Most LIMS buyer’s guides assume a testing laboratory. A sample arrives, gets analyzed, a result is released, and the sample is eventually discarded. A biobank works the other way around: the sample is the product. It may stay in a freezer for decades, be split into dozens of aliquots, travel to research groups in other countries, and remain usable only as long as the biobank can show what the donor agreed to and how the material was handled before it was frozen.
That difference explains why so many biorepositories still run on spreadsheets, or on a general LIMS that was never set up for donors and consent. An internal audit at UT Southwestern found that five of its eight in-scope biorepositories kept their specimen inventory in protected Excel workbooks, and flagged the risk of losing track of specimens after staff turnover or during an emergency. When the University of Iowa moved its research biobanking onto one shared LIMS, it retired eight separate systems, several of them spreadsheets or desktop databases, according to the team’s 2025 write-up in the Journal of Clinical and Translational Science.
This guide covers what the standards actually ask of the software, the three families of products on the market, the few prices vendors publish, and the failure points that tend to appear after go-live. It is written for biobank managers, core facility directors and the IT teams that support them.
Why a biobank is not a testing lab
A testing-lab LIMS is built around the chain from sample to test to approved result. A biobank system is built around a different set of records: the donor or participant, the consent and its version, the lineage of every aliquot and derivative, the exact storage position, and the history of every request and shipment. James Harrison, a University of Virginia informatician interviewed by CAP TODAY in 2022, said that any LIMS, even one designed for biobanking, typically requires significant customization.
The gap shows up concretely when a general system is adapted. The developers of Baobab LIMS, built for a human biobank in South Africa, had to custom-build kit assembly, shipping and storage management modules on top of the Bika LIMS base before it fit biobank work, according to their own 2017 paper. That is a developer’s account, but it matches the customization point above.
Four capabilities separate a biobank system from a generic LIMS:
- Donor and consent as first-class records. Every aliquot should inherit the consent scope of its donor, including the consent version it was collected under. Iowa listed tracking the consent version as a core requirement, because it decides what a specimen can be used for.
- Lineage. Parent and child relationships across aliquots, pools and derivatives such as extracted DNA, so that a withdrawal or a quality issue can be traced to every descendant.
- Storage to the exact position. Site, room, freezer or tank, rack, box and slot, plus the movement history. NCI’s best practices state that the database must be able to identify each position in storage (NCI, 2016 edition).
- Distribution. Requests from researchers, approvals, material transfer agreements, shipments and chain of custody, often with invoicing attached.
If your facility needs only the third item, a freezer inventory tool may be enough. If it needs all four, a sample-centric LIMS with a few extra fields will hold you back. Our guide on when spreadsheets stop being enough covers the earlier decision of whether to buy anything at all.
What the standards ask of the software
No standard certifies software for biobanking. ISO 20387 accredits the biobank, and the CAP Biorepository Accreditation Program inspects the biorepository. But both, along with the ISBER and NCI best practices, translate into concrete requirements a system has to support. The table below maps the main ones to a test you can run during a vendor demo.
| Requirement | Where it comes from | What to test in a demo |
|---|---|---|
| Locate any material and its associated data at any time | ISO 20387:2018, clause 7.5 on traceability | Pick ten aliquots from your own export and ask for their current box position and full movement history |
| A defined system for tracking information and data, with integrity, security and backup | ISO 20387:2018, clause 7.10 | Ask where backups run, who can restore them, and when a restore was last tested |
| Link material and data to consent and use restrictions | ISO 20387:2018, clause 7.5.1; CAP checklist item on informed consent records | Withdraw consent for one donor and check that every derived aliquot is blocked from distribution |
| Unique identifiers and parent and child relationships | CAP Biorepository checklist, inventory items | Create aliquots, pool two of them, derive DNA, then open the lineage view |
| Chain of custody for every transfer | CAP Biorepository checklist | Ship a box to an external site and receive it back |
| Record of who entered or changed data | CAP checklist; 21 CFR 11.10(e) where FDA predicate rules apply | Edit a record, then pull the audit trail while logged in as a non-administrator |
| Pre-analytical annotation | SPREC version 3.0, developed by ISBER | Ask whether pre-analytical codes are structured fields or free text |
The ISO clause numbers come from the public self-assessment worksheet published by NATA, the Australian accreditation body, since the standard itself is sold by ISO. The CAP items come from the 2017 public edition of the Biorepository Checklist; the current edition is available to accredited biorepositories and for purchase, and wording may have changed.
