Branditify

Lab Management System

A container on the bench is not a sample yet.

One sample, from the moment it arrives to the moment its report is released — the accession that makes it workable, the checks assigned to it, the results recorded against it, and the review that has to happen before anything leaves the building.

Where lab management ends
SM-2048Water qualitySite A · outflowRack B · Position 07
Identity
Passed
Request
Passed
Condition
Passed
Quantity
Passed
  1. Check ABench 1Recorded
  2. Check BBench 1Recorded
  3. Check CBench 3Recorded
Accessioned
SA-2048
Results
3 of 3
Review
Pending
Release
Blocked

NextSend it to whoever is authorised to review it

Illustrative interface · sample dataAccession is earned, not delivered.

Results recorded. Accessioned. Not released. Next: Send it to whoever is authorised to review it.

Somebody has put a container on the bench. Nobody has yet established that it is the right sample, that anyone asked for it, or that there is enough of it to work with.

All four checks pass, so the sample takes an accession reference and becomes something the laboratory can work on. A container turned into a record, because somebody decided it had.

One sample, three defined checks, each with a bench that owns it. The worklist is the queue the laboratory works through — it is not a scientific decision and it never was.

One result in, two benches still working. Nothing is complete because somebody said so — it is complete when every check on the list has a result against it.

Every check has a result and nothing can leave the building. Recording what was found and being allowed to issue it are two different things.

A reviewer with the authority to issue it has done so, and their name is on the release. Six months from now, this is the state that answers what was done and who signed it out.

Receipt and accession

Arriving and being workable are two different states.

A courier can deliver a container. Only the laboratory can decide it is a sample, and it decides on four things.

What physically turned upOne container, labelled, left at the receiving bench

  1. IdentityIs the label legible and does it match the request?
  2. RequestIs there a request saying what this sample is for?
  3. ConditionDid the container arrive intact and within its holding conditions?
  4. QuantityIs there enough for the work being asked for?
  1. Received0 of 4 checks passedCannot be accessioned
  2. Accessioned4 of 4 checks passedCan be accessioned
Sample
SM-2048
The reference every result, review and report hangs off
Accession
SA-2048
Issued at accession, not at receipt — it is what says the lab took it on
Type
Water quality
What kind of sample it is, in the laboratory’s own vocabulary
Source
Site A · outflow
Where it came from, at whatever depth the operation records
Storage
Rack B · Position 07
Where it physically sits, so somebody can go and find it
Received by
Receiving bench
Who took it in, which is the first entry in its history

Nothing here promises a turnaround time. When work will be finished depends on the laboratory, the method and the queue in front of it — those are facts the operation supplies, and no clock starts running because a page said so.

Most laboratory disputes are about a sample nobody should have accepted in the first place.

What is lab management software?

Lab management software runs the operational side of a laboratory: the sample arriving, the acceptance checks that turn it into a workable record, the accession reference it takes, the tests or checks assigned to it, the bench worklist those create, the results recorded against each one, the review that has to happen before anything is issued, the release itself, and the history left behind. It is the record of a sample, not a judgement about what the result means.

What is a laboratory management system, and what does it manage?

Samples and the requests behind them, acceptance and accessioning, storage references, the work assigned to each sample, the bench worklists that work creates, results recorded against the work, review and release by whoever the laboratory authorises, reports issued, the full history of every one of those steps, exceptions that need a person, and who in the laboratory is allowed to see or change each part of it.

What is sample accessioning?

Accessioning is the point where a laboratory formally takes a sample on: it is checked against what was requested, given a unique accession reference, and registered with its type, source, condition and storage location. Before accession a container is just something on the bench. After it, there is a record the laboratory is accountable for — which is why it is a decision somebody makes rather than something that happens on arrival.

Sample and worklist

One sample becomes a list of work with owners.

The checks a sample needs, the bench each belongs to, and what each one is waiting on. An operational queue, not a scientific instruction.

Down one sampleWhat is outstanding on SM-2048?

  1. Check ABench 1
  2. Check BBench 1
  3. Check CBench 3

SM-2048

Across one benchWhat is outstanding on Bench 1?

