Spend Diagnostic.

• Sourcing Graph, deployed in your environment. Suppliers, items, and categories resolved to canonical identities inside your Snowflake or Databricks. The canonical layer, the similarity index, and the classification engine stay after the engagement ends. Your data never leaves your platform.
• A written diagnostic, typically 15–25 pages. Top opportunities surfaced across supplier fragmentation, item duplication, price variance, frozen-price detection, and tail spend. Sized against canonical entities, not approximations.
• A defensible savings number. Quantified, traceable to specific suppliers and items, defensible to your CFO and to a sourcing committee. The math sits on top of the canonical layer, so the numbers are reproducible after the engagement ends.
• A 90-day plan. What to do first with the findings, sequenced by leverage. Whether with me or without.

• ·Supplier master across ERP, AP, and sourcing systems
• ·Item catalog and PO line history (12–24 months ideal)
• ·Existing category taxonomy, or what should replace it
• ·Contract structure and sourcing event history
• ·Where the spend cube currently breaks down and why
• ·AP reconciliation and how spend ties (or doesn't) across systems

1. 01

   Kickoff (60 min)

   Scope, data sources, access, who to talk to. Confirm the engagement fits before any data work begins.

2. 02

   Data access and Sourcing Graph deployment

   Week 1. Suppliers and items run through canonicalization and the similarity index. First passes of fragmentation and duplicate detection produce a working draft of opportunities.

3. 03

   Working sessions

   Weeks 2–3. With your sourcing analysts, category managers, and AP team. Validation of canonical entities, review of surfaced opportunities, prioritization by leverage.

4. 04

   Write and revise

   Week 3–4. I produce the diagnostic. You review. I revise once.

5. 05

   Readout (60 min)

   Walk through findings, savings numbers, and the 90-day plan with your team and leadership. The Sourcing Graph deployment stays live in your environment.

Two to four weeks, end to end. Remote-first, with optional onsite for kickoff and readout.

Fixed-fee, scoped per engagement. 50% on signing, 50% on delivery. If the scope is wrong before we start, I'll say so and refund the deposit.

You're a good fit if you're:

• ·A procurement, sourcing, or supply chain leader with real data fragmentation problems and the authority to act on the findings.
• ·Operating in industrials, manufacturing, distribution, or any high-SKU, multi-supplier environment.
• ·Past the “maybe we should look at spend” phase and ready to look at the actual data.
• ·A PE operating partner sizing a procurement program across one portfolio company or several.
• ·Considering a larger procurement transformation and wanting a defensible savings number and a working canonical layer before committing.

You're not a good fit if your spend data lives entirely in legacy systems you can't extract from, if you need outputs only as a slide deck rather than a working system, or if you're earlier than “is spend even a problem worth diagnosing?” I'll tell you that on the first call and point you somewhere useful.

The technical analog of this engagement, scoped for AI and data systems more broadly rather than procurement specifically, is the Architecture Review.

Email fbaig@htsmcp.com with a few sentences about your environment: data platform (Snowflake, Databricks, other), ERP, rough spend volume, and what triggered the conversation. I respond within two business days.

Have a system you're stuck on? Talk it through with me in 30 minutes →