One Set of Numbers for Ops, Sales, and Finance: An Apparel Playbook

It is Tuesday morning at a $12M womenswear brand. The head of wholesale is on a call promising a Nordstrom buyer that 480 units of a bestseller will ship by Friday. The ops manager is in Slack telling the DTC team to pause paid spend on the same style because the 3PL flagged a count discrepancy. Finance is closing October and cannot decide whether to book the units sitting in a container at Long Beach as inventory-in-transit or defer them. Three people, three different answers about the same 480 units. Nobody is wrong. They are just reading from three different systems that were never taught to agree.

What are shared operational numbers in an apparel business?

Shared operational numbers apparel teams can actually run on are a small set of figures, updated in near real time, that ops, sales, and finance all pull from the same underlying record. In practice that means one on-hand inventory count per SKU per location, one open-order figure by channel, one shipped-and-invoiced revenue figure, and one landed cost per unit that drives both margin reporting and inventory valuation on the balance sheet.

The test is simple. If the sales director’s pipeline view of committed units, the ops manager’s ATS view, and the controller’s inventory asset account all trace back to the same transactional spine, the numbers are shared. If any of those three views comes from a spreadsheet, a channel-specific report, or a manually reconciled export, they are not shared. They are three versions of the truth held together by whoever is best at pivot tables.

This is Breakpoint 6 territory in the 6 Breakpoints framework. Reporting stops being operational and becomes political the moment two departments cannot agree on the base numbers. You know you are there when meetings open with fifteen minutes of arguing about whose figure is right, before anyone gets to what to do about it.

Why do the numbers diverge in the first place?

The configuration choices my customers wrestle with most, especially in the first month of implementation, are almost always about where the source of truth lives for a given number. And the honest answer is that in most $10M to $20M brands, nobody ever decided. The Shopify install came first. Then a 3PL got bolted on with its own WMS. Then wholesale got a B2B portal, or worse, ran through the DTC storefront with tag-based hacks. Then QuickBooks or Xero was set up by an outside accountant who never saw a purchase order.

Each of those systems keeps its own count. Shopify tracks what it thinks it has in the sellable pool. The 3PL tracks what is physically on the shelf. The wholesale tool tracks what has been committed to open orders. The accounting system tracks what was capitalized when the PO landed. Four counts, four update cadences, four definitions of what counts as available.

The cost of that divergence is boring and continuous. For a $15M brand running wholesale, DTC, and a 3PL, we consistently see 6 to 9 hours per week of pure reconciliation work, usually falling on one operations analyst who has effectively become full-time data plumbing. At peak, when velocity is highest and the counts are most out of sync, oversell rates hit 2 to 3 percent. That is not a rounding error. On a Black Friday weekend that is thousands of units promised to customers that do not exist, followed by cancellations, refunds, and a stack of chargebacks from wholesale accounts who were told the same units were reserved for them.

What does the daily reality look like when the numbers agree?

From the training rooms I lead each month, the shift is visible in the first week of go-live. Not because a magic dashboard appeared, but because the daily standup gets shorter. The ops lead opens the inventory view, the sales lead opens the open-order view, and the finance lead opens the shipped revenue view. All three views come from the same underlying ledger of transactions. There is nothing to reconcile before the meeting starts.

Committed units are visible to sales the moment a wholesale PO is confirmed, and those units disappear from the DTC sellable pool automatically. When a return posts at the 3PL, it hits inventory the same day, not three weeks later when a spreadsheet gets updated. When a container clears customs, landed cost gets allocated across the units on that PO, and margin reports for those styles reflect the actual duty and freight bill, not an estimate from six months ago.

That is what channel-aware ATS actually means in practice. It is not a marketing phrase. It is the difference between telling a buyer 480 units are available because Shopify says so, and telling a buyer 480 units are available because the system has already subtracted the DTC velocity forecast, the units committed to two other wholesale orders, and the safety stock rule for that SKU class.

Which numbers actually need to be shared, and which do not?

A common mistake in the first configuration pass is trying to share everything. You do not need marketing’s CAC and finance’s contribution margin on the same dashboard. You need the small set of numbers that appear in more than one team’s decisions.

In my experience the shared set is roughly this:

• On-hand inventory by SKU by location, updated on every warehouse movement
• Available-to-sell by channel, with wholesale commitments and DTC velocity already netted out
• Open wholesale orders by ship window, with cancel dates and retailer compliance flags visible
• Shipped and invoiced revenue by channel by day, tied to the GL
• Landed cost per unit, refreshed as receipts and duty invoices post
• Returns in flight, and their expected inventory impact by week

Everything else can live in a channel-specific tool. Ad spend belongs in the marketing stack. Payroll belongs in HR. Cash forecasting can live in a treasury tool. But those six items above are the ones that get argued about in every meeting where ops, sales, and finance are in the room together, and those are the ones that must resolve to a single source.

Where does accounting fit into shared operational numbers?