Bench 1

  1. SM-2048Check A
  2. SM-2048Check B
  3. SM-2051Check A
  4. SM-2066Check B

Bench 3

  1. SM-2048Check C
  2. SM-2039Check C

A shared spreadsheet gives you the left-hand column and a conversation. Everything that goes missing later happens in the right-hand one.

How work finds a bench

Most laboratories route on something the sample already carries — its type, the method required, the section that owns that method, or who is qualified to run it. That routing is configured against your own sections and your own people. There is no default laboratory structure assumed here, and nothing decides that a particular instrument should run a particular sample.

Assigning work is a decision the laboratory makes and the system records. Nothing here schedules a method, selects an instrument, calculates capacity or predicts when a bench will be free.

A sample that is “in the lab” is not a plan. Three checks with three owners is a plan.

Can tests or work be assigned to a sample?

Yes — that is the middle of the record. A sample carries the checks it needs, each with the section or bench that owns it and its own state, so a supervisor can see what is outstanding on a sample and a bench can see what is outstanding on it. What the system does not do is decide which method is scientifically appropriate; that is defined by the laboratory and configured, not inferred.

Can laboratory instruments connect to the system?

Where a real interface exists, yes, and it is scoped rather than assumed. Analysers typically expose a serial or file interface, an ASTM or HL7 message, a vendor API, or middleware sitting between the instrument and the system — and each analyser family is its own piece of interface engineering rather than a setting. We start by identifying the instrument, what it can actually emit, and what somebody has to do at the bench, before anything is promised.

Can lab management software support barcodes and labels?

Sample identifiers, barcodes, QR labels and rack or position references are all normal parts of a laboratory record, and the system holds them. Printing them on your label stock and reading them with your scanners depends on the hardware you actually run, so that part is scoped around those devices. Branditify does not supply laboratory hardware and no printer or scanner is claimed to work until it has been identified.

Review and release

Recorded is not released.

Every check has a result and nothing can leave the building. Information and authority are two different things to be missing.

Work
Three of three recorded
Counted off the worklist, never declared complete
Results
Held against the sample
Each one attached to the check it belongs to

A report is releasedOn every result and an authorised review

Review
Pending
Whoever the laboratory authorises to review this kind of work
Release
Blocked
It cannot be issued while the review is outstanding
Report
Not issued
RP-2048 exists once release happens, and not before
  1. Results recorded3 of 3 recorded · review pendingCannot be released
  2. Released3 of 3 recorded · review completeReleased as RP-2048

Who is allowed to release is your decision

Recording a result, reviewing it and issuing it are separate permissions, because in most laboratories they belong to different people — and in some, to different people depending on the method. That routing is built from how your laboratory actually delegates authority, with the reviewer’s name and the state of the record at the moment they agreed kept on the sample.

What the system does not decide

It does not interpret a result, diagnose anything, decide scientifically whether something passes, or stand in for a qualified reviewer. It records what was found, holds it against the sample and the check it came from, and routes it through the review the laboratory defines. Where a laboratory has a deterministic rule of its own — a specification, a limit, a method’s own acceptance criteria — that rule can be configured and applied, and it remains the laboratory’s rule rather than the software’s opinion.

No result on this page is real and none is interpreted. The states shown are the operational states of a record — recorded, reviewed, released — and nothing here validates that a measurement is correct.

A laboratory is not judged on the measurement. It is judged on what left the building and who let it.

Can results be recorded and reviewed before release?

Yes, and separating those two is the point. Results are recorded against the check they came from, the sample carries how many of its checks are complete, and release stays blocked until an authorised reviewer has signed it off. Because completeness is counted from the worklist rather than set by hand, a report cannot be issued for a sample that still has work outstanding.

Does a lab management system automatically interpret results?

No, and it is worth being plain about it. It records the result, keeps it attached to the sample and the check, and moves it through review and release. It does not decide what a value means, diagnose anything, or certify that a measurement is right. Where a laboratory has its own specification or acceptance criteria, those can be configured and applied as the laboratory’s rule — which is a very different thing from software forming a scientific opinion.

Does LIMS software make a laboratory compliant or accredited?

No. Accreditation is granted on a laboratory’s own quality system — its methods, its people, its controls and its records — not on the software it runs. What good software does is make that system easier to operate and easier to evidence: consistent workflow, review and release that cannot be skipped, and a history that can be shown. Which requirements apply to you, and what has to be demonstrated, is established during scope rather than claimed on a page.

