Why do DSO and DPO metrics differ across PE portfolio companies?

Each company typically uses a different denominator (trailing 30-day, 60-day, or 90-day revenue) and a different aging bucket cutoff in their ERP. Without a shared KPI contract, the same company can report DSO figures that differ by 8-15 days depending on whose definition is used.

What is the fastest way to normalize working capital metrics across a multi-ERP PE portfolio?

The fastest approach is to define a shared KPI contract first — agreed definitions for DSO, DPO, and CCC with explicit denominator and aging bucket rules — then build a Fabric normalization layer that applies those definitions to ERP exports from each portfolio company. This can deliver board-ready comparability in 2-4 weeks without ERP consolidation.

Which working capital KPIs matter most for PE exit readiness?

DSO and DPO trends over a 24-month period, cash conversion cycle, inventory turns (for product companies), AR aging quality, and covenant-specific ratios. Quality of earnings processes will scrutinize working capital normalization, so auditability of the metric calculation is as important as the number itself.

Working Capital Benchmarking Across PE Portfolios

The Comparability Problem in PE Portfolios

Most PE operating teams discover the working capital comparability problem when they first try to produce a portfolio-level DSO trend for a board or lender meeting. The numbers from each company come back incompatible: one uses trailing 30-day revenue as the denominator, another uses trailing 90-day, a third uses annualized prior-year revenue. The resulting DSO figures can differ by 8–15 days for the same business, purely from definition variance.

Inventory turns, DPO, and cash conversion cycle have equivalent definition drift problems. By the time the operating partner reconciles the numbers manually, the discussion has shifted from performance to methodology; board time is consumed defending the measurement rather than acting on it.

The underlying issue is not the ERP systems or the quality of the data within them. It is the absence of a shared definition layer that sits above all of them.

Normalization Model: Three Components

A working capital normalization model that works across a multi-ERP portfolio requires three components, in this order:

Standardized KPI contract: approved definitions for DSO, DPO, CCC, inventory turns, and any covenant-specific ratios. Each definition must specify the exact denominator (trailing 90-day revenue, for example), the aging bucket cutoffs, and how to handle partial-month periods, acquisitions, and currency translations. This document should be signed off by the operating partner and each portfolio company CFO before data work begins.
Standardized aging bucket and posting calendar alignment: each ERP has different default aging bucket boundaries and month-end posting cutoff logic. The normalization layer must translate each company's ERP aging data into a common bucket structure (0–30, 31–60, 61–90, 90+) and apply a consistent month-end cutoff date for comparability.
Entity-level mapping to a common data model: each portfolio company's GL, AR, and AP data is mapped to a common chart-of-account structure in the shared Fabric Lakehouse, with entity-specific overrides documented for unusual commercial terms, intercompany transactions, or currency effects.

Governance Model

Comparability degrades quickly without governance. Three practices maintain it:

Central KPI definitions approved by a finance transformation or operating office: definitions should not change at the portfolio company level without a formal change request. When a company CFO proposes a change (because their commercial terms genuinely require it), the change goes through a governance review, the impact on comparability is quantified, and if approved, a footnote is added to the portfolio-level reporting for the period of transition.
Automated validation checks before lender and board packs: before each board or lender submission, the data pipeline runs a set of validation checks: aging bucket coverage rates, AR and AP balance reconciliation to ERP trial balance, and metric drift alerts if a KPI moves more than 15% month-over-month without a flagged exception (acquisition, large one-time transaction, etc.).
Exception workflows for unusual commercial terms: portfolio companies with extended payment terms, consignment arrangements, or factoring programs need exception handling in the normalization layer. These should be documented and coded into the data model, not handled by manual adjustments in the dashboard.

Outcomes Operating Partners Should Expect

When normalization is done correctly, PE operating teams gain three categories of value:

Actionable comparability: the ability to rank portfolio companies on working capital performance on a like-for-like basis, identify which entities are outliers, and direct operating resources toward the highest-impact improvement opportunities. The ranking is defensible to both internal and external stakeholders.
Exit readiness: normalized, auditable working capital metrics significantly reduce the friction in quality of earnings processes. Buyers and their advisors can verify the metrics directly from the data model rather than through manual reconciliation. This compresses the QoE timeline and reduces the risk of metric-related purchase price adjustments.
Lender confidence: covenant reporting generated from the same certified model that powers board reporting, with automated checks that the covenant ratios use the definitions specified in the credit agreement. Audit trail from covenant ratio to source AR or AP aging record is available on demand.