See It Work

Before Basal AI

Monday morning in Portland. The Hawthorn & Reed planning office is one room — eight desks, two Merchants who are also doing Sales Planning because the team is that size, and a VP of Planning named Claire who has been opening the same Excel workbook every Monday for four years. The workbook is called MASTER_SS26_v14_FINAL_FINAL_USE_THIS_ONE.xlsx. It has twenty-two tabs. Three of them have broken formulas that Claire has learned to work around. One of them has not been touched since February because someone made a copy and the team lost track of which version was current.

The Spring-Summer 2026 range review finished two weeks ago — or mostly finished. There are fourteen styles confirmed and two that are still in conversation with the supplier. Planning for confirmed styles has already started, because it has to. The rest of the team is planning around the two open styles, building forecasts with a placeholder unit count that everyone knows is wrong but no one is willing to call provisional in writing, because once it is provisional in writing, someone will ask Claire to re-run the whole file and there are only so many hours in a week.

No system tells Claire when it is safe to start planning. That judgment is hers, made on the basis of experience, instinct, and the number of supplier emails currently sitting in her inbox without a confirmed answer. She is the Merchant, the Sales Planner, and the person who will rebuild the size curve estimates by hand when the numbers come back wrong at the end of the season. She is very good at her job. She is also doing the work of three people.

Stride Meridian

Velocity Track

A mid-market footwear brand at $240M revenue, Distributed Org with Global office in Munich and Regional office in North America, Velocity Track. This narrative demonstrates how the system handles multi-office planning, change events from Global, and top-down/bottom-up reconciliation.

Before Basal AI

Monday morning in New York. Sarah Okubo, Regional Merchant for North America, has fifty-three unread emails. Forty-one of them arrived over the weekend. Three are from Munich.

The most important email came in Friday at 5:14 p.m. It is from the Global Planning Director. The subject line is: SS26 Line Update — Action Required. Three colorways have been dropped from the line, effective immediately — a manufacturing constraint at the factory in Vietnam that no one flagged until final sign-off. The email is polite. It is also six years into Sarah's tenure, which means it is the sixth time a version of this email has arrived on a Friday afternoon at the start of a planning weekend.

Sarah knows what the next forty-eight hours look like. She will open the regional Sales Planning file — a Looker export that was current as of Thursday — and begin manually identifying every Key Account forecast that references one of the three dropped colorways. There are, she estimates, somewhere between forty and sixty records. She will update each one. She will notify each KAM who carries those accounts. She will re-run the OTB position, which means calling the Demand Planner in Chicago on Saturday morning. She will do all of this before the Monday stand-up, because if she does not, the Monday stand-up will be about the gap rather than about the commercial response to it.

The three colorways will be out of the plan by Monday. The accommodation window for the factory decision will have quietly closed over the weekend. No one will know how much that cost because no one will have had time to calculate it.

The Commencement Declaration

The range for SS26 is at the point Claire has determined to be sufficiently confirmed — not perfect, but stable enough that planning decisions made against it are more likely to stand than to be materially revised before sell-in. Fourteen of sixteen styles are confirmed. The two open styles have been formally excluded from the planning scope pending supplier confirmation. Claire logs her judgment in the four-field ledger entry: her Librarian ID, the UTC timestamp, the range confirmation basis — fourteen confirmed styles, two excluded pending supplier resolution — and the documented justification. The Commencement Declaration is issued through the authenticated Librarian interface.

The planning chain activates. Not because a timer fired. Not because a threshold was crossed automatically. Because a named human reviewed the conditions, made a judgment, and initiated the chain.

Stride Meridian — Velocity Track and FLR

Stride Meridian's engagement begins three weeks earlier. The Velocity Track integration is faster — the Adapter Layer connects to the Looker export and SPS Commerce API, field mapping is stable after one clean run, and the Threshold Assessment places all five planning functions at OPERATIVE THRESHOLD by Week 2. The data history is two full seasons at Account Name level across the North American account portfolio. The account architecture is already present in the system: KEY_ACCOUNT, ACCOUNT, and REGION_FIELD designations mapped and validated.