Two legal frameworks sit on top of this. In the EU, the GDPR treats genetic and health data as special categories, requires the controller to be able to demonstrate consent, and gives the data subject the right to withdraw it at any time (Regulation 2016/679, Articles 7 and 9). In the US, the revised Common Rule allows broad consent for storage and secondary research use, but if a person was asked for broad consent and refused, an IRB cannot waive consent for that use (45 CFR 46.116). Both rules turn a refusal or a withdrawal into a record that the software has to enforce, not just store.
21 CFR Part 11 is often listed as a biobank requirement, but its scope is narrower. It applies to electronic records maintained under FDA predicate rules (21 CFR 11.1), and FDA has said since its 2003 scope and application guidance that it reads that scope narrowly. Our reading: a biorepository holding specimens and data for an FDA-regulated trial falls inside it, while an academic research collection usually does not. Our explainer on what 21 CFR Part 11 requires goes into the details.
ISO 20387 is changing: what that means if you buy now
ISO 20387 was first published in August 2018. ISO decided in December 2023 that it should be revised, and the second edition, ISO/FDIS 20387, was registered for formal approval on 22 June 2026, according to ISO’s page for the draft. It will replace the 2018 edition, and the title shifts from requirements “for biobanking” to requirements “for biobanks”. As of 9 October 2026, ISO gave no publication date.
Accreditation against the standard is already offered in several countries, for example by A2LA and ANAB in the United States, by DAkkS in Germany, and by UKAS, which announced its first UK biobank accreditation in March 2025 for the NHS Greater Glasgow and Clyde Biorepository (UKAS).
For a buyer, three points matter:
- A software claim is not an accreditation. CloudLIMS markets its biobanking product as a way to achieve ISO 20387:2018, and Agilent states that OpenLab Sync can support ISO 20387:2018 requirements. Both are vendor claims tied to the 2018 text. The accreditation body assesses your processes, records and competence, with the software as one tool among others.
- Ask for the mapping, not the badge. A clause-by-clause document that shows which feature supports which requirement is more useful during an assessment than a logo on a brochure.
- Put the revision in the contract. We have not seen the final text, so we cannot say which requirements will move or change. What you can control is who updates the mapping, and at what cost, once the second edition is published.
Two free references were also updated recently. ISBER published the fifth edition of its Best Practices for Repositories in December 2023, and NCI released the fourth edition of its Best Practices for Biospecimen Resources in April 2026, with an informatics section covering selection, validation and regulatory issues. Both are worth reading before you write a requirements document.
The three families of biobank software
Products marketed to biobanks fall into three families. The examples below are based on how each vendor positions its own product; the fit and limits are our assessment.
| Family | Examples | Where it fits | Where it struggles |
|---|---|---|---|
| Dedicated biobank systems | OpenSpecimen (Krishagni), MBioLIMS BioBanking (Modul-Bio), Limfinity Biobanking LIMS (Azenta), SoftBiobank (SCC Soft Computer) | Hospital, cohort and population biobanks where donors, consent and distribution are the core workflow | Facilities that also run regulated assays or testing workflows, which may need a separate LIMS or LIS and an interface between them |
| General LIMS with a biobanking configuration | LabVantage Biobanking, LabWare, Agilent OpenLab Sync for Biobanks, CloudLIMS, Matrix Gemini LIMS Biobank Manager (Instem), Sapio Biorepository Management | Biorepositories inside pharma, CROs or large institutions that want one platform for storage and testing | Biobank features are configured on a sample-centric core, so setup and validation take longer and the depth of consent handling varies |
| Sample inventory and freezer tools | Freezerworks, FreezerPro (Azenta), LabKey Sample Manager, Mosaic (Cenevo), SciSure | Research groups and core facilities whose main question is what is stored where | Donor-level consent and distribution management are thin or absent on the vendor pages we reviewed |
A few observations from reading the vendors’ own pages in October 2026:
- Consent tracking is not universal. It is stated explicitly by OpenSpecimen, Modul-Bio, Azenta for Limfinity, Agilent, CloudLIMS and Instem, and LabWare lists consent among the subject details users can manage. LabVantage offers a participant-centric data model but mentions consent only in a privacy-compliance context. We did not find consent tracking on the biobanking pages of Sapio, Freezerworks, FreezerPro, LabKey, Mosaic or SciSure. That does not prove the feature is missing, only that the vendor does not lead with it. Ask.
