planner chooses the units, because a navy carryover into the next season is more sellable than dyed yardage; the point is that the choice is made and priced, not discovered. The buy is now 9,600 units against a net plan of 9,000.

ColorSeason plan (units)Season buy (units)Fabric (meters)Over plan (units)
Black4,5004,5005,6250
Ivory2,7002,7003,3750
Navy1,8002,4003,000600
Total9,0009,60012,000600

The round-up carries money. At $11.00 landed, the plan's cost commitment was 9,000 × $11.00 = $99,000 and the buy's is 9,600 × $11.00 = $105,600, so $6,600 goes back to the open-to-buy in August, the month the navy lands, before the order is cut. Either the planner takes it from another class in the same month or the plan accepts the higher inventory and the exit risk on navy deliberately.

The greige minimum then decides how the fabric is committed. Delivery 1's fabric is 6,400 × 1.25 = 8,000 meters and Delivery 2's is 3,200 × 1.25 = 4,000 meters; each on its own is under the 10,000-meter greige minimum, so the greige is knitted as one 12,000-meter lot on Delivery 1's dates and dyed in two lots, one for each delivery. The second half of the greige waits eight weeks between coming off the knitting machine and its dye order. That is the two-points-of-no-return structure planning against fabric minimums describes: one fabric commitment, two garment orders.

Step 3: Lead-time offsets

Back-scheduling from each in-store date through the six intervals gives every commitment its date. The garment purchase order is cut on the day the dye order is placed, because the color quantities on the order drive the dye lots.

MilestoneOffsetDelivery 1Delivery 2
In storeMon 30 Aug 2027Mon 25 Oct 2027
DC receipt2 weeks before in-storeMon 16 Aug 2027Mon 11 Oct 2027
Ex-factory5 weeks before DC receiptMon 12 Jul 2027Mon 6 Sep 2027
Production start6 weeks before ex-factoryMon 31 May 2027Mon 26 Jul 2027
Dyed fabric at the factory1 week before production startMon 24 May 2027Mon 19 Jul 2027
Dye order placed and garment PO cut4 weeks before fabric at factoryMon 26 Apr 2027Mon 21 Jun 2027
Greige commitment5 weeks before dye orderMon 22 Mar 2027Consolidated into 22 Mar 2027

The season's real start is 22 March 2027, 23 weeks before the first unit is on the floor and 31 weeks before the second delivery's in-store date. A plan that reports commitment from the purchase order date of 26 April understates the fixed portion of the season by five weeks, and a plan that reports it from the second order's date of 21 June understates it by thirteen.

Steps 4 to 7: Capacity, buffer, drops and the chase

Delivery 1's 6,400 units run on the line from 31 May to 12 July, across June and the first half of July, and Delivery 2's 3,200 units from 26 July to 6 September, across August; against 5,000 units a month held for the brand, both windows fit. The capacity is reserved provisionally on 22 March, when the greige is committed, and confirmed with each purchase order on 26 April and 21 June. The seasonal one-shot carries no safety stock; the carryover is a replenished core with its own trigger and is outside this example. The drops are already split, with navy consolidated into the first. For the chase, the planner asks the mill to hold the option on a further greige lot: with greige held, the chain from dye order to in-store is 4 + 1 + 6 + 5 + 2 = 18 weeks, so a chase that has to be on the floor by Monday 1 November 2027, ahead of the holiday peak, has a last responsible dye date of Monday 28 June 2027, one week after Delivery 2's dye order. That date is nine weeks before Delivery 1 reaches the floor on 30 August, so on this chain a chase for the holiday peak cannot be triggered by this group's own selling at all: it has to be decided on the wholesale order book and the pre-season read, with the held greige as the option that makes the decision cheap. A dye order placed after the first two weeks of selling, on Monday 13 September 2027, lands in store on Monday 17 January 2028 on the same 18-week chain, which is a spring receipt rather than a chase.