The planning chain for Stride Meridian does not wait for a Commencement Declaration. It activates on a different trigger: the Final Line Review. The Product Line Declaration arrives from Munich on a Thursday morning. The Global Planning Director registers the confirmed SS26 product universe — one hundred and twelve style-color combinations, twelve dropped from the prior development line, the Regional Sales Target for North America set at $28M. Both are registered in the system as received declarations. Both are live from the moment they arrive.

The five planning functions receive their inputs simultaneously. No email goes to the North American team saying “the FLR is done, you can start planning.” No meeting is called. The planning chain activates against the registered Product Line Declaration and Sales Target automatically. The Merchandise Planner in New York opens the system Monday morning and finds the framework already built against the confirmed line. The Sales Planner finds the Account Name architecture already populated with the SS26 product universe. The Demand Planner finds the initial Material/Capacity Forecast already submitted from the Merchandise function.

Sarah Okubo, for the first time in six years, does not spend Monday morning telling the team that the FLR just happened and they need to start planning.

Account-Level Forecasting — Stride Meridian

The Sales Planning function at Stride Meridian is building account-level forecasts for North America. The Sales Planner, a senior member of the team named David, is working through the SS26 sell-in plan for the Key Account portfolio. There are eleven KEY_ACCOUNT records in the North American account architecture — department store groups and specialty chains that together represent roughly 60% of regional revenue. Each is planned individually at Account Name level.

What David is not doing is eligibility logic. The Account Name Type and Account Name Tier architecture does that work automatically. When David opens the forecast interface for a specific account — call it a regional department store group with KEY_ACCOUNT / Tier B classification — the product universe visible to him is already filtered to the products that account is eligible to receive. A limited-edition franchise that Stride Meridian has restricted to KEY_ACCOUNT / Tier A only is simply not in the build. The PEM_ACCOUNT_TIER_VIOLATION does not fire because the ineligible product never appears in the sell-in tool for that account. David does not have to remember eligibility rules for eleven Key Accounts across one hundred and twelve style-color combinations. He focuses on the commercial relationship: what this account's buyer is likely to commit to, where there is open-to-buy opportunity, and where the prior season's performance points to a growth conversation.

For the REGION_FIELD account portfolio — sixty-three smaller independent and regional accounts across North America — the forecasting proceeds at Account Channel level rather than individual Account Name level. The system has flagged three REGION_FIELD accounts with ORDER_ECONOMICS_FLAG signals from the prior season: their order sizes are below the minimum economically viable order threshold defined in the Hawthorn & Reed Bespoke Library. David's Account Manager is alerted before the sell-in window opens, not after the order arrives.

The LOW Confidence Output — Hawthorn & Reed, Month 2

In Month 2 of the planning season, the Demand Planning function produces an output for Hawthorn & Reed's knitwear category. The output is a Demand Plan for three styles in the Extended Sizes range — styles that, per the Threshold Assessment Record, fall under the Indicative Threshold limitation identified at engagement start.

The output carries a confidence rating of LOW. The Output Limitation Disclosure reads:

“This Demand Plan is produced from less than one full season of SKU-level sell-through data for Extended Sizes in the knitwear category. Size curves are derived from category-level estimates rather than style-specific history. PO Monthly Flow calculations do not have a statistically defensible size distribution at the style level. Treat this output as directional. Do not act on it without supplementary Merchant judgment and explicit Arbiter authorization.”

Claire sees the rating and the disclosure before anyone else does. The system requires her acknowledgment before the output is cleared for wider presentation. She reads it. She confirms it in the ledger. She notes in her review commentary that the Extended Sizes knitwear gap was documented at engagement start and that the plan was always to supplement system output with Merchant judgment for this category in Season 1.

