Background
Crème Collective has outlined an ambitious vision: to deliver the most advanced retail intelligence reporting in the beauty industry. This document organizes that vision into a phased roadmap based on a data availability audit of existing pipelines, APIs, and integration surfaces.
Phases are sequenced by data readiness — what we can deliver soonest with the data already flowing, versus what requires new platform integrations, credential handoffs, or third-party evaluations. This approach ensures value is delivered from Day 1 while more complex integrations are built in parallel.
Source Tag Data already flowing in pipeline
Source Tag API available — needs credentials or configuration
Source Tag Requires new integration or browser automation
Source Tag Manual input or no programmatic access
Why first: Approximately 95% of the data required for this module is already flowing through the Camelot 3PL OData API into our data warehouse. This is the fastest path to a live, working dashboard — and it directly supports one of Crème's core revenue lines (beauty-specialized fulfillment). Delivering this first proves the platform's value with minimal integration risk.
📦
Warehouse & Fulfillment Dashboard
Real-time visibility into inventory health for fulfillment clients. Proactive alerts replace reactive fire drills.
- Weeks of supply by SKU
- Out-of-stock risk alerts (60-day horizon)
- Slow-moving inventory flags
- Fill rate by retailer
- Operational risk meter (Green / Yellow / Red)
Camelot 3PL — Inventory Items
Camelot 3PL — Piece Inventory
Camelot 3PL — Posted Lines
Camelot 3PL — Pending Lines
Camelot 3PL — Item Master
Camelot 3PL — All Transactions
Why second: Both modules in this phase pull primarily from HubSpot, which has one of the strongest CRM APIs available — and we already have an adapter scaffolded in the codebase. Door tracking data lives in HubSpot custom properties and internal Google Sheets, both of which have existing adapters. These are configuration-level integrations, not new builds. High perceived value for Leilah, low integration effort.
📈
Sales Pipeline Probability Model
Transform the HubSpot pipeline from a list of opportunities into a revenue-weighted forecast.
- Opportunity size (annualized)
- Stage tracking (intro → buyer follow-up → sample → planogram → closed)
- Close probability percentage
- Expected close quarter
- Revenue-weighted pipeline value
HubSpot CRM — Deals API
HubSpot CRM — Pipelines API
🗺️
Distribution Map & Door Tracking
Visual distribution intelligence — see where brands are placed and where they're growing.
- Door counts by channel (specialty, prestige, mass, spa, resort, DTC)
- Geographic map with retail location pins
- Net door growth over time
- Door quality scoring (A / B / C tier)
HubSpot CRM — Companies / Custom Properties
Google Sheets — Internal Records
Door Quality Scoring — Manual Classification
Why third: This is the core client-facing deliverable, but it depends on retailer portal access — and each portal is its own mini-project. We already have Camelot data (fulfillment side), email-based ingestion for Goop, and a working Playwright Lambda framework for portal scraping (Revolve is built). But scaling this to Nordstrom, Sephora, Target, and others requires portal credentials, testing per site, and handling format variations. The framework is in place — execution scales linearly with portal count. Starting this in Phase 3 gives time to coordinate access while Phases 1–2 are delivering value.
📊
Monthly Client Sales Report (Automation)
Automate the existing monthly report that currently requires manual data pulls. Sell-in, sell-through, and inventory metrics aggregated per brand, per retailer.
- Net sales by period (MTD / QTD / YTD)
- Sell-in vs. sell-through comparison
- Open POs and expected ship dates
- Forecast vs. plan variance
- Commission earned / accrued
Camelot 3PL — Sales & Transaction Data
Gmail — Retailer Email Reports (Goop)
Revolve — S3 Direct Access
Nordstrom Vendor Portal — Playwright Automation
Sephora Vendor Portal — Playwright Automation
Target Partners Online — Playwright Automation
Other Retailer Portals — Per-Account Build
Google Sheets — Forecast & Plan Data
Commission Rates — Internal Spreadsheets
🏪
Account Status Dashboards
Per-retailer view showing the full health picture for each key account. Builds on the same portal data as the monthly report, adding account-level analysis.
- Sales velocity by SKU within each retailer
- Category ranking (where available)
- Promo performance recaps
- Margin performance
- Replenishment status and expansion probability
Retailer Portals — Same sources as Monthly Report
Camelot 3PL — Replenishment & Inventory
Google Sheets — Margin & Promo Tracking
Why fourth: This is Leilah's most personally important ask — replacing the manual Word document her Controller fills out weekly. However, it has the widest integration surface of any module: QuickBooks (already via Fivetran, but limited to P&L and balance sheet), bank account feeds, credit card balances across four issuers, Bill.com for accounts payable, and a payroll provider. The key decision here is whether to use Plaid as a banking aggregator (one integration for bank + credit card feeds) or connect to each institution individually. That evaluation, plus credential coordination with Andrea, makes this the right place in the sequence — not because it's less important, but because it's the most dependent on external access and platform decisions.
