Fully autonomous project. The Procurement Codex is built, verified, and published end-to-end without manual authoring. Its core logic — the spine, the layer model, and the platform comparison rubric — is rebuilt and improved on every iteration for continuous method validation. Content is generated programmatically and refined each cycle: treat it as a directional learning aid, verify against primary sources, and send corrections — accuracy and fairness compound with each pass.
Episode 02 · Seeing the money
You can't source, negotiate, or manage risk on money you can't see. This step turns messy ERP and AP transactions into one clean, classified, enriched view — then decides, category by category, where to compete, consolidate, or partner. It is where raw spend becomes leverage.
Below: the plain concept → how every platform does it (fair A vs B vs C) → then we stack process mining, AI, orchestration and people on top until you see why standalone is never enough.
Cleanse, classify and enrich historical spend into a single trusted view, then organise it into categories — each with a deliberate strategy. It is procurement's visibility and planning layer.
Because value starts with visibility. Classified, enriched spend reveals where the money goes, who the real (parent-level) suppliers are, and which categories are worth competing, consolidating, or partnering on — before any sourcing event.
>95% classified spend · continuous refresh · mapped to a taxonomy (UNSPSC or custom) · supplier parent-child normalised · Kraljic-segmented · a named strategy and savings pipeline for every material category.
Same rubric for every vendor, 1–5. We state explicitly what each is best and worst at. Toggle platforms to compare.
| Platform | Best at | Watch-out | |
|---|---|---|---|
Scores are directional teaching aids based on typical deployments, not vendor benchmarks. Your mileage varies by configuration, module licensing, data quality, and integration maturity.
Stack layers onto plain spend analysis and watch the architecture — and the outcome metrics — change. This is the whole thesis of the Codex in one control.
A real best-of-breed spend & category architecture — no single vendor owns all of it. The value lives in the seams.
flowchart LR ERP[(ERP + AP
S/4HANA / Oracle)] --> CL[Cleanse + normalize
supplier parent-child] CL --> AI[L4 ML classification
Sievo / Zycus Merlin] AI --> TAX[Taxonomy
UNSPSC / custom] TAX --> SEG{Kraljic segmentation} SEG --> STRAT[Category strategies] PM[L3 Process mining
Celonis / PM4Py] -.off-contract & tail.-> STRAT BM[External benchmarks
Sievo / Coupa community] --> STRAT STRAT --> ORCH{L5 Orchestration
opportunity to sourcing} ORCH --> SRC[L1 Sourcing / RFx suite]
Standalone, a suite's spend module shows a dashboard but classifies in batches and rarely enriches with external data. Best-of-breed ML (Sievo, Zycus Merlin) lifts classification past 95%; process mining exposes the off-contract and tail spend the cube hides; benchmarks tell you whether a price is actually good; orchestration turns an opportunity into a live sourcing event. No one tool does all five — the leverage is in wiring them together.
Pick one from each column. The Codex assembles the composite and calls out where the seams need engineering. Shown here for spend & category; the same engine powers every episode.
Turn messy transactions into a clean, classified, enriched spend view — then set a deliberate strategy for every category.
% classified · % under management · classification accuracy · savings identified vs. realised · refresh latency.
Value × risk → leverage (compete), strategic (partner), bottleneck (secure supply), non-critical (automate/catalog).
Sievo leads best-of-breed analytics · Coupa on community benchmarks + UX · Ariba/Oracle on ERP-native depth · GEP/Ivalua on configurability · Zycus on AI value.
L3 mining exposes off-contract & tail · L4 AI classifies >95% · L5 orchestration turns insight into sourcing · L6 gives every category an owner.
No single tool cleanses, classifies, enriches, benchmarks and executes. Value is in the seams.