What goes back to the plan

The receipt plan asked for 6,000 units in August and 4,000 in October. The supply plan delivers 6,400 purchased units in August plus 400 on order and 600 already on hand, 7,400 units against the August plan of 6,000: 800 of the 1,400 difference is navy pulled forward from October and 600 is the fabric minimum. October receives 3,200 against a plan of 4,000, the 800 navy having moved. The season's supply is 10,600 units against a plan of 10,000, the 600-unit difference being the minimum, at a cost commitment of $6,600 above plan. Every one of those numbers is written back to the receipt plan and the open-to-buy on 22 March, when the fabric is committed, which is the only date on which the plan could still have chosen differently.

Common failure modes

The failures below recur because each one is invisible from where it is made. The plan looks balanced, the order looks complete, and the problem lands at the dock.

Planning supply from last year's receipts rather than this year's plan. The quickest way to build a supply plan is to re-key last season's purchase orders with new dates, and the reason it fails is that the receipt plan has changed and the orders have not: a class that the merchandise plan cut has its old quantity back, a delivery that moved a month is still in the old one, and a color mix the assortment changed is unchanged on the order. The supply plan has to be netted from this season's receipt plan in step one, every season, and the test is simple: every order line should trace to a line in the current receipt plan, and every line in the receipt plan should trace to an order, a stock position or a deliberate gap.

MOQ round-ups that silently inflate the buy. The order is sized to the minimum because the factory will not take less, the extra units are not written back to the receipt plan, and the open-to-buy discovers them when they land. The worked example shows the honest version: the round-up is found at sizing, the alternatives are compared, the chosen one is priced in units and money and the money moves before the order is cut. The failure version skips the write-back, and the symptom is a month that lands over plan with no decision anyone can point to.

Capacity booked before the line is locked. A provisional booking is a forecast of the buy, and the line review changes the buy. Booking too early at full volume commits the brand to capacity it then has to fill or pay for; booking too late finds the factory full in the month the arithmetic needs. The workable pattern is a provisional range at material commitment, a confirmation at the purchase order and a factory relationship in which the difference between the two is expected rather than argued about. The failure is a confirmed booking against a line that is still moving, which is why the booking record should carry its status and the line plan's version it was made against.

The plan and the T&A calendar disagreeing about a receipt month. The receipt plan holds a delivery in October. The T&A calendar, read against the WIP report, shows the ex-factory date has slipped three weeks. In the worked example that moves Delivery 2's ex-factory date from 6 September to 27 September, the DC receipt from 11 October to 1 November and the in-store date from 25 October to 15 November: the receipt has left October's open-to-buy period and entered November's, with the money still in October. The failure is not the slip, which is a production fact; it is the two documents holding two months for the same receipt. The fix is the weekly translation in step eight, from a milestone slip to a receipt-month effect, written back to the plan.

Netting against a dirty on-order. Cancelled lines left open, orders past their cancel date, re-dated deliveries still bucketed by the original month: each one holds money against a receipt that will not arrive as the plan says, and the net requirement comes out too low. The on order entry covers the discipline; the supply plan depends on it.

Safety stock on a one-shot buy. A seasonal style with no reorder is given a safety stock because the template has a column for it, and the result is a deeper buy with no more protection, because there is no replenishment order for the buffer to protect. The lever for a one-shot is planned depth and a dated chase reserve, and the buffer belongs only to the part of the range that repeats.

What to look for in supply planning software

Software for this work ranges from a purchase order tracker beside a spreadsheet receipt plan to a dedicated supply chain planning engine, and this section does not name products. It describes the mechanisms a buyer should test, because the mechanisms are what decide whether the eight steps above can be run as a weekly discipline or only as a season-start project.