The output goes to Marcus — acting as Arbiter for material planning decisions — with the disclosure attached. Marcus reviews the Gap Limitation, asks whether the direction of the estimate is plausible given what the team knows from wholesale sell-in conversations, receives confirmation that it is, and authorizes the team to treat the output as directional. He does not authorize it as a plan. He authorizes it as a starting point, explicitly supplemented by the Merchant's judgment on the Extended Sizes distribution for the knitwear category. The authorization is logged with his identity and timestamp.

The planning team now has a directional estimate and a clear record of exactly what it is and how it should be used. In the prior season, the same estimate would have been a number Claire put in a cell with no record of how she got there, no confidence rating, and no authorization trail. A LOW confidence output with an acknowledged disclosure is not a failure of the system. It is the system being honest.

Stride Meridian — The Friday 5 p.m. Email

The email arrives on a Friday at 4:58 p.m. Munich time. The subject line: SS26 Line Update — URGENT. Three colorways dropped. Effective immediately. The same message Sarah Okubo has received, in some form, six times in six years.

What happens next is not what used to happen.

The change is registered against the Stride Meridian Product Line Declaration within the hour. Three CLASS 1 and CLASS 2 change events — two colorway cancellations and one volume reduction — are classified at ingestion. The Blast Radius simulation runs simultaneously for all three events. The North American Regional Pushback Package generates automatically.

Sarah Okubo does not open the system until Monday morning. When she does, the following is waiting:

The complete financial impact of the three changes on the North American plan — revenue exposure by account type, OTB release by category, and the factory accommodation window calculation. The factory for one of the cancelled colorways is still within the accommodation window. The window closes Tuesday at end of business. If North America wants to negotiate for a partial production run, there are thirty-two hours to make that case.

The impact on Key Account forecasts: six KEY_ACCOUNT records carry material exposure to the cancelled colorways. Two of them — the accounts with KEY_ACCOUNT / Tier A classification — are carrying the largest exposure. The system has pre-ranked the accounts by impact severity and KAM coverage.

The sample priority implication: physical sample sets for the cancelled colorways have already been prioritized for three KEY_ACCOUNT sell-in visits scheduled for the following week. The system flags the conflict and recommends revising the travel set configuration before the samples are pulled.

Sarah reviews the simulation. She forwards the Pushback Package to the Global Planning Director before 9 a.m. Monday with a cover note that took her fifteen minutes to write because the numbers were already in the package. She authorizes the change event with Option B — request accommodation for the factory with the narrowest window — and accepts the change for the other two events. The Arbiter authorization is logged. Every downstream record updates.

The North American team spends Monday making commercial decisions. The two KAMs with KEY_ACCOUNT exposure have numbers before their 10 a.m. calls. The sample team has revised travel set instructions before the weekly samples meeting. The Demand Planner in Chicago does not receive a Saturday phone call.

SAP handles the record of what happened. Basal handles the speed of what to do about it.

The Reconciliation

The Librarian — in this case, the Regional Planning Manager who acts as Librarian for the Stride Meridian engagement — initiates the Reconciliation Event. The system generates the Parameter Opportunity Set: the levers available within the Statistical Credibility Ceiling. These are not invented. They are the specific planning parameters that, if adjusted within statistically defensible bounds, would contribute to closing the gap.

The REGION_FIELD channel estimation is the first lever. The channel-level estimate is provisional — $800K of the gap exists because the bottom-up account-by-account build has not been completed. Completing that build, using the system's REGION_FIELD intelligence layer to fill accounts where historical data is sufficient, is expected to recover $600K to $700K of the gap without requiring any stretch assumptions. This is legitimate. The gap exists because the estimate is incomplete, not because the business cannot perform.

