Most planning discussions stop one step too early. A brand invests in better demand insight, builds a forecast it trusts, agrees on targets, and produces a clear picture of where the business is heading. Then someone has to turn all of that into an actual purchase order, with quantities, sizes, colors, suppliers, and delivery dates.
That final step is where a large share of planning time is spent, and it is rarely the part anyone plans for. The analysis is intellectually satisfying. The order is administrative. But the order is the only part that changes anything, and the quality of the translation from forecast to order determines whether the analysis was worth doing at all.
Fashion Planner treats this step as a first-class part of the planning process rather than an afterthought. Order suggestions and size breakdown are where the numbers stop being a report and start being a decision the business can execute.
The gap between knowing and ordering
A planner can know with complete confidence that a product needs replenishment and still be a long way from placing the order. Knowing is the easy half. The remaining work is mechanical and surprisingly heavy.
The quantity has to be calculated from expected demand, current stock, incoming supply, and the target coverage. It has to be checked against the supplier's minimum order quantity and rounded to a valid pack or carton. It has to be split into sizes using a curve that reflects how the product actually sells. If the product exists in several colors, the exercise repeats for each. If stock sits in multiple warehouses or channels, the allocation has to be considered too.
For one product this is fifteen minutes of careful work. For a category review covering two hundred products it is several days, and by the time the last product is reached, the numbers used at the start are no longer current. The process is not difficult so much as it is relentless, and relentless work performed under time pressure is where errors enter.
This is the point where many brands quietly compromise. The order gets placed using last season's quantities plus a percentage, or using a size curve someone set up years ago, because there is not enough time to do it properly for everything. The compromise is invisible in the moment and expensive over the year.
What an order suggestion actually is
An order suggestion is a proposal, not an instruction. It is the system doing the arithmetic so the planner can spend their attention on judgment.
The calculation itself is not mysterious. The system looks at expected demand over the relevant horizon, subtracts what is already covered by stock on hand and orders in transit, applies the target coverage and any safety stock, respects the supplier's constraints, and breaks the result into sizes. Each of those steps is something a planner would do anyway. The difference is that the system does it identically for every product, every time, in seconds.
What makes a suggestion trustworthy is transparency. A number with no explanation invites either blind acceptance or blanket rejection, and both are bad outcomes. A planner should be able to see why the suggestion is what it is: this is the demand assumption, this is the current coverage, this is the lead time, this is the minimum order quantity that rounded the figure upward.
When the reasoning is visible, the planner can disagree with a specific input rather than with the result as a whole. That is a much more productive conversation, and it is how a planning system earns confidence over time instead of demanding it up front.
The planner is still the decision maker
There is a persistent worry that automated suggestions replace planning judgment. In practice the opposite happens, because judgment is the scarce resource and arithmetic is not.
A system does not know that a major customer is opening fifteen doors next quarter. It does not know that a campaign is planned for March, that a supplier is unreliable in the summer, that a product is being phased out at the end of the year, or that last month's spike was one unusual bulk order rather than a change in demand. All of that is commercial knowledge, and it belongs in the order.
What the system can do is remove the work that does not need a human. It can calculate, apply constraints, break into sizes, and present a defensible starting point. The planner then does the part that requires knowing the business: adjusting the twenty products where reality differs from the data, and approving the rest.
This changes what a planning review feels like. Instead of building numbers and running out of time before the thinking starts, the team begins with numbers and spends the session on the exceptions. The output is better because attention was spent where it mattered, not because the model was clever.
Size breakdown is the step that gets rushed
Ask a planner which part of the process they would most like to stop doing manually, and size breakdown is a common answer. It is repetitive, it is fiddly, and it has an outsized effect on whether the product is actually available to customers.
The problem is that a size curve is not a fixed property of a product. It varies by market, because body demographics and fit preferences differ between countries. It varies by channel, because the customer buying online is often not the customer buying in a partner's store. It varies by customer type, because a discount retailer and a premium boutique attract different shoppers. And it drifts over time as the brand's customer base evolves.
Using a single historical curve across all of that produces a slow, compounding imbalance. The sizes in the middle of the curve sell out first and get reordered in proportion, so they sell out again. The sizes at the edges accumulate. After a few cycles the style shows perfectly healthy total inventory while the sizes customers actually want are missing.
The damage is easy to underestimate because the reporting hides it. A style at eighty percent sell-through looks successful. If the remaining twenty percent is concentrated in two sizes and those sizes are the ones the market wants, the style has been unavailable in practice for weeks while looking fine on every summary.
Deriving the size curve from what actually sold
The alternative is to let the size split follow observed demand rather than an assumption made when the product was introduced.
This requires being careful about which sales data is used. Sales of a size that was out of stock understate real demand, because a customer who cannot find their size does not register as demand at all, they simply leave. A naive curve built on raw sales therefore reinforces exactly the imbalance it should be correcting. It reads the absence of sales in a stocked-out size as a lack of interest.