💰
Weekly Cash Flow Dashboard
Automate the weekly financial snapshot currently filled out manually by the Controller. Live data replaces a static Word document.
- Cash position across Crème Collective + Fulfillment (vs. $750K target)
- Accounts payable summary (next 60 days)
- Payroll coverage status by entity
- Credit card balances — AMEX, Chase, Capital One, BREX (on-track-to-$0 tracking)
- Loan balances and payoff timelines
- 6-month cash trajectory with projected vs. target
- Overall health indicator (Green / Yellow / Red)
QuickBooks Online — P&L, Balance Sheet (via Fivetran)
Bill.com — Accounts Payable API (REST, OAuth 2.0)
BREX — Card Balances API (REST)
Plaid — Bank Feeds + Credit Card Aggregation (evaluation needed)
Payroll Provider — API TBD (Gusto / ADP / other)
Loan Balances — Manual Entry or Google Sheets
Why fifth: These modules are a computation and presentation layer on top of Phases 1–4. No new data sources are required — the value comes from synthesizing existing data into executive-ready formats. This phase becomes possible (and meaningful) once the underlying data from earlier phases is flowing reliably.
🎯
Executive Snapshot (2-Page Board Brief)
A single, synthesized view designed to be shared with investors, board members, or brand founders.
- Composite "Financial Health Score" (1–10 rating)
- Growth trajectory, cash conversion, inventory efficiency
- Gross margin snapshot and A/R aging summary
- Key risks and opportunities in plain language
Derived from Phases 1–4 data
🔮
Forecasting & Scenario Planning
Move from single-point forecasts to scenario-based planning so leadership can see implications before making decisions.
- Base case, upside case, and conservative case forecasts
- Cash flow and inventory implications per scenario
- "If retailer X expands by 100 doors, here's what it requires operationally"
Derived from Phases 1–4 data
✍️
Founder's Strategic Letter
A structured, AI-assisted template that gives each report a founder-level perspective — personal, strategic, and forward-looking.
- 3 wins this month
- 2 strategic risks
- 1 bold recommendation
- 1 macro industry observation
- Overall Crème growth update
Derived from Phases 1–4 data
AI-assisted narrative generation
Why last: These modules represent Leilah's most ambitious vision — evolving Crème from a sales agency with good reporting into a beauty intelligence company. The data sources here are largely unstructured (web scraping, manual curation, meeting transcripts) and don't have clean API paths. Scope and feasibility will be defined after Phases 1–5 are operational and delivering value.
🧴
Product & Market Feedback
- Structured feedback: packaging, pricing, positioning
- Consumer response themes
- SKU rationalization insights
Manual Input / Surveys
Google Sheets — Structured Forms
🔍
Competitive Intelligence
- 3 brands gaining traction in category
- 3 SKUs driving velocity in comparable sets
- Retail trend watch (curated, not copied)
Web Scraping — Competitor Sites
Manual Curation
📣
Marketing & Retail Activation
- Retailer campaign calendar
- Promo performance recaps
- In-store events & BA metrics
- "Retailer Mood Index"
Retailer Portals — Campaign Data
BA Metrics — Manual Reporting
- AI-summarized meeting notes from notetakers
- Key highlights extracted per account
- Action items tracked automatically
Meeting Notetaker — API (Fireflies / Otter / etc.)
Key Considerations
Existing Infrastructure
The Camelot 3PL OData pipeline (18 endpoints), Gmail email parsing, Google Sheets adapters, and a Playwright-based Lambda framework for portal scraping are already built and operational. This platform builds on top of that foundation.
Retailer Portals Scale Linearly
The browser automation framework (Playwright on AWS Lambda) is architected for extensibility — each new retailer requires a workflow class, not a new system. Revolve is live. Nordstrom, Sephora, Target, and others are a matter of credentials + per-portal development. Each portal is an independent unit of work.
Plaid Decision Required for Phase 4
The Cash Flow Dashboard's biggest open question is how to access bank account and credit card balances. Plaid would provide a single integration point for Chase, AMEX, and Capital One feeds. Without Plaid, each institution requires individual handling (some have no API, requiring manual entry or scraping). This decision should be made before Phase 4 development begins.
Credential Coordination
Multiple phases depend on the Crème team providing access: retailer portal logins (Phase 3), Bill.com and banking credentials (Phase 4), payroll provider details (Phase 4), and HubSpot API key (Phase 2). Early coordination prevents these from becoming blockers.
Target Launch — August 2026
A brand launching in Target in August will need dedicated sell-through and inventory reporting. This fits into Phase 3 (Monthly Sales Report + Account Dashboard) as a specific retailer portal integration. Portal access setup should be initiated early.