One Portal Is Not the Whole Study.
A central lab portal shows you honestly what the central lab knows. It cannot show you the parts of your study that aren’t in the central lab. That gap is where reconciliation quietly lives.
What the portal shows you — and what it doesn’t
Every sponsor with a biospecimen-heavy study has, at some point, logged into a central laboratory portal and felt a wave of reassurance. Subject IDs, accession numbers, receipt timestamps, QC status, aliquot yields, results as they come in — the portal is well-built, the data is clean.
And then, a week later, the CRO shares a spreadsheet of samples that appear to be missing, and the numbers don’t match. The portal shows one thing. The spreadsheet shows another. The reconciliation begins.
This isn’t a failure of the portal. The portal is faithfully reporting what the central lab knows. The problem is that a modern biospecimen operation isn’t run inside the central lab. Everything before a specimen arrives at its doors — the visit that was supposed to happen, the tube drawn at the site, the shipment in transit, the aliquot routed to a specialty lab — is outside the portal’s field of view. Not because the portal is bad. Because it was never asked to see those things.
Eight sources of truth. The portal is one.
Any biospecimen-heavy clinical trial is, whether the sponsor articulates it or not, an operation running across roughly eight independent sources of truth. Each is honest about the slice it covers. None can see the others.
- The EDC knows what was supposed to happen — the protocol, the visit schedule, the specimen types expected at each timepoint.
- The site LIMS or eSource knows what was actually collected at the visit — the tubes that were drawn, the times, the collector’s initials.
- The central lab portal knows what arrived at the central facility, was accessioned, processed, aliquoted, tested, and reported.
- The PK/PD lab knows only the pharmacokinetic aliquots that were routed to it, and only after they arrived.
- The biomarker vendor knows only its own assay panel — the samples it received, the results it produced, nothing upstream.
- The logistics providers know what’s in transit, where, and whether a temperature excursion happened between pickup and delivery.
- The sponsor CTMS knows the protocol amendments, the site activations, the current version of the schedule — which the other seven may or may not have.
- The kit vendor knows what was shipped to which site, and what’s expiring next — but not what was consumed, what’s sitting idle at a slow-recruiting site, or which site is a week away from a stock-out.
The central lab portal is one of these. It is often the biggest of them — the one with the best UI, the cleanest data, the most polished executive view. It is never the whole picture. It cannot be. The physical operation the sponsor is paying for spans all eight, and no single vendor sits above all eight.
- The portal answers: is our lab OK?
- The study asks: is our specimen operation OK?
What the portal can’t tell you
The cleanest way to see the gap is to line up the questions the portal cheerfully answers next to the ones the study actually needs answered.
- Which samples arrived here, and when?
- What’s the accession status of each?
- What aliquots did we produce?
- Which results are ready, and which are pending?
- How is our lab’s turnaround trending?
- Was this subject’s C1D1 PK draw actually collected at the site — and is it on its way?
- Is the visit still within the collection window, or did the site miss it silently?
- Did the biomarker aliquot make it to the specialty vendor, or is it stuck between them?
- Does the EDC row for this visit match the tubes the site says they drew?
- Which sites are trending toward a kit stock-out next week — and which are sitting on expiring inventory nobody’s going to use?
Every question on the right requires stitching together at least two of the eight sources. A “missing” sample isn’t missing anywhere in particular — it could be absent from the EDC, present in the site’s notes, on a shipping manifest, or at the wrong lab entirely. The portal shows a clean empty row and has no way of knowing whether that row should be full.
The reconciliation you’re already paying for
Most sponsors don’t realize the portal isn’t the whole picture — until they notice that someone, or an entire team, is quietly stitching it together in the background. A program lead exports the central lab portal. A separate export comes from the EDC. Another from the PK vendor, another from biomarker. A coordinator merges them by hand, VLOOKUPs subject IDs, chases what doesn’t match, and waits — every week or two, on some cadence tied to a governance meeting.
That effort is the tell. If the portal really gave the whole picture, the reconciliation wouldn’t exist. The head-count exists precisely because the picture is partial — and the cost is buried in labor, not in a line item on the tooling budget, which is why it’s so easy to leave alone.
Kits make the shape of this most obvious. The central lab tells you, accurately, that they’ve shipped the kits — and that is the last honest number anyone has. How many were actually consumed? Which sites are sitting on kits about to expire? Which site enrolled twice as fast as forecast and is a week from a stock-out? None of that lives in the portal. Kit shipment is a vendor deliverable; kit consumption, expiry, and forecast are operational questions no single vendor was ever going to answer.
What the layer above the portal actually does — and where SpeciGen fits
None of this is an argument against central lab portals. Keep them — they’re the right instrument for reporting on the lab’s own work. What’s missing is the layer above them: one that reads all eight sources and understands the operation as one thing rather than eight overlapping fragments.
SpeciGen
Operational Intelligence for Biospecimen Management
That layer is what SpeciGen is built to be. It sits above the central lab portal, the EDC, the specialty labs, the logistics providers, the kit vendor, and the sponsor’s own systems — ingesting their operational signals continuously, encoding the protocol so the software knows what should have happened, applying reconciliation rules that live with the operations team, and surfacing what needs attention while there is still time to act on it.
Ingestion happens across whatever channels the vendors offer — APIs where available, system integrations where practical, automated file-based workflows where those are all a partner will send. What changes is not how the data arrives; it is that the data arrives into a layer built to reconcile it against the protocol, continuously, on the sponsor’s side of the wall.
- The central lab portal keeps its job — and the sponsor gains a view above it that answers the whole question.
- Cross-vendor gaps surface as they emerge, not at end-of-study reconciliation.
- Kits are tracked past the shipment. Consumption, expiry, and forecast become continuous signals rather than periodic emails.
- Every specimen carries a defensible, auditable operational history — assembled across all eight sources, from bench to database lock.