What is a WSSI and how do you build one?

WSSI stands for Weekly Sales, Stock and Intake.

Those three words are the whole idea. Every week of a season has a number for what you sell, a number for what you are holding, and a number for what is arriving. The WSSI is where those three numbers are planned together, compared against last year and against budget, and re-forecast as the season actually trades.

The weekly part is not a convention, it is the point. A monthly view is too coarse to catch a trend while you can still act on it: by the time the month closes, the intake is landed and the decision is gone. Daily is mostly noise. A week is long enough to be a signal and short enough to still change the buy, which is why it is the tempo the whole trading calendar runs on.

It is planned at retail value rather than units because that is the language the budget and the open-to-buy already speak. One currency runs from the finance target all the way down to what the buyer commits. And it is built by hierarchy, department to category to subcategory, so a planner can look at the entire business or at one shelf of it without changing grid.

Effectively, yes. Both describe the same job: planning sales, stock, intake and open-to-buy at weekly level. WSSI is the common term in UK and Australian retail. MFP (Merchandise Financial Planning) is more common in the US. If someone hands you an MFP and someone else hands you a WSSI, you are looking at the same work under two names.

The distinction people sometimes draw is that MFP names the discipline and WSSI names the thing it produces, the weekly grid itself. Almost nobody is that careful in practice, and both terms get used for both. It is not worth correcting anyone over.

Where it does matter is in a business running both. The vocabulary is not the risk, the construction is. Before you compare a US plan to a UK one, check they are on the same trading calendar and that stock is stated the same way in both, at cost or at retail. Two grids with the same name can still be answering different questions.

Every WSSI is a table of weeks across the top and rows down the side. The rows vary between retailers, but these are the ones that do the real work.

That is the whole engine. Each week's closing stock becomes the next week's opening stock, so a change in week two ripples through to week twenty. It is also why a WSSI kept in a spreadsheet gets dangerous: one broken formula in one week quietly rewrites the rest of the season.

Numbers make this obvious in a way definitions never do. Here is one category, planned across six weeks, at retail value in thousands of dollars. Sales are shown as one line here to keep it readable, though a real WSSI splits full price from markdown. Then here is what happens when it trades ahead of plan.

The plan

Read one row and you have read them all. Week 1 opens at 500, takes in 120, sells 95, and closes at 525. That 525 opens week 2. The plan walks stock down from 500 to 290 across six weeks while selling 670 in total, which is exactly the intention: end the phase clean, without a mountain of stock heading into markdown.

What happens when it trades ahead

Now the category sells about 12% ahead of plan for three weeks. Good news, and the WSSI is where you find out what it costs you.

Now check cover. At the end of week 3 the plan expected 450 of stock against roughly 120 of average forward weekly sales, which is 3.8 weeks of cover. What you actually have is 410 against forward sales now running nearer 134 a week, which is 3.1 weeks of cover. The category is selling better and heading for empty shelves at the same time.

That gap is the entire value of a WSSI. It does not tell you the good news, your sales report already did that. It tells you the consequence of the good news, early enough to do something about it: chase more intake, pull stock forward, or accept the sell-out and protect the margin.

Open-to-buy is the answer the WSSI exists to produce. It is how much more stock you are allowed to commit to, and it falls out of the plan rather than being decided separately.

The calculation, at retail value, for any given period:

In plain terms: work out where you need to end up, add what you expect to sell on the way, then subtract what you already have and what is already coming. Whatever is left is what you can still buy.

Put the category from earlier through it. At the end of week 3 you are holding 410. You still want to close week 6 at 290, you now expect to sell 403 across the last three weeks rather than 360, and 200 of intake is already on order. That is 290 plus 403, minus 410, minus 200, which leaves 83 of open-to-buy.

On the original plan that number was zero: 290 plus 360, minus 450, minus 200. The buy was fully committed. Selling ahead is what opened the 83, because finishing in the same place while selling more means more stock has to come in. That is the chase the falling cover was asking for, and the WSSI is what sized it.

This is why the WSSI and the buy cannot live in different places. If the WSSI moves and the OTB does not move with it, the buyer is spending against a number that stopped being true weeks ago.

A WSSI is not a document, it is a weekly meeting with a grid attached. The rhythm is roughly the same everywhere, and so is the part that never makes it into the process doc: how much of the meeting goes on agreeing the numbers before anyone gets to talk about the decisions.

Then it happens again next week. A WSSI that gets built once a season and never touched again is not a WSSI, it is a budget with weeks drawn on it.

Almost every retailer starts here, and for a while it is completely fine. It stops being fine for reasons that have nothing to do with how good the planner is.

This is the gap the industry keeps measuring. IHL Group found that AI and machine learning are already producing results in demand planning and forecasting for 76% of retailers, while fewer than one in four have rolled it out in the areas where inventory distortion actually happens. Forecasting was never the hard part.

In Merchmix the WSSI does not start from a blank grid. The budget arrives from Budget Builder, business targets already broken down to department level, so the weekly plan is anchored to the number finance signed off rather than a number someone retyped. Option plans set the constraints on the forecast. Sales and stock come in live from your connected POS and ERP, alongside pricing and promotion data, your trading calendar and your product hierarchy.

Planners edit the forecast directly in the grid, with version control, so a plan can be saved, compared against an earlier one, and restored when the call turns out to be wrong. It carries this year against last year and the year before, splits full price from markdown, tracks discount depth, and moves from department to category to subcategory without changing grid. Once a version is committed it becomes the baseline: the system of record for the weekly trading plan.

The grid also does some of the looking for you. Merchmix AI reads the same weekly numbers and surfaces what is worth a second look: where cover is drifting, where stock is at risk, and where there is upside worth chasing, with the reasoning shown alongside it. The planner keeps the pen.

That is where it stops being a grid. The committed WSSI produces the weekly sales forecast, the stock flow plan, the markdown plan and the variance against last year, and those become the inputs the rest of the platform runs on. Allocation and replenishment take their chase and distribution logic from it. Continuity reads it to protect the never-out lines. Markdown decisions are made against it and write back to it when a price changes. It feeds Budget Builder for financial alignment, and the dashboards, trade reviews and finance forecasts all read the same committed numbers.

So when a planner re-forecasts on a Monday, the open-to-buy moves with it, the chase recommendation updates, allocation reads the new position, and the executive view is already telling the same story. Your team still makes every call. The platform makes sure the call reaches the places it needs to reach, while it still counts.