0600
Decision Support Platform

Decision support for consumer brands

An operator-built, AI-native inventory and sales intelligence platform. Live in production today, onboarding new brands in days, not quarters.

First read of the day
© 0600 LLC
The problem

Brands run on systems that don't talk to each other

A typical brand's operating picture is split across an ERP, an e-commerce platform, wholesale portals, 3PL reports, and a layer of spreadsheets holding it all together. Nobody holds a source of truth, so the questions that drive margin get answered late, or not at all.

ERPShopifyWholesale portalsRetail / POS 3PL / warehouse reportsRetailer selling reports…and the spreadsheets stitching them together
01
"How are we actually doing, and where?"
Revenue vs plan and the channel mix across DTC, wholesale, and own retail. The business-level read lives in five systems and one person's head.
02
"How are we selling at our key accounts?"
Retail-partner sell-through by account, door, and style arrives as scattered selling reports, so the brand finds out when the markdown request lands, not before.
03
"What's aging, where, and what's it worth?"
Consolidating stock across locations is a manual, days-long exercise, and aged inventory quietly compounds. One ops lead spent three hours per aged-stock report.
04
"What should we buy next season?"
Buy depth, size curves, and channel mix get decided on instinct plus last year's spreadsheet, not on the full selling history.
The platform

Any data in. Decisions out.

The platform ingests whatever a brand can export, with no system replacement and no IT project, normalizes it into one canonical model, computes the analytics deterministically, and puts an AI layer on top that explains, but never invents.

01 · Ingest

Any export, recognized

Excel, CSV, even PDFs, from ERPs, Shopify, wholesale portals, and 3PLs. Files are recognized by content, not by filename.

02 · Normalize

One canonical model

Each brand's raw vocabulary maps to a shared schema via a per-brand config: configuration, not custom code.

03 · Compute

Deterministic analytics

Aging, tiers, sell-through, realizable value, size curves: computed by verified logic, locked by automated checks.

04 · Decide

Views and AI on top

Role-appropriate dashboards, exports, and an AI analyst that answers questions against the computed numbers.

Design principleThe AI never does the math. Every figure on every screen, and every number the AI cites, comes from the deterministic engine and is traceable to source data. That's what makes it trustworthy enough to run a business on.
In production

Real brands run on it, every day

This is not a roadmap. The platform runs live, from full-platform flagship deployments to a single focused module, for brands across contemporary fashion, sportswear, and outerwear. Same core, configured per brand, with data isolation between brands as a hard rule.

Selected deployments · identities under NDA
Flagship · Full production

NY contemporary fashion house

Global inventory visibility, aging, order book, and realizable value across warehouses and regions.

Deepest · Full platform

Heritage sportswear brand

Sales, returns, customer health, sizing, seasonal planning, performance, and the AI analyst.

Focused · Single module

Global outerwear brand

Retail-partner sell-through by account and door. The platform scales down as cleanly as it scales up.

And for anyone who asks: a fully working demo brand running real production data, deterministically anonymized and leak-verified. Not screenshots. The live product.

Days
From exports to live
Weekly
Release cadence
10
Analytics modules
1,600+
Automated tests
The modules

Ten modules today, more with every brand

Each module is independently switchable per brand: start with what your data supports, light up more as sources come online. And the library itself compounds. New modules are scoped with a real brand that needs them, then productized for every brand after.

Inventory

Unified stock state, aging, and tier classification across every location.

Order Book

Open orders, WIP, in-transit, and pipeline coverage against stock.

Sales History

Shipped sales, velocity, sell-through, and pacing by channel.

Returns

Return rates, RMA history, and problem-product detection.

Customer Health

Account base, top customers, and early warning on slowing accounts.

Sizing Analysis

Category-by-channel size curves and season-over-season trend.

Realizable Value

What aged stock is actually worth, graded by size-run integrity.

Seasonal Planning

Buy-decision worksheets grounded in real selling history, with exportable planned orders.

Performance

Full-price vs off-price sell-through, discount depth, and movement.

Key Account

Retail-partner sell-through by account, door, and style.

+

The next module

Scoped with the brand that needs it. When it lands and your data supports it, it simply appears in your workspace.

Plus the cross-cutting surfaces every brand gets: composed views for location, tier, velocity, and aged detail, a flexible export builder, and self-serve data upload.

