The procurement visibility gap that costs more than materials
Most procurement teams know their spend number. They do not know whether that number is higher than it should be, which suppliers are drifting from contracted prices, or whether lead times on their critical components are quietly getting worse. That information is in the ERP. It is just not surfaced.
ERP procurement analytics is the practice of pulling those signals out continuously, connecting them across suppliers and sites, and making them actionable before they become problems. The eight signals described in this article are each derivable from transaction data most manufacturers already have. The question is whether the system is configured to surface them.
The pattern across most PE-owned manufacturers: procurement data is captured in the ERP, exported to a spreadsheet, analyzed once a month by one analyst, and communicated to the team two weeks after the period closes. By then, the invoices are paid and the pattern has continued for another cycle.
Purchase Price Variance: from month-end report to real-time alert
PPV is the difference between what you expected to pay for an item (your standard cost or contracted price per unit) and what you actually paid on a given PO. Positive PPV means you paid more than expected. Negative means you paid less. The total number seems straightforward, but the interpretation is not: the same $600K in unfavorable PPV can come from a commodity index passthrough your contract explicitly allows, from a buyer ordering off-contract from a spot supplier, or from a supplier quietly raising prices outside the renegotiation cycle. All three require completely different responses.
Your ERP records the standard price alongside the actual receipt price on every PO line. PPV is calculable at the line item level, by supplier, by category, by plant, and by period. The problem most teams face is not data availability but analysis structure: extracting PPV by root cause, not just in total, and surfacing it within days of a receipt rather than weeks after close. The bridge below shows what that decomposition looks like when structured correctly.
Real-time PPV monitoring means the alert fires when a receipt posts, not when an analyst runs the month-end export. A buyer who ordered off-contract gets a notification the same day. The category manager sees the cumulative pattern within the week. Corrective conversations happen before the invoice clears rather than after.
Demonstrated lead time: what your suppliers actually deliver
Every PO carries a promised delivery date. The ERP records the actual goods receipt date. The gap between the two is that order's demonstrated lead time performance. A supplier who quotes a 14-day lead time but consistently delivers in 22 days has an 8-day negative lead time variance, and your planning system likely does not know it. Over multiple orders, the trend tells you whether a supplier's reliability is improving, degrading, or holding steady.
Demonstrated lead time matters well beyond on-time delivery rates. Buyers who know a supplier delivers late pad their PO timing. That padding inflates the stated lead time in the planning system, which inflates safety stock for every component that supplier makes, which ties up working capital. Getting demonstrated lead time accurate and current for every supplier is the first step to right-sizing inventory buffers across your supply base.
Quoted lead time is what a supplier told you at negotiation. Demonstrated lead time is the rolling average of actual delivery lag across recent POs. When they diverge, you have either a measurement problem or a performance problem worth addressing before it compounds into a stockout.
A single late delivery is noise. A 90-day moving average of delivery lag that is growing quarter over quarter is signal. Lead time trend monitoring identifies supplier capacity or prioritization issues early, before they cause a missed production schedule.
Supplier scorecarding: one accountability view per supplier
A supplier scorecard combines multiple performance signals into a single view that updates each period. The goal is not to reduce a supplier relationship to one number. It is to surface which signals have changed and need attention versus which are holding steady, and to do that automatically rather than requiring someone to build the comparison each month.
The inputs for a standard manufacturing supplier scorecard are all in the ERP or derivable from it: PPV, on-time delivery rate, lead time accuracy relative to quoted, pricing compliance rate, and payment terms captured versus negotiated. The scorecard below shows what a single-supplier view looks like when those signals are combined. The IQ score flags the relationship's overall health and drives the alert routing.
Data enrichment: the foundation every scorecard needs
A supplier scorecard is only as useful as the supplier master underneath it. In a single-ERP environment, the supplier master may be reasonably clean. In a multi-ERP environment, the same physical supplier typically appears under dozens of name variants across plants: "Midwest Precision," "Midwest Precision Castings Inc.," "MPC Castings LLC," each carrying different payment terms, different contract records, and different spend histories. A scorecard built on fragmented records treats one supplier as twelve and produces results that are impossible to act on.
Data enrichment creates a golden supplier record by de-duplicating and normalizing supplier records across every ERP. Once that master exists, all spend, PPV, lead time, and compliance data consolidates under one supplier identity. The scorecard works. The contracted payment terms apply at every plant. The IQ alert fires on the right vendor, to the right buyer, with the full relationship history attached.
A common finding in multi-ERP environments: supplier name duplication rates of 15 to 40 percent before enrichment are typical. A supplier appearing under 12 name variants is effectively paying 12 sets of payment terms, being tracked against 12 separate performance records, and generating PPV numbers that cannot be consolidated without a unified master.
Payment terms rationalization: the savings hiding in your AP aging
Your AP system knows exactly what payment terms are attached to each vendor record. It knows which invoices were paid on day 10, which on day 30, and which on day 45. It also knows, in most cases, whether an early-pay discount clause exists and whether it was captured. Terms rationalization is the analysis of what was negotiated versus what is being executed, run across your full AP portfolio rather than supplier by supplier.