- Pre-analytical coding is rare. Only one vendor page we reviewed, Agilent’s, names SPREC.
- Names have moved. Agilent now markets SLIMS as OpenLab Sync; our Agilent SLIMS overview was written under the old name. Mosaic is now sold under the Cenevo brand, as covered in our Mosaic by Cenevo overview, and SciSure is the company formed when eLabNext and SciShield merged in January 2025.
What biobank software costs: the few published prices
Most vendors in this market quote on request. The table lists the published figures we found, in US dollars, as displayed on each vendor’s pricing page on 9 October 2026.
| Product and tier | Published price | Model |
|---|---|---|
| OpenSpecimen Starter, up to 5 users | 75,000 one time | Hosted on Krishagni Cloud only, plus annual support from 35,000 |
| OpenSpecimen Medium-Scale, up to 10 users | 100,000 one time | Hosted on Krishagni Cloud only, plus annual support from 35,000 |
| OpenSpecimen Institutional, unlimited users | 150,000 | On-premise or cloud; the page labels the price both one time and fixed yearly |
| CloudLIMS, in-house biobank, 3 users | 310 per user per month, billed annually (11,160 per year) | SaaS |
| CloudLIMS, third-party or commercial biobank, 5-user minimum | 285 per user per month (17,100 per year) | SaaS |
| LabKey Sample Manager Starter, 5 users included | 6,540 per year | Hosted subscription |
| LabKey LIMS Starter, 10 users included | 34,080 per year | Hosted subscription |
| Freezerworks | On request | Perpetual licence, optional support renewal at 20% of the software cost after the first year |
Annual totals for CloudLIMS are our arithmetic from the per-user prices. Sources: OpenSpecimen, CloudLIMS, LabKey, Freezerworks FAQ.
Two things stand out. First, OpenSpecimen’s code is open source under a BSD-style licence, yet its supported editions cost as much as a commercial system, and its pricing page lists indicative fees of 35,000 for integrations, with annual fees on top, and 40,000 for data migration. Open source changes who controls the code, not the cost of running it in production. Second, the licence is rarely the largest line. In a book chapter on biobanks in low- and middle-income countries, a Makerere University author reports one commercial LIMS purchased at 120,000 USD and stresses that many vendors have no local support (IntechOpen, 2021). The Iowa consolidation ran over several years and depended on staff time to transform legacy data. For broader context, see our LIMS pricing benchmark.
Where biobank software projects go wrong
Consent is the migration nobody budgets for
Moving freezer locations into a new system is tedious but mechanical. Moving consent is not. In a digital maturity survey of Italian biobanks, piloted in 2023 and published in 2025, about two thirds of the 37 respondents that agreed to publication reported having a biobank management system, yet 94.6% of the 37 still traced informed consent to biobanking for future use on paper forms only (Rossi et al., 2025). Backfilling consent means reading forms, mapping each version to a usage scope, and deciding what to do with ambiguous cases. That is expert work, and it rarely appears in a vendor’s migration quote.
Iowa’s experience points to a second trap: inconsistencies appeared because several people transformed data for upload in different ways. Assign one owner for transformation rules before the first import. Our guide to migrating between LIMS covers the mechanics.
The database is only as accurate as the freezer
The UK Human Tissue Authority reviewed 87 inspections of research establishments carried out between 2010 and 2015 and reported that, during some audits, samples could not be located or sample identifiers were duplicated (HTA, 2016). We found no peer-reviewed figure for the share of biobank samples that cannot be located. What the evidence supports is that software makes discrepancies visible, while the physical inventory still has to be checked: NCI recommends random quality control checks of what is actually in storage, and a system is only as reliable as the last time someone compared it with the freezer.