Sample history

Every sample carries its own account of itself.

Not an audit report. The states one sample went through, in order, with who moved it and what reference it took.

Sample historySM-2048SA-2048

  1. ReceivedContainer logged at the receiving benchReceiving
  2. AccessionedFour acceptance checks passed · SA-2048 issuedReceiving
  3. Work assignedThree checks created against the sampleSection lead
  4. ProcessingBench 1 and Bench 3 working their own listsBench
  5. Results recordedEach result attached to the check it came fromBench
  6. ReviewedSigned off by somebody authorised for this workReviewer
  7. ReleasedRP-2048 issued · nothing left outstandingReviewer

On the phrase “chain of custody”

It is used precisely in some laboratories and loosely in others, and in the strict sense it carries evidentiary requirements about handover, sealing and documented possession that belong to a laboratory’s own procedure rather than to software. What this record holds is a sample history: the states a sample went through, in order, with who moved it and when. Where a formal custody procedure is required, it is scoped against that procedure instead of being implied by the existence of a log.

How far back history goes, what is retained and for how long are decisions with real storage and policy consequences. They are settled with the laboratory rather than defaulted, and nothing here claims a retention period nobody agreed.

Six months later, nobody remembers the sample. The record is the only thing that does.

Can lab software track sample history?

Yes, and it is usually the reason a laboratory replaces its spreadsheets. Every state a sample moved through is kept in order against the sample itself — received, accessioned, assigned, worked, recorded, reviewed, released — with who did it and which references were issued. How much depth that history needs, and how long it is kept, is scoped from what the laboratory actually has to be able to show.

What needs a person now

Four samples that are not going to move by themselves.

Not a dashboard. Four records stuck for four different reasons, and what each one is waiting on.

  1. SM-2071Label unreadable, and no request matches itIt cannot be accessioned, so no work can be created against itGo back to whoever submitted itNeeds a decision
  2. SM-2066Arrived outside its holding conditionsAccepting it would put a result on the record that the method cannot supportDecide: reject, or accept with the condition notedNeeds a decision
  3. SM-2059Two of three results in, third bench blocked on materialThe sample cannot complete, so it cannot reach reviewChase the material, or reassign the checkNeeds a decision
  4. SM-2044Complete and waiting on an authorised reviewerThe report cannot be released by anybody currently on shiftHeld — waiting on a reviewerWatching

And one thing this list will not tell you

None of these are predicted. Each is either a state somebody recorded or a condition the record can work out for itself — a sample that failed an acceptance check, work with no result against it, a completed sample with no review. There is no risk score here, nothing is escalated automatically, and no judgement is made about the science.

A list where every sample is urgent is a list nobody opens. Three of these need a decision today and one is simply being watched, which is what the difference is for.

A laboratory system that only shows the samples going well is a system that gets checked once.

One sample, six questions

Everyone asks about SM-2048. Only one of them is asking about the laboratory work.

Six systems, one sample, and this page is the one holding the bench.

What is happening to this sample and its laboratory work?

Lab Management SystemThis page

Receipt, acceptance, accession, the work assigned to it, the bench that owns each check, the results recorded, the review, the release and the history. This page.

SM-2048

  1. Who is the patient, and what happened at the visit?

    Clinic & Practice Management

    The patient, the appointment, the consultation and the practice. A clinic system may raise a request that becomes a sample; it does not run the bench, the review or the release.

    Clinic Management System
  2. What reagent and consumable stock exists across the business?

    Inventory

    What is held, what is left and where. Inventory answers how many units of a reagent exist; it does not answer whether this sample’s assigned work has what it needs.

    Inventory Management System
  3. Who do we buy the reagent from, and did it arrive?

    Vendor & Procurement

    The supplier, the request, the quote, the approval, the purchase order and the receipt. A low reagent level may eventually start that process; it does not make the two one system.

    Vendor & Procurement Management
  4. What may the client see for themselves?

    Client portal

    A controlled outside view of a request, its state and the report once it is issued. It is a surface onto some of this record, decided deliberately, not the record itself.

    Client Portal
  5. How is the laboratory doing across every source?

