Perspective
Serialisation Runs on the Stock Record, So We Put the Record Right First
A serial number turns every unrecorded stock movement into a visible exception. We close the gaps in the stock record first, so serialisation is built on figures your team believes.
4-minute read5 November 2026
In Brief
- Serialisation gives a company two accounts of its inventory, one by batch and one by pack. Movements that even out by batch at month end stay visible as exceptions at pack level.
- We see four actions: list the movements that change a pack count, name an owner for each, reconcile the batch master, and let the ERP decide what status a pack holds.
- Thinqmesh finds where stock moves without being recorded and closes those gaps before go-live, then brings a first module live on your system within four weeks.
Serialisation programmes in pharmaceutical companies usually begin on the packaging line, with printers, cameras and a repository for the numbers. That is understandable, because the line is where the obligation is most visible. The harder part sits in the stock record. A serial number gives each pack an identity, and from then on you hold two accounts of your inventory, one by batch and quantity and one by pack. We correct the first account before the second is built on it.
An Approximate Stock Record Shows Up One Exception at a Time
India's rule for the brands listed in Schedule H2 asks for a bar code or QR code carrying, among other things, the batch number and the dates of manufacture and expiry. The United States asks for tracing at the level of the package, and FDA has extended its timetable after trading partners reported difficulty resolving missing or erroneous data.
Wherever the data is wrong, it began somewhere, and an internal record that disagrees with the warehouse is the obvious place to look. A sample drawn from a sealed case, a return waiting in a corner or a conversion held in one person's head each leave the batch total looking right and the pack account wrong. Each one then costs a manual investigation, repeated at every release.
Four Actions That Give Serialisation a Record to Stand On
Each makes the stock figure more dependable, and each is useful in your current system too, so the work holds its value from the first day.
List the Movements That Change a Pack Count
Samples, returns, rework and write-offs each alter the packs in a case without touching the batch total.
A quality sample drawn from a sealed shipper, a return from a stockist, a rework and a destruction all change which packs exist. We trace each on your floor with the people who handle it, and compare that map with the adjustments posted over the past year. The adjustments point to the movements no one records.
Get started
- List the ten items adjusted most often over the past year.
- Walk each from receipt to dispatch with the person who handles it.
Name an Owner for Every Movement
Each kind of movement gets one role that records it, close to the moment it happens.
Where recording in the moment is impractical, we change the work with scanning, a simpler document or a different layout, so that the easiest way of doing the job is the way that records it. That is what makes the change last once the system is live.
Get started
- Write down who records each sample, return and rework today.
- Mark any movement that no one records at all.
Reconcile the Batch Master
Printed batch and expiry values should come from a master that has been settled with the people who use it.
Units, conversions, item codes and locations shape the configuration, so we agree them early and merge duplicates, with a clear decision on whose history survives. A batch master that finance, quality and the warehouse all read the same way gives each code on a pack a single source.
Get started
- Pick one product and list every code and batch format it carries.
- Check that quality, finance and the warehouse agree on each.
Let the ERP Decide What Status a Pack Holds
One system holds the status of a pack, and every other system reads it.
When a repository beside the ERP holds its own view of status, two accounts of the same stock must be kept in step by hand. We connect the systems so the ERP or warehouse system decides, and the others follow. Repeated counts of a sample of items then show whether the process works.
Get started
- Write down which system today decides whether a pack is saleable.
- Count twenty items twice, a fortnight apart, and trace each variance.
How Thinqmesh Solves This
We begin on your floor with your stock report, and the first module runs beside the systems you already use, so you can judge it on real work.
- Week 1
Where Stock Moves Without a Record
We stand up the platform with data representative of your business and walk the floor to find where stock goes unrecorded.
- Week 2
Your Rules, and an Owner for Every Movement
Approvals, access and costing rules are configured the way you operate, and each type of movement gets a role that records it.
- Week 3
Your Records Moved Across and Checked
We connect the systems you rely on and migrate your records. If your current system cannot export its data, we tell you first, and extraction gets its own date.
- Week 4
Real Work Running on It
Stock accuracy or job costing goes live on your own system, and further modules are agreed one at a time, in writing.
You finish with one job running on your own system, on stock figures your team trusts, with the places stock went missing found and closed.