An inventory system is not an alarm system
In December 2023, temperatures rose in 16 cryogenic tanks at Karolinska Institutet, and about 34,400 biobank samples and 3,800 samples from animal models were lost. The internal report found several causes, including an alarm that was announced as reset but was not, and gaps in how alerts were delivered and acted on while service work was under way (Karolinska Institutet). Monitoring and escalation belong in a dedicated system with tested alerts. The inventory software’s job is to link an excursion to the specimens stored in the affected unit, so their quality can be annotated afterwards. CloudLIMS and Modul-Bio, for example, list integrations with several monitoring systems on their product pages. Equipment is also where biorepositories inspected under the CAP program most often fell short: in 88 inspections between 2017 and 2022, equipment accounted for 24% of deficiencies and information technology for 13% (Leelatian et al., 2026).
Check what the free or legacy edition actually contains
Two current examples show why the version matters as much as the product. In April 2025, Krishagni wrote on the OpenSpecimen forum that it intends to keep the Community Edition free and open source, and pointed users who want that to version 10.3. In June 2026, a user on Community Edition 12.2 reported being unable to add specimens because a workflow module was not installed, and was asked to email the company for access; the thread did not say whether that module is paid. Separately, Freezerworks asks existing customers to move to version 14.01.03 by 31 December 2026, is discontinuing its desktop client, and plans Freezerworks 2 for late 2026 as a self-hosted product rather than SaaS (Freezerworks). Neither point is a verdict on these products. Both are reasons to confirm in writing which version you will deploy and what it includes. Our guide to open-source LIMS covers the wider trade-offs.
Adoption is a governance problem
Iowa’s shared system now serves more than 380 users across 18 research groups, and hundreds of thousands of aliquots have been added since go-live, but the team still found it hard to convince new researchers to store their data there. Software cannot mandate its own use. If the institution does not require new collections to be registered in the system, spreadsheets will grow back around it.
Questions to put to every shortlisted vendor
- Show a consent withdrawal propagating to every derived aliquot and blocking it from distribution.
- Can one donor hold two consent versions with different scopes, and which one applies to each specimen?
- Which ISO 20387:2018 clauses does your documentation map, and who updates that mapping, at what cost, when the second edition is published?
- Are pre-analytical variables, such as SPREC codes, stored as structured fields?
- How is a temperature excursion linked to the specimens stored in the affected freezer or tank?
- What exactly does the migration quote cover, and who writes the rules for transforming our spreadsheets and paper consent forms?
- For community, legacy or entry tiers: which modules are required for core specimen workflows in the current release, and are they included?
Ask for each answer on your own anonymized data, not on a demo dataset. Our guide to running a LIMS proof of concept explains how to set pass and fail criteria before the trial starts.
The bottom line
Choose the family before you choose the product. A cohort or hospital biobank whose work revolves around donors, consent and distribution will usually be better served by a dedicated biobank system. A biorepository inside a testing or pharma organization may accept the extra configuration of a general LIMS to keep one platform. A research group that mainly needs to find tubes can stop at an inventory tool. In every case, the costs that decide the project are the ones outside the licence: turning paper consent into enforceable records, reconciling the database with the freezers, and getting researchers to use the system. With ISO 20387 in its final draft stage, also settle in the contract who keeps the compliance mapping current.
This site is editorially independent. No vendor named in this guide paid for its inclusion or reviewed this piece before publication.
FAQ
What is the difference between a biobank LIMS and a regular LIMS?
A regular LIMS is organized around samples, tests and released results, and usually assumes the sample is consumed or discarded. A biobank system is organized around the donor, the consent and its scope, the lineage of every aliquot and derivative, the exact storage position, and distribution to outside researchers. Some general LIMS platforms offer a biobanking configuration that adds these records, but it is configured rather than native, so budget more setup and validation work.
Can software make a biobank ISO 20387 compliant?
No. ISO 20387 accreditation is granted to the biobank and its processes by an accreditation body such as A2LA, ANAB, UKAS or DAkkS, not to a software product. Software can support specific requirements, such as locating any material and its associated data at any time or keeping data integrity and backups under control. Treat a vendor’s ISO 20387 claim as a statement that its features can support those clauses, and ask for a clause-by-clause mapping.