    Dashboards

    Reporting that spans more than the bench, and usually more than one system. Operational views inside this record answer what is outstanding now; that is a different question from how the laboratory is performing.

    Dashboards

A lab management system is not a clinic system with a sample field bolted on, and it is not an inventory system wearing laboratory labels. It is the operating record of a sample.

And the terminology everybody argues about

LIMS stands for Laboratory Information Management System and grew up in research, testing and industrial laboratories, where sample management and traceability are the centre of the work. LIS grew up in clinical diagnostics, where the centre is the patient order, the result report and the connection to hospital systems. Most vendors and most buyers now use the terms loosely and many products cover both. This page owns the broad sample-centred capability under whichever name a laboratory searches for it, and the specifically clinical requirements — patient identifiers, order interfaces, hospital system integration — are scoped against the laboratory that actually has them.

Every one of these is a real system with a real owner. A laboratory page that answers all six is a page nobody will trust with any of them.

Lab Management System vs LIMS — is there a difference?

In practice, less than the vocabulary suggests. LIMS is the established acronym for Laboratory Information Management System, and Lab Management System is what most buyers actually type. Both describe software built around the sample: receipt, accession, work, results, review, release and history. We use the plainer term on this page and answer to the acronym, because a laboratory searching for one of them means the other.

Lab management vs clinic management — which do we need?

They answer different questions about the same day. A clinic system owns the patient, the appointment and the consultation. A laboratory system owns the sample, the work done to it, the result and the release. A clinical laboratory naturally references a patient or an order — that reference does not make it a clinic system, and running both usually means a request raised in one becomes a sample record in the other rather than being retyped.

Can lab software connect to inventory and procurement?

That is the shape worth building. The laboratory record knows this sample’s work needs a unit of a reagent; the stock system knows how many units exist; the procurement record knows who they are bought from and whether the last order arrived. Each connection is scoped against what the other system can accept, and this record deliberately keeps neither a second stock ledger nor a second set of supplier terms.

Can clients access reports through a portal?

Where a portal is in scope, yes — a controlled view of a request, its state and the report once it has been released. It is a decision rather than a default, because giving outsiders a login changes what every internal screen is allowed to contain, and because nothing internal to the bench should become visible by accident.

What it meets · what changes size

The parts every laboratory build gets, and the parts that are a decision.

Some of this is the system. The rest is scoped from how the laboratory actually runs.

  1. Samples and requestsIn every buildWhat arrived, who asked for it, what it is and where it came from — at whatever depth the laboratory records.
  2. Acceptance and accessioningIn every buildThe checks that decide whether a container becomes a sample, and the accession reference that follows when they pass.
  3. Storage referencesIn every buildRack, position, or whatever the laboratory calls the place somebody goes to find it.
  4. Work and test assignmentIn every buildThe checks a sample needs, the section or bench that owns each, and the state each is in.
  5. Bench worklistsIn every buildWhat is outstanding, seen from the bench rather than from the sample — the same records, the other way round.
  6. Result recordingIn every buildResults held against the check they came from, by whoever recorded them, with the sample carrying how many are complete.
  7. Review and releaseIn every buildWhoever the laboratory authorises, with release blocked until every result is in and the review is done.
  8. Sample historyIn every buildEvery state the sample went through, in order, with who moved it — which is what answers questions months later.
  9. Roles and accessIn every buildWho may accession, who may assign, who may record, who may review, who may release and who may only look.
  10. Barcode and label workflowsScoped per projectThe identifiers are core; printing them on your label stock and reading them with your scanners is scoped around the hardware you actually run.
  11. Instrument interfacesScoped per projectEach analyser family is its own interface — a serial or file feed, an ASTM or HL7 message, a vendor API, sometimes middleware in between — and it is engineered per instrument rather than switched on.
  12. Clinical and hospital connectionsScoped per projectPatient or order references, and connections to a clinic, hospital or external system, change what the record holds and who may see it. They are settled before the model is built.
  13. Quality control workflowsScoped per projectControls, calibrations, repeats and the rules a laboratory applies to its own runs are real work and specific to the methods being run.
  14. Batches, lots, expiry and retentionScoped per projectWhere a laboratory needs reagent lots tied to results, or defined retention of samples and records, that shapes the model rather than sitting on top of it.
  15. Client portal accessScoped per projectLetting a client see a request or download a released report is a deliberate decision about what outsiders may read.
  16. Multi-site laboratoriesScoped per projectSeveral laboratories sharing methods, people or samples changes accession, routing and review in different directions.
  1. How many samples arrive on a normal dayTwenty is a list. Four hundred is a receiving model, and it changes the first screen anybody sees.
  2. How many different kinds of work you runOne method is a form. Forty methods with their own fields, units and reviewers is the single biggest lever in the build.
  3. Who is allowed to review and releaseIf that depends on the method or the section, the authority model is the design rather than a permission list.
  4. Whether instruments have to feed itInterfaces are engineered per analyser family, so the honest question is which instruments, not how many.
  5. What the laboratory has to be able to showRetention, history depth and evidence requirements come from the laboratory’s own obligations, and they shape storage as much as screens.
  6. Which systems it has to meetA clinic system, a stock system, a procurement record, a client portal — each connection is real work and worth doing deliberately.