  • The receipt plan and the on-order at the same grain and period. Netting is only possible if planned receipts and open orders can be read against each other by period, class, channel and delivery, with the landing period, not the order date, as the bucket. Ask to see a receipt moved a month and watch what the open-to-buy does.
  • Lead time by stage, by vendor and by freight mode, with back-scheduling that produces commitment dates from an in-store date and recomputes them when one interval changes. A single lead-time number per vendor is a tracker, not a plan.
  • Minimums, case packs and material minimums visible at sizing, with every round-up shown against the plan quantity and its cost, and the consolidation alternatives available in the same view.
  • Capacity by factory and month against provisional and confirmed bookings, with the factory calendar in it, so that bookings against capacity can be read in aggregate across programs.
  • The write-back. A re-dated delivery, a round-up or a cancelled line should reach the receipt plan and the open-to-buy as a matter of structure rather than as a re-key. If the planning layer and the supply layer are two systems, ask which direction the integration runs and how often.
  • Planned and actual dates for every milestone, with the WIP feed from factories or agents landing against the T&A plan, and exceptions surfaced by receipt-week effect rather than by order count.
  • A chase reserve with last responsible dates, held material and capacity and the open-to-buy left to fund them, so that the in-season read has something to act on.
  • A named system of record for the purchase order. The buy plan is a planning artifact; the order is a commitment, and one system has to win when two disagree about quantity, cost or date.

Questions to ask planning software vendors gives the question families for testing the data model and the implementation model behind any of these, and supply chain software vs merchandise planning software maps which class of system is the record for each decision from the buy plan to the receipt. The honest test of any product, including ours, is which of the eight steps it becomes the system of record for and which it only reads.

Where RetailNorthstar fits

RetailNorthstar is a merchandise planning platform, and supply planning as this guide defines it is not what it is for. The two meet at the buy. The buy planning page describes buy planning as converting assortment decisions into structured purchase commitments, quantities by style, color, size, vendor and channel, reconciled against the OTB budget in real time, with buy quantities assigned to vendors with cost, lead time and minimum order quantities tracked, and unit cost and landed cost captured per style-vendor combination so that the buy plan rolls up to total cost commitment. The OTB planning page describes receipt plan management as tracking confirmed PO commitments against planned receipts by period, vendor and category. The buying and vendor execution page describes purchase orders generated from the finalized buy plan, WIP tracking of production milestones from fabric sourcing through final shipment, style by style, suppliers updating production status directly, and receipt variance management that tracks what was ordered versus what was received. Those are the netted buy, the vendor and lead-time attributes, the commitments against the receipt plan and the milestone actuals that the steps above produce and consume, held on the same data model as the plan they came from.

What it is not is a supply chain planning or factory capacity system. As the demand vs merchandise vs supply planning guide states, a merchandise planning platform stops where the question becomes factory capacity, raw material commitment, production scheduling, container consolidation or landed cost by freight mode; those are supply planning problems, they need vendor and materials data a planning platform does not hold, and a system that pretends otherwise produces dates nobody in production believes. Where a brand runs a dedicated supply engine, the sensible arrangement is integration rather than replacement: the merchandise plan supplies the financial envelope and the assortment intent, and the engine returns dates and feasible quantities as constraints priced into the plan. The evaluation points above apply to RetailNorthstar as much as to any product a brand compares it with.

Related resources

See how RetailNorthstar reconciles the buy plan against open-to-buy in real time and tracks PO commitments against planned receipts by period, vendor and category.

Start Free Trial →

No credit card. No commitments.

Common questions

What is supply planning?

Supply planning is the process of deciding how, when and from where the quantities in a demand or merchandise plan will be supplied. It converts planned receipts into purchase orders, production bookings, raw-material commitments and inventory targets against lead times, factory capacity and minimum order quantities, so that stock arrives where and when the plan needs it. The demand plan and the merchandise plan say what the business intends to sell and own; the supply plan says what will be ordered from whom, when each commitment has to be made and when the goods will land.

What are the steps in the supply planning process?