The KEY_ACCOUNT adjustment is more delicate. The three accounts with softer pre-order signals represent $1.4M of the gap. The Parameter Opportunity Set identifies that the prior-season growth trajectories for two of those accounts are consistent with a recovery pattern in Q3 — the sell-in signals may be soft because buyers are managing open-to-buy cautiously early in the season, not because the relationship is declining. An assumption adjustment within the Statistical Credibility Ceiling — based on the two-season trajectory and the documented early-season caution pattern — could support $500K to $700K of recovery in the KEY_ACCOUNT forecast without exceeding what the historical record can support.

The Reconciliation Blast Radius runs against both proposed adjustments simultaneously. The Librarian prepares the Strategic Justification payloads: for the REGION_FIELD gap, the business rationale is straightforward — the estimate was incomplete, not wrong. For the KEY_ACCOUNT adjustment, the rationale references the documented buy pattern of the two accounts, the two-season trajectory, and the early-season caution pattern observed in prior years. Both payloads are submitted with the RE-ID from the Gap Declaration.

The Reconciliation Blast Radius report goes to Sarah Okubo acting as Strategic Arbiter. Sarah reviews the Gap Anatomy Report, the two Strategic Justification payloads, and the combined downstream impact of both proposed adjustments. She authorizes both. The adjusted plan aggregate reaches $26.3M — a $2.1M recovery against the gap.

That leaves $1.7M unrecovered. The system cannot support closing this portion of the gap through any planning assumption that meets the Statistical Credibility Ceiling. The inventory assumptions required, the growth assumptions required, or the market size assumptions required to reach $28M from the current bottom-up state would exceed anything the historical record can defend. The system generates the Variance Declaration.

“The current plan cannot reach the stated target of $28M through planning assumptions within statistical credibility bounds. The residual gap of $1.7M requires either a revision of the top-down target or the explicit authorization of assumptions that exceed historical precedent. This declaration is logged in the immutable ledger and requires Arbiter disposition.”

Sarah reviews the Variance Declaration. She chooses Option A: revise the target. She takes the reconciliation record — the Gap Declaration, the Gap Anatomy Report, the Reconciliation Blast Radius, the two authorized parameter adjustments, and the Variance Declaration — to the Global Planning Director in Munich. She does not walk into that meeting asking for relief. She walks in with the math.

The Global Planning Director revises the North American Sales Target to $26.5M. The plan is honest. The Arbiter has signed it. The gap between what the plan can credibly support and what Global originally wanted is documented, named, and resolved — on the record, in the immutable ledger, with every decision traceable to the human who made it.

For the VP of Planning reading this: think about how many end-of-season reviews have been spent trying to explain a gap that no one wrote down at the moment it was identified. This is what it looks like when the gap is written down at the moment it is identified, with the math attached.

Stride Meridian

At the end of the season, the North American Regional team has something it has never had before: an immutable record of every planning decision made in the SS26 cycle. The Product Line Declaration, the four change events that arrived from Munich, the Blast Radius simulations, the Arbiter authorizations, the Gap Declaration, the Reconciliation Event, the Variance Declaration, the revised Sales Target. Every decision, every authorization, every change event — logged, timestamped, and traceable.

The Monday Allocator Gem has become the document the Regional Merchant actually uses to open the week. Not a spreadsheet that was exported Thursday and is already four days stale by Monday morning. A live output built from the weekly REGION_FIELD sell-through signals and the current account architecture, with every mid-tier SKU reviewed, every ORDER_ECONOMICS_FLAG surfaced, and the top items requiring commercial action ranked by financial impact.

The reconciliation record for the SS26 season goes to Munich with the end-of-season debrief. Global can see, for the first time, exactly where the $1.7M Variance Declaration originated — not as an explanation of why the region missed the target, but as a documented record of why the target was revised and what the plan can credibly support at the current trajectory. The conversation between Munich and New York about the SS27 target begins from that record, not from a gap that no one wrote down.

Sarah Okubo will still receive emails from Munich on Friday afternoons. The line will change. Products will be cancelled. Colorways will be dropped. That is the nature of mid-market fashion. What will be different is what happens on Monday morning.