What we check before anything is quoted

How many samples arrive and in what shape, how many kinds of work the laboratory runs, who may review and release and whether that varies by method, which instruments genuinely have to feed the system, what the laboratory has to be able to show and for how long, and which existing systems must keep working. That conversation shapes the build far more than a feature list does.

What determines the scope of a lab management project?

Mostly four things: how many samples arrive, how many different kinds of work the laboratory runs, whether review and release authority varies by method, and which instruments have to feed the system. Feature lists predict a laboratory build badly; those four predict it well.

Who owns the system and the data after a custom build?

You do. The code, the database and the hosting account are yours, and every sample, result, review and report in it is yours. There is no per-sample charge, nothing meters how much work you run through it, and nothing stops working if a relationship ends.

What should a laboratory digitise first?

Samples and acceptance, then work assignment. A clean record of what arrived and whether it was accepted is what everything else hangs off, and work with a named owner is what makes anything else answerable. Digitising results or reports before those exist just moves the confusion into a nicer screen.

Going live

Getting a laboratory onto it.

The hard part is not the software. It is the first fortnight nobody keeps a parallel register “just in case”.

  1. What exists nowA sample register, worklist spreadsheets, an old LIMS nobody can get data out of, instrument exports in a shared folder, and reports as documents. We look at what can genuinely be exported before promising anything.
  2. One section firstReal samples, one section, one fortnight. It surfaces what a schema never does — three names for the same method, samples accepted that should not have been, results recorded against the wrong check.
  3. Map the four recordsSample, work, result, report. Everything else hangs off those, so getting them right is the whole migration.
  4. CleanDuplicate sample types, methods renamed over the years, sources recorded five ways. Decided with you, not silently.
  5. Import and checkLoaded, then walked through against the samples the laboratory believes are open today. Open samples and where they physically are must agree first.
  6. ReadyRoles set, the review and release routes tested with the people who actually hold that authority, and the two screens a bench will live in worked through with them.

Before it is the system of record

  1. Open samples agree with the racksEvery sample the system calls open is one somebody can physically go and find.
  2. A bench can record a result unaidedWithout asking anyone how, on the screen they will actually use.
  3. Nothing can be released without reviewTry it. The block is the point of the system, and it has to hold on day one.

SM-2048The system of record

Only then does the register stop being the source of truth. Running both for a fortnight is normal and worth planning for.

Nobody can promise years of old results, raw instrument files and scanned reports come across intact. What matters at go-live is that today’s picture is right — which samples are open, where they are and what is outstanding on each — and that the history worth keeping is identified rather than assumed.

A laboratory system goes live on the day somebody stops writing the accession number on a whiteboard, and not before.

Can old sample and result records migrate from spreadsheets or legacy software?

Usually, and the first job is finding out what the current tools can export rather than assuming. Samples, methods and open work normally move cleanly once duplicates have been decided. Years of raw instrument files and scanned reports depend entirely on what shape they are in and whether that history is worth the work — which is a decision made with you rather than for you.

Custom lab software, or an off-the-shelf LIMS?