This is the part most mid-market apparel brands get wrong, and it is why so many end up with a beautiful ops system and a completely disconnected P&L. If accounting is a bolt-on that receives a nightly export of journal entries, the numbers will drift. Not because the export is broken, but because the definitions inevitably diverge. Ops calls something a transfer. Finance books it as a movement between two inventory sub-accounts. When the reconciliation fails, nobody can trace the transaction back to the physical event.

The stronger pattern is accounting as a first-class module on the same operational spine. Every physical event, a receipt, a shipment, a return, a transfer, a write-off, generates the journal entry at the same moment it updates inventory. The controller sees the same transaction the ops manager sees, from a different angle. When margin looks off on a style, finance can drill from the P&L line back to the specific PO, the landed cost allocation, and the shipment records without leaving the system.

For smaller brands or brands with strong existing relationships with a Xero or QuickBooks environment, integration is a perfectly good path. But once you are in the $10M to $20M zone with multiple entities, multiple warehouses, or international duty allocations that need to hit the right cost layer, native accounting stops being a nice-to-have. It is the only way the numbers stay shared instead of merely reconciled.

What is the anti-pattern to watch for?

The most common anti-pattern is the reconciliation dashboard. A brand realizes the numbers disagree, so someone builds a BI layer, usually Looker or a Google Sheet with imported ranges, that pulls from Shopify, the 3PL, the wholesale tool, and the GL, and reconciles them into a single view. It looks great in the demo. It buys about six months of relief.

Then it breaks. Shopify changes an API. The 3PL adds a new movement type. A new wholesale channel gets added and nobody updates the pipe. The dashboard now shows a number that agrees with nothing. Worse, people have been trusting it, so when it starts lying, decisions get made on bad data for weeks before anyone notices.

The rule I would put on the wall: never solve a source-of-truth problem with a reporting layer. Reporting layers are downstream. If the underlying systems disagree, adding a dashboard on top just gives everyone a shared hallucination. Wholesale should not run through Shopify’s native flow. Inventory truth should not be assembled nightly from four exports. The fix has to happen where the transactions are recorded, not where they are visualized.

How do you actually get there without a two-year project?

The sequencing matters. Brands that try to move everything at once, PLM, inventory, orders, warehouse, accounting, in a single cutover, usually stall. The sequence that works, in the go-lives I have run this year, is roughly this:

1. Consolidate inventory first. One count per SKU per location, sourced from the operational system, with the 3PL feeding movements in near real time. This alone kills most of the 6 to 9 hours a week of reconciliation.
2. Bring wholesale orders onto the same spine. Committed units now net out of the sellable pool automatically. Oversell drops.
3. Bring DTC channel connections in, so Shopify reads ATS from the operational system rather than holding its own count.
4. Move accounting onto the same spine, either natively or via a well-defined integration, so journal entries fire from physical events.
5. Retire the reconciliation spreadsheets and the shadow dashboards. This step is emotional. Somebody built those spreadsheets. Somebody is proud of them. They still have to go.

Most brands in the $10M to $20M band can get through steps one through three in the first eight to twelve weeks. Step four takes longer if there is an entity restructure or a multi-brand catalog involved, which is where brands like Lufema, running multiple brands and a B2B portal across entities, spend real configuration time. It is worth doing right. A shared chart of accounts across brands is what turns consolidated reporting from a monthly ordeal into a Monday-morning view.

What about brands with drop cycles and same-day fulfillment?

The pressure on shared numbers is highest at brands running frequent drops with same-day or next-day DTC fulfillment. Magnolia Pearl is a useful example here: high drop cadence, international customers, duty complexity, and a returns flow that has to close quickly to keep inventory in play for the next drop. If returns take three weeks to post to inventory, a bestseller sits in a returns bin during the exact window when it could be resold on the next drop.

The operational rule is straightforward. Returns should post to inventory in days, not weeks. That is not a warehouse KPI. It is a shared-numbers problem, because until the return posts, sales does not know the unit is back available, finance does not know the revenue was reversed, and ops does not know the physical count changed. Three teams operating on stale data, all because a returns process is running on a different clock than everything else.

What this means for an apparel operations team

The practical takeaway is that shared operational numbers are an architecture choice, not a reporting choice. If ops, sales, and finance are pulling from the same transactional spine, alignment is the default. If they are pulling from different systems and reconciling downstream, misalignment is the default, and no amount of dashboarding will fix it.

The cost of the wrong architecture is measured in the small daily frictions. Six to nine hours a week of reconciliation. Two to three percent oversell at peak. One analyst who cannot do analysis because they are cleaning data. A weekly ops meeting that starts with fifteen minutes of arguing about whose number is right. None of it shows up as a single line item. All of it shows up in decisions made a beat too late.

The brands that make it through the $10M to $20M zone cleanly are the ones that stop treating reporting as a separate function and start treating it as an output of a well-configured operational spine. The numbers do not need to be reconciled if they were never allowed to diverge in the first place.