Laboratory Network & Operating Model
Which laboratory does what: research, quality control, microbiology, stability, bioanalytical, central and contract laboratories, and the point-of-use testing that happens outside all of them. Testing scattered across laboratories with different quality systems produces results that are individually defensible and collectively inconsistent — and the network model is what decides where a result can be trusted to mean the same thing.
What an explainer is not
A topic explainer is SPEQ’s synthesis of what a practice involves, cited to the standards that govern it. It does not reproduce their text, and it does not determine which of them apply to your product or process.
[ POSITION IN THE FRAMEWORK ]
7 DIMENSIONS · 24 LINKSA testing network is only as resilient as its least transferable method: a laboratory holding a technique nobody else can run is a single point of failure that no supplier assessment will surface.
06 · QUALITY MATURITY — LABORATORY NETWORK & OPERATING MODEL, REACTIVE TO ADAPTIVE
Testing happens where it always has. Which laboratory can run which method is knowledge held by schedulers.
A capability matrix exists per laboratory, listing methods rather than demonstrated current competence in them.
Capability is mapped against demand and criticality, methods that exist in only one place are known, and contract laboratories are qualified per method rather than as organisations.
Transfer capability is maintained deliberately — a second site is kept current on critical methods rather than nominally approved.
The network is arranged so testing can move without a project, and where a method must stay in one place that is a recorded decision with mitigation.
SPEQ’s shared five-stage progression, labelled synthesis — not the FDA QMM rating scale. Where does your organization sit? Score your quality system →
07 · REGULATORY & EVIDENCE
GOVERNING STANDARDS · 5
Derived from the 5 standards SPEQ maps to this subject, across 5 regulatory bodies: EMA, ICH, USP, ISO, ILAC.
RECORDS & OBJECTIVE EVIDENCE
- The capability map by laboratory and method, with currency of each capability
- Methods available in only one laboratory, with the mitigation for each
- Contract laboratory qualification, per method rather than per organisation
- Method transfer records where testing moved between laboratories
- Demand and capacity data supporting the allocation of testing
COMMON INSPECTION FINDINGS
- A critical method that only one analyst in one laboratory can perform
- Contract laboratories qualified as organisations, with method scope never verified
- A capability matrix listing methods a laboratory has not run in years
- Testing moved between laboratories without a transfer exercise
- Capacity constraints resolved by moving work to laboratories not qualified for it
The same method in two laboratories is not the same test
A method transferred to a second laboratory runs on different instruments, with different analysts, different reference standards and a different local interpretation of the procedure. Transfer studies establish that the two produce equivalent results at the point of transfer; nothing guarantees they still do three years later unless something is watching.
Cross-laboratory comparison — the same sample tested in both, periodically — is the control that detects drift, and it is rarely run outside contract-laboratory oversight. Where it is run, it reliably finds differences: a bias that developed after an instrument replacement, a divergence in integration practice, a reference standard used past its intended life. Each is small and each moves results in one direction.
Accreditation is not a quality system, and is not a substitute
ISO/IEC 17025 accredits a laboratory’s competence for specific methods and specific matrices, which is why the accreditation schedule matters more than the certificate — accreditation is granted per method and per range. ILAC recognition is a property of the accreditation body, not of the laboratory, and the distinction is routinely blurred in supplier assessment.
Accreditation also does not establish GMP compliance. A laboratory can be accredited and not operating under a pharmaceutical quality system, which affects change control, deviation handling, data integrity and how out-of-specification results are investigated. For contract laboratories performing GMP testing, both questions have to be asked, because neither answer implies the other.
The testing that happens outside the laboratory
Point-of-use testing — in-process checks on the line, environmental monitoring plates read locally, rapid methods at the point of manufacture — sits outside the laboratory quality system and frequently outside anyone’s. The instrument may have no calibration owner, the analyst no method qualification, and the result no audit trail, while feeding a batch record.
This is where laboratory network design has its largest blind spot. Any test generating a GxP result belongs inside a defined quality system regardless of where it physically occurs, and the network model should enumerate those tests explicitly rather than defining the network as the set of rooms called laboratories.
SPEQ interpretation — decide where a decision is made, not just where a test is run
Network design conversations focus on where testing happens: consolidate or distribute, in-house or contract, central or local. The more consequential question is where the resulting decision is made — who dispositions an out-of-specification result, who decides an investigation has found root cause, who approves a method change.
A distributed testing network with centralised decision-making is coherent. A distributed network with distributed decision-making produces different answers to the same situation in different places, and that inconsistency is what an inspector finds by comparing two investigations of the same phenomenon. Deciding the decision rights explicitly, separately from the testing footprint, is what keeps a network consistent as it grows.
FREQUENTLY ASKED
Does a transferred method stay equivalent?
Only if something is watching. Transfer establishes equivalence at the point of transfer; different instruments, analysts, reference standards and local interpretations drift afterwards. Periodic cross-laboratory comparison of the same sample is the control, and where it is run it reliably finds bias from an instrument replacement, divergent integration practice, or an over-aged reference standard.
Does ISO/IEC 17025 accreditation establish GMP compliance?
No. It accredits competence for specific methods and matrices — which is why the schedule matters more than the certificate. A laboratory can be accredited without operating under a pharmaceutical quality system, which affects change control, deviation handling, data integrity and OOS investigation. For GMP testing both questions must be asked.
What about testing done outside the laboratory?
In-process checks, locally read environmental monitoring plates and rapid methods at the point of manufacture often sit outside any quality system while feeding a batch record — no calibration owner, no method qualification, no audit trail. Any test generating a GxP result belongs inside a defined quality system regardless of the room it happens in.
What matters more than where testing happens?
Where the resulting decision is made: who dispositions an OOS result, who decides an investigation found root cause, who approves a method change. A distributed network with centralised decision-making is coherent; distributed decision-making produces different answers to the same situation in different places.