Off-the-shelf is the right answer more often than agencies admit, and there are really three options rather than two. A packaged LIMS suits a laboratory whose workflow is standard. A configurable platform — a commercial product you build your own workflows on — covers a great deal of the middle ground without a build. A custom system earns its place when the last part will not configure: an unusual sample or review model, authority that varies by method, instruments nobody supports, or several existing systems that have to stay in step rather than be replaced.

Does every laboratory need custom software?

No. A small laboratory that needs sample tracking, results capture and a history it can show is well served by an established product, and commissioning a build would mean paying for complexity it will never use. Very few laboratories are genuinely unique — most differ in the type, number and flow of samples rather than in the shape of the work. Custom becomes the right answer when real workflow or integration gaps justify it, and that is a judgement worth making before a build starts rather than halfway through one.

Straight answers

What laboratory managers ask before they commit.

We run the lab on a sample register and spreadsheets. Is that actually a problem?
It works until somebody is away, or until a client asks what happened to a sample from March and the answer is in a file nobody can find. A register holds today well and holds very little afterwards. If it is one section and steady work, that is fine. If results are being re-entered, or samples are being looked for, that is the thing failing.
What happens when a sample is received?
It is checked before it is accepted — that the label is legible and matches a request, that somebody actually asked for the work, that the container arrived intact and within its holding conditions, and that there is enough of it. Only when all four pass does it take an accession reference and become a record the laboratory can assign work to. Before that it is a container on a bench.
Can the system reject a sample?
It records that the laboratory rejected it and why, which is not the same thing. Acceptance is a judgement a person makes against the laboratory’s own criteria; the record holds the decision, who made it and what happened next — returned, recollected, or accepted with the condition noted. Nothing is rejected automatically.
Can a sample carry a patient or client order reference?
Yes, where the laboratory needs one. A clinical laboratory naturally references a patient or an order, and a commercial testing laboratory references a client and a job. What stays true either way is that this record is built around the sample — the patient and the practice belong to a clinic system, and how much of that reference lives here is decided during scope rather than assumed.
Can we run several sections or several laboratories on it?
Yes, and it is worth naming early because it changes the shape rather than the size. Sections with their own methods, reviewers and worklists are normal. Several laboratories sharing samples, people or methods is a different model again, and which one you need is a question about how the organisation is run rather than about the software.
Does it produce the report?
It holds what a report is made of — the sample, its work, the results, the review and the release — and it can produce a report from that where report generation is in scope. What the report has to look like, what it must contain and whose format it follows is specific enough that it is designed with you rather than assumed, and nothing is issued before the review is done.
Can it handle quality control runs and calibrations?
Where quality control is in scope, yes — controls, calibrations and repeats recorded alongside the work they belong to. The rules a laboratory applies to its own runs are specific to its methods, so they are configured rather than shipped, and the system applies the laboratory’s rule instead of forming a view of its own.
Can we see how long samples are taking?
You can see what is outstanding and where it is sitting, derived from the states samples are actually in — what is awaiting accession, what is on each bench, what is complete and waiting on review. What you will not find on this page is a turnaround-time promise or a benchmark, because those belong to your methods and your queue rather than to software.
Who can release a report?
Whoever you decide, and in most laboratories that is not the person who recorded the result. Recording, reviewing and releasing are separate permissions, and where authority depends on the method or the section, that is built from how the laboratory actually delegates it. Reopening something already released is usually the permission that needs to belong to someone specific.
How long before the laboratory is running on it?
It depends on how many kinds of work you run, how complicated review and release authority is, and what state the current records are in — which is why no timeline appears on this page. What we can say is the sequence: samples and acceptance first, then work assignment, then results and review, with the open-samples check before any of it becomes the system of record.
What happens after it goes live?
The system is yours — code, database and hosting account. Most laboratories want changes in the first few months as benches discover what they actually need on the result screen, and that is either an agreed period of work or an ongoing arrangement. Neither is a licence.

Start a project

Tell us about the samples.

The useful first conversation is about what arrives and who may release it, not about features.

  • Roughly how many samples arrive on a normal day
  • How many different kinds of work the laboratory runs
  • What has to be true before a sample is accepted
  • Who is allowed to review and release, and whether that varies
  • Which instruments genuinely have to feed the system
  • Which systems have to keep working
ERP & Operations