The suite you join is the smallest it will ever be.
The AI layer

An analyst in the dashboard, grounded in real numbers

  • Ask anything, in plain language. The AI queries the same deterministic engine the dashboard uses, by channel, season, customer, or category, and answers with the actual figures.
  • Charts on demand, data you can keep. Answers render as live visuals, and every visual downloads as Excel or CSV that matches the platform's numbers byte-for-byte.
  • Auto-generated briefings. The landing view opens with a computed summary of what changed since the last data drop, so no one has to go looking.
  • Guardrailed by design. The AI reads computed results only. It cannot fabricate a number, and month-in-progress data is explicitly badged so comparisons stay honest.
✦ AI Analyst · illustrative exchange
Which categories are driving our aged inventory, and what's it worth if we move it now?
Four categories hold most of your 365-day-plus stock. At current recovery grades, realizable value is concentrated here:
Aged units by category · illustrative↓ XLSX · CSV
Outerwear
2,410
Knits
1,675
Dresses
1,232
Accessories
734
Outerwear is 38% of aged units but holds size runs graded A/B: strongest candidates for a curated off-price offer rather than deep markdown.
How it's built

AI-speed development, production-grade discipline

The platform is built AI-natively, which is why a lean team ships weekly what a traditional vendor ships quarterly. What makes that safe is the verification architecture around it.

Correctness

1,600+ automated tests and locked canaries

Every brand has a canary: a locked set of expected values recomputed on every change. If a number moves that shouldn't, the change doesn't ship.

Verification

Independent parity oracles

Separate verification programs recompute the analytics from raw data by a second path and must agree byte-for-byte with production.

Privacy

Brand data isolation

Per-brand data stores, NDA-grade confidentiality, role-based access, and a demo environment whose anonymization is machine-verified for leaks on every refresh.

The differentiator

Built by an operator, not a software vendor

The platform wasn't designed from a feature list. It encodes the questions an operator actually asks, and the decisions those answers have to feed.

  • Every module began as a real operator problem. Aging, realizable value, buy depth, key-account sell-through: each exists because a brand operator needed the answer, not because a roadmap committee imagined it.
  • Onboarding runs in the brand's language. Implementation is led by someone who has lived merchandising calendars, wholesale relationships, markdown pressure, and 3PL realities. Discovery takes days, not months of requirements workshops.
  • Operator judgment, AI-native execution. The operator sets what matters; AI-native development ships it weekly, gated by production-grade verification. Enterprise output at a pace and price growth-stage brands can actually access.
Erik Ulin
Founder · New York + Stockholm

Two decades running fashion and retail businesses: President USA at J.Lindeberg, President of Men's Fashion at UBM (PROJECT, The TENTS), VP Marketplaces at NuORDER, and SVP Revenue and Operations at Ghost.

Before operating: management consulting and restructuring at Arthur D. Little and FTI Consulting (Lehman Brothers, Delphi, Intrawest). Columbia MBA. Swedish military intelligence background, which is where the name 0600 comes from.

The platform exists because he needed it himself, in every one of those seats.

The practice

Consulting, with the platform as the working medium

0600 is also a consulting practice for fashion and retail. Strategy engagements run alongside the platform, and the platform is often how the work lands: the deliverable is a running system, not a deck.

Wholesale strategy

Growth without dilution

Channel strategy, key-account planning, and sales-organization design, from specialty doors to majors, built by someone who has carried the order book.

Workflows and decision systems

People, process, platforms

How the team works, decides, and reports. AI enablement and decision-support tooling, with the premise of improving work, not replacing people.

Special situations

When it matters most

Diligence, restructuring, market entry, and interim operating leadership, drawn from a restructuring and P&L background.

Working with 0600

Designed for a fast yes, and a compounding relationship

Step 1 · Days

White-glove onboarding

We take the brand's existing exports and stand up a configured environment. Analysis first, so the first thing the team sees is their own data telling them something new.

Step 2 · Modular

A base, plus the modules that matter

Brands run a lightweight base subscription and add modules as needs surface. No monolithic license, no seat-count games. The platform earns each expansion.

Step 3 · Compounding

Design-partner economics

Early brands shape the roadmap and keep preferential terms as the platform matures. Advisory work is available as a separate, clearly scoped stream.

No rip-and-replaceThe platform sits on top of whatever the brand runs today. It is deliberately additive, which is why it can be live and useful in the first week, and why it survives, and de-risks, ERP transitions rather than competing with them.
Next step
See it live, on real data.

A fully working demo environment, running real production data that has been deterministically anonymized, is available on request. Fifteen minutes in the live product says more than any page.

Contact
Erik Ulin · Managing Director, 0600 LLC
Base
New York + Stockholm