The $12,000 freezer.

An operator scanned material for a GMP batch. The lot number looked right. The label said it was good until next month. What the label didn't say: the freezer had been alarming for three days. Nobody had checked whether the material was still viable.

The batch failed release testing. Investigation: two weeks. Cost: $12,000. Root cause: temperature excursion during storage. The material should have been quarantined. It wasn't, because nobody knew.

This is what inventory failure looks like. Not theft. Not fire. Just a gap between what you think you have and what you actually have.

The spreadsheet is lying to you

You've tried spreadsheets. Everyone has. Katie maintains one for Lab 4. Manufacturing has their own. There's a "Master Sheet" that nobody trusts because it's always two weeks behind.

The spreadsheet says you have 47 units of Taq Polymerase. Reality: 12 are expired. 8 are in a freezer that's been alarming. 3 were used last week and nobody logged it. The actual usable quantity is 24. But the spreadsheet doesn't know, and neither do you—until someone needs the material and can't find it.

Spreadsheets fail because they track what people remember to enter, not what actually happens. A scientist grabs a vial and forgets to log it. A box moves to a different shelf. A shipment sits in receiving for three days. The spreadsheet diverges from reality within hours.

Your ERP is no better. It knows what you ordered and what you paid for. It doesn't know what's actually on the shelf, what's expired, what's quarantined, or what's been partially consumed.

The costs are real

A stockout stops a $50,000/day manufacturing line while you wait for overnight shipping. An expired material in a GMP batch triggers a $5,000-15,000 deviation investigation. A failed audit observation around inventory control costs $100,000+ in remediation. Organizations carry 20-30% excess inventory as buffer against uncertainty they can't see.

This isn't about better record-keeping. It's about whether production runs on time, whether batches release without deviations, and whether auditors find problems you didn't know you had.

Container-level truth

Seal doesn't track SKUs. It tracks containers.

Not "Taq Polymerase, 47 units." This specific aliquot: LOT-2024-0892, 250µL, in Freezer ULT-03 on Shelf 2 in Box C, received January 15th, expires March 2025, 180µL remaining, last used March 12th by J. Smith for Batch B-2024-041.

When you need material, you don't get a count that might be wrong. You get a list of actual containers—where they are, how much remains, when they expire, whether they're released. Click to see the full history: when it arrived, who handled it, where it's been, what it's been used for.

When a lot is recalled, you don't dig through records for days. You query and know—in seconds—exactly which containers you have, where they are, and what batches already used them.

Enforcement, not tracking

Most inventory systems track status. Seal enforces it.

A quarantined container cannot be selected for any operation. An expired container doesn't appear in pick lists. A container released for R&D only cannot be selected for GMP batches. There's no checkbox to override. There's no "continue anyway." The control is architectural.

When an operator scans material for a batch, the system checks: Is this the right material? Is it released? Is it within expiry? Is the storage unit in range? If everything passes, the operator proceeds and quantity deducts automatically. If anything fails, the system blocks them with a clear explanation.

This is the difference between "we track inventory" and "we control inventory." Tracking tells you what happened after the fact. Control prevents the problem before it occurs.

Connected to operations

Standalone inventory systems create reconciliation nightmares. The batch record says you used 500mL. Inventory says you have 600mL left. You started with 1000mL. Where did the other 100mL go? Someone spends a day figuring it out. Or doesn't, and the discrepancy just... exists.

Seal integrates inventory with lab and manufacturing because they're the same platform. When a scientist runs an experiment, they scan materials—lot numbers record with the experiment, quantities deduct automatically. When manufacturing executes a batch step, operators scan materials—the batch record captures which lots were used, inventory reflects what was consumed.

Same data, captured once, at the point of use. No reconciliation meetings. No "inventory says X but batch records say Y."

The integration matters most when things go wrong. Friday afternoon: a supplier calls about a contamination issue with lot 2024-0892. How many batches used it? Which are still in quarantine? Which shipped? In a standalone system, that's a weekend of digging through records. In Seal, it's a query: every container of that lot, where they are, what batches consumed them, which products are affected. Thirty seconds. Recall scope defined before you leave for the day.

Setup in hours, not months

Traditional inventory implementations die in configuration. Defining 500 materials. Entering specifications. Setting up storage conditions. Re-typing from supplier datasheets. Months of work before you track a single container.

Drop your supplier documentation—datasheets, specs, safety data sheets. AI extracts material properties, storage conditions, handling requirements. A stack of datasheets becomes structured material definitions in hours. When shipments arrive, drop the CoA. AI extracts lot numbers, expiry dates, test results. Receiving becomes verification, not data entry.

Every AI extraction is reviewable. You see exactly what was extracted. Edit what's wrong, approve what's right. Nothing enters the system without human verification—which matters for GxP.

Getting Started

You don't need to barcode everything overnight. You don't need to count every freezer. You don't need a six-month implementation.

Start with one material class—your critical raw materials, or your expensive reagents, or whatever causes the most pain. Define those materials. Set up storage locations. Print labels for new receipts. Start tracking containers as they arrive.

Your existing inventory stays where it is. As you consume it, it leaves. As new material arrives, it enters the system properly. Within months, your critical materials are tracked at the container level. Expand from there.

The next time a freezer alarms at 2 AM, you'll know exactly which containers are affected. The next time a supplier issues a recall, you'll have the impact assessment before lunch. The next time an auditor asks about chain of custody, you'll show them—in seconds.

That's what visibility looks like. Start there.