A more useful approach looks at periods where the full size range was available, and gives more weight to recent behavior than to old behavior. It also considers whether comparable products in the same category and market share a curve, which matters for new products where there is no history of their own yet.
None of this is beyond a planner's ability. It is simply not something anyone can do by hand for every product, every season, across every market. That is precisely the kind of work worth systematizing: well understood, repetitive, and consequential when it is skipped.
Constraints belong in the suggestion, not after it
A suggestion that ignores supplier constraints is not a suggestion. It is a first draft that someone has to rework before it can be used.
Minimum order quantities change what a sensible order looks like. If the calculation says two hundred units and the supplier's minimum is five hundred, the real decision is whether to order five hundred or nothing. That is a genuine commercial choice about tying up cash for a longer coverage period, and it should be presented as such rather than discovered later by the purchasing team.
Pack sizes, carton quantities, and size ratios impose the same kind of discipline. A theoretically ideal size split that cannot be ordered is worse than a slightly imperfect one that can. Currency, freight thresholds, container fill, and consolidated shipping windows all shape what is actually orderable.
Building these constraints into the suggestion keeps the conversation honest. The planner sees what is possible rather than what would be ideal in a world without suppliers, and the order that comes out of the review can be placed as it stands rather than negotiated back into feasibility afterwards.
Order calendars and the cost of being late
Replenishment is not only about how much. It is about when, and the when is often set by someone else. Suppliers have production windows, factory holidays, and booking deadlines. Freight has cut-offs. Warehouses have receiving capacity.
An order that is right in quantity and two weeks late in timing can be worse than an approximate order placed on schedule. The stock arrives after the demand peak and then sits, which converts a service problem into an inventory problem without solving either.
Order suggestions become significantly more useful when they are aware of the calendar. Instead of only answering what should be ordered, they answer what must be decided this week because the supplier window closes. This reframes the planning routine from a periodic analytical exercise into an operating rhythm that matches how the supply chain actually works.
It also protects the brand from a common failure pattern, where an accurate review happens on a comfortable internal schedule that has drifted out of alignment with the production calendar. The analysis is correct and the timing makes it useless.
Why the spreadsheet version stops scaling
Almost every brand has built this in Excel at some point, and for a while it works. A capable planner can construct a genuinely good replenishment sheet with coverage calculations, size curves, and order proposals.
It stops scaling for reasons that have nothing to do with the quality of the spreadsheet. The data has to be exported and refreshed manually, so the numbers are only as current as the last export. The logic lives inside one person's file, so it cannot be reviewed, shared, or inherited safely. Formula errors are silent, and a wrong range reference produces a plausible number rather than an obvious failure. Multiple copies appear, and nobody is certain which is authoritative.
The deeper cost is the effect on frequency. Because the process is heavy, it runs monthly instead of weekly, or quarterly instead of monthly. Every reduction in frequency increases the average age of the information behind each order, which is the single biggest driver of inventory error in most brands.
The point is not that spreadsheets are bad. It is that a spreadsheet's cost is paid every time the process runs, while a system's cost is paid once. For a routine that should run every week across hundreds of products, that difference compounds quickly.
What changes when the routine gets lighter
Brands that move this work into a shared planning process tend to notice the same set of changes, and they are mostly about time and attention rather than about the model.
Reviews get more frequent, because the preparation that made them expensive has been removed. Orders get placed closer to the moment the need is identified, which shortens the effective planning horizon and reduces the uncertainty every order has to absorb. Size availability improves without a corresponding increase in total stock, because the same units are distributed more sensibly. And the discussion in planning meetings shifts from checking numbers to deciding what to do about them.
There is also an organizational effect that is easy to overlook. When suggestions, constraints, and reasoning live in a shared system, planning stops depending on one person's file and one person's memory. New team members can be productive faster. Holidays stop being risk periods. The process survives the growth of the business rather than being rebuilt each time it outgrows the previous setup.
Closing the last mile of planning
The last mile of planning is not glamorous. It is quantities, sizes, minimums, and dates, and it rarely appears in a strategy discussion about demand planning.
It is also where the value of everything upstream is either realized or lost. A brand with an excellent forecast and a slow, manual ordering process will underperform a brand with an average forecast and a fast, disciplined one, because the second brand acts on current information more often.
Fashion Planner focuses on that last mile: turning demand insight into concrete order suggestions, breaking those suggestions into sizes based on how products actually sell, respecting the constraints suppliers impose, and presenting the result in a form a planner can review and approve rather than rebuild.
The goal is not to remove judgment from ordering. It is to make sure the judgment is spent on the twenty decisions that need it, instead of on the arithmetic behind the other two hundred.