Common findings when this analysis runs for the first time: suppliers being paid meaningfully earlier than their terms require, early-pay discount clauses that are going uncaptured because AP did not flag them, and terms that were negotiated at the acquiring entity but never pushed down to acquired plants running different ERPs. Each gap is recoverable. Each requires a different fix. The analysis is straightforward once supplier records are unified across systems.
Pricing compliance: three prices in play, only one is right
On every purchase, three prices are active: the contract price (what was negotiated with the supplier), the PO price (what the buyer entered at time of ordering), and the invoice price (what the supplier actually billed). Full pricing compliance means all three match for every line on every order. Any gap between these three figures is a signal, and each type of gap has a different cause and requires a different fix.
A gap between contract and PO price typically means the buyer did not have the contracted price loaded as the default for that supplier-item combination, or ordered from the wrong supplier tier. A gap between PO price and invoice price means the supplier billed differently than the confirmed PO, which may be a supplier error, a deliberate price change, or a quantity bracket applied incorrectly. Both types are identifiable from ERP transaction data without any external data sources. The question is whether anyone is looking systematically.
Price list assignment matters here too. When a supplier has multiple price lists by customer tier or volume bracket, the wrong tier applied at PO entry creates a compliance gap that looks like a data problem but behaves like a pricing problem. Clean supplier master data combined with price list verification at the PO line level closes this gap before the invoice arrives.
AI contract capture: closing the loop between negotiation and execution
Most procurement contracts live in PDFs, email chains, and shared drives. They contain negotiated prices by item or category, material index escalation clauses, payment terms, quantity commitments, and quality specifications. None of that information is in the ERP unless someone entered it manually, and manual entry at the granularity of a full supply contract is rarely complete. The result is a gap between what was negotiated and what the system enforces on every purchase.
AI contract capture reads supplier contracts and extracts key commercial terms: base prices, index-based escalation rates and their current values, effective dates, payment terms, and quantity commitments. Those extracted terms are then mapped to PO and invoice records in the ERP, and the system monitors compliance automatically as transactions post. When a supplier invoices at a price that does not match the extracted contract terms, whether because of a stainless surcharge that was miscalculated or a base price that drifted outside the renegotiation window, the discrepancy surfaces without requiring anyone to locate the contract document and perform the comparison manually.
The result is a closed loop. The procurement team's negotiation work actively governs every purchase rather than sitting in a folder. Compliance reporting updates as invoices post. Exceptions route to the right buyer with the relevant contract clause attached. Over-billings are identified and recovered. Under-spend against quantity commitments is flagged before the contract period closes.
AI reads supplier contracts in their native format and extracts commercial terms: base prices, material index clauses, payment terms, quantity minimums, and effective date ranges. No manual data entry. No translation spreadsheet. The terms go directly into the compliance layer.
Each invoice line is matched against the extracted contract terms for that supplier and item. Price deviations, index miscalculations, and quantity bracket mismatches are flagged at the invoice level with the expected value, the actual value, and the estimated recovery amount.
IQ Insights: monitoring without dashboard checking
The value of all the analytics described in this article depends on how quickly they reach the right person. A PPV trend that surfaces through a monthly review cycle three weeks after the receipts posted cannot stop the problem. A lead time degradation that requires someone to log into a dashboard and run a query will be found occasionally, if at all. Procurement analytics has to be pushed to the team, not pulled by it.
IQ Insights is Marquis's AI monitoring layer that watches supplier signals continuously and routes alerts to the procurement team when a threshold is crossed. The alerts are generated from the same ERP data but delivered proactively, to the right person, with enough context to act without additional research.
PPV alerts fire at the PO receipt level, by supplier, by category, by plant. When a supplier's actual price departs from contracted or standard by a defined threshold, the buyer who placed the order receives the alert before the invoice clears AP, while there is still time to act.
When a supplier's 90-day demonstrated lead time increases materially versus their quoted lead time, the alert goes to the category manager. Early warning of capacity or prioritization issues at a supplier gives the team time to pull forward POs or activate alternates before production is affected.
When a supplier's pricing compliance rate falls below threshold over a rolling window, IQ Insights alerts the procurement lead and the AP team simultaneously. The alert identifies which POs are driving the gap, what the estimated exposure is, and whether the pattern is new or an ongoing drift.
When AI contract monitoring detects a supplier invoicing outside extracted contract terms, including price deviations, index miscalculations, and payment term mismatches, the alert routes to the right buyer with the specific contract clause attached and the estimated over-billing calculated.
How Marquis Procurement IQ surfaces all eight signals
Procurement IQ connects directly to your ERP and surfaces every signal described in this article: real-time PPV by root cause, demonstrated lead time versus quoted, supplier scorecards updated each period, pricing compliance at the PO line level, payment terms captured versus negotiated, and IQ Insights alerts that route to the right person when a threshold is crossed. AI contract capture closes the loop between what was negotiated and what the system enforces on every purchase.