Eight steps, run before the season and re-run as it trades. Net the receipt plan against stock on hand and on order to find what still has to be bought. Size the buy to minimum order quantities, case packs and fabric minimums. Offset lead times backwards from the in-store date to find the date of every commitment, back to the fabric. Book factory capacity by month. Set safety stock on replenished items and apply dating rules where shelf life applies. Split the buy into deliveries and drops. Reserve the chase and in-season re-supply with a last responsible date for each lever. Review exceptions weekly against the plan as production reports in.

How is supply planning different from demand planning and merchandise planning?

Demand planning decides what customers will buy, in units by item, location and week. Merchandise planning decides what the business will own and what it must return, in money first and options and units second. Supply planning decides whether and how those units can be made and landed on the dates the plan assumes, in purchase orders, capacity, material commitments and lead times. The three meet at the buy quantity: demand and merchandise planning agree how many to buy, and supply planning turns that number into dated commitments, sending it back changed when a minimum, a capacity ceiling or a lead time will not allow it.

What is the difference between supply planning and supply chain planning?

Supply planning is one business converting the quantities in its own plan into purchase orders, production bookings, material commitments and inventory targets. Supply chain planning is the wider discipline that also covers network design, inventory positioning across warehouses and nodes, transport planning and detailed production scheduling, often across many products and many businesses' demand. Supply planning is the part of supply chain planning that a brand or retailer does for its own receipts; S&OP or integrated business planning is the monthly cycle that reconciles the supply plan with the demand plan and the financial plan; replenishment planning is the rule-based refill of continuing items that draws on the lead times, minimums and reserved receipts the supply plan set.

What are the inputs and outputs of a supply plan?

The inputs are the receipt plan by period and channel, stock on hand by location, open orders with their cancel dates, the bill of materials and consumption per unit, vendor lead times by stage, minimum order quantities, case packs, capacity by month and holiday calendars, size curves and pack configurations, freight modes and transit times, and dating or shelf-life rules. The outputs are a dated purchase order schedule by vendor and style-color-size, material commitments to mills and component suppliers, capacity bookings by factory and month, a delivery and drop schedule, inventory targets and safety stock parameters, a chase reserve with last responsible dates, an exception list, and a revised receipt plan written back to the open-to-buy.

How does supply planning differ in apparel from consumer packaged goods?

The textbook supply planning model assumes a continuing item with a long sales history, forecast weekly by item and location, replenished on repeating orders, and buffered with safety stock against forecast error and lead-time variability. Seasonal apparel breaks every assumption. The buy is placed once, before any sales exist, so there is no reorder for safety stock to protect; new styles have no history of their own, so the quantity comes from a plan and an analogue rather than a forecast; size-level demand is sparse, so the run is bought as a curve; the fabric is committed weeks before the garment order, so there are two points of no return rather than one; and the factory calendar, with holiday shutdowns and peak-season congestion, decides when a receipt can land as much as the plan does.

Does a merchandise planning platform do supply planning?

Only up to the buy. A merchandise planning platform owns the money, the assortment, the buy plan and the allocation of what was bought, and it can hold the vendor, cost, lead time and minimum order quantity attributes the buy carries and track purchase order commitments against the receipt plan. It stops where the question becomes factory capacity, raw material commitment, production scheduling, container consolidation or landed cost by freight mode. Those are supply planning problems that need vendor and materials data a planning platform does not hold. Where a brand runs a dedicated supply engine, the sensible arrangement is integration: the merchandise plan supplies the envelope, and the engine returns dates and feasible quantities as constraints.

RetailNorthstar Editorial Team
RetailNorthstar ·

Share this guide with your team

Copy a link or a pre-written message for Slack, Teams, or email.

// Know where your operation stands

Apply this to your planning operation.

The free Apparel Planning Maturity Assessment benchmarks your operation and tells you exactly which gaps to fix first.

Take the assessment →

Apply these insights with RetailNorthstar.

See how modern apparel brands use RetailNorthstar to put this planning framework into practice.

No credit card. No commitments.

Connected merchandise planning — live in weeks, not quarters.