How much does biobank software cost?
Few vendors publish prices. Among those that do, as of October 2026: CloudLIMS lists 310 USD per user per month for an in-house biobank with three users, billed annually; LabKey lists its Sample Manager Starter plan at 6,540 USD per year for five users; OpenSpecimen lists enterprise licences from 75,000 USD one time plus annual support from 35,000 USD. Most other vendors quote on request, and migration and configuration services often add a large share on top of the licence.
Is open-source biobank software a realistic option?
It can be, but open-source code does not mean a free deployment. Someone has to host, validate, upgrade and support the system, and some open-source products reserve modules or support for paid editions. Before planning around a community edition, check in the current release notes and user forum which modules are needed for core specimen workflows and whether they are included.
Do biobanks need 21 CFR Part 11 compliant software?
Only when the biobank’s electronic records fall under FDA predicate rules, for example specimens and data that support an FDA-regulated clinical investigation. FDA has said it interprets the scope of Part 11 narrowly. A purely academic research collection is usually outside that scope, although audit trails and access controls are still good practice and are expected by the CAP Biorepository Accreditation Program.
Should freezer temperature monitoring be part of the biobank LIMS?
Alarms and on-call escalation are best handled by a dedicated monitoring system with regularly tested alerts. What the inventory software should do is link a temperature excursion to the specimens stored in the affected unit, so the biobank can annotate their quality afterwards. Several vendors list integrations with monitoring systems, so ask to see that link working in a demo.
Sources
- ISO 20387:2018 Biotechnology, Biobanking, General requirements for biobanking (accessed 2026-10-09)
- ISO/FDIS 20387 Biotechnology, Biobanking, General requirements for biobanks (accessed 2026-10-09)
- ISO 20387:2018 Assessment Worksheet (NATA) (accessed 2026-10-09)
- Biorepository Accreditation Program (College of American Pathologists) (accessed 2026-10-09)
- Biorepository Checklist, 08/21/2017 edition (College of American Pathologists) (accessed 2026-10-09)
- ISBER Best Practices for Repositories, Fifth Edition (accessed 2026-10-09)
- NCI Publishes Updated Best Practices for Biospecimen Resources (accessed 2026-10-09)
- Building and implementation of a common infrastructure for specimen and data storage at an academic medical center (Santillan et al., J Clin Transl Sci, 2025) (accessed 2026-10-09)
- Implementation of a Digital Maturity Framework for Biobanking (Rossi et al., J Biomed Inform, 2025) (accessed 2026-10-09)
- UT Southwestern Tissue Collection and Storage audit report (UT System, 2024) (accessed 2026-10-09)
- Research sector review (Human Tissue Authority, 2016) (accessed 2026-10-09)
- Several causes of the freezer failure according to internal report (Karolinska Institutet) (accessed 2026-10-09)
- The College of American Pathologists Biorepository Accreditation Program: Results from the First 10 Years (Leelatian et al., Biopreserv Biobank, 2026) (accessed 2026-10-09)
- NCI Best Practices for Biospecimen Resources, 2016 edition (accessed 2026-10-09)
- NCI Best Practices for Biospecimen Resources, 4th edition (PDF) (accessed 2026-10-09)
- NHSGGC Biorepository awarded UK’s first UKAS accreditation for biobanking (UKAS) (accessed 2026-10-09)
- Baobab Laboratory Information Management System (Bendou et al., Biopreserv Biobank, 2017) (accessed 2026-10-09)
- Laboratory information management systems for biobanks in low- and middle-income countries (IntechOpen, 2021) (accessed 2026-10-09)
- Newsbytes, October 2022 (CAP TODAY) (accessed 2026-10-09)
- OpenSpecimen pricing (accessed 2026-10-09)
- OpenSpecimen source code and licence (GitHub) (accessed 2026-10-09)
- Is the Community Edition being discontinued? (OpenSpecimen forum) (accessed 2026-10-09)
- Impossible to add specimens on Community Edition v12.2 (OpenSpecimen forum) (accessed 2026-10-09)
- CloudLIMS pricing (accessed 2026-10-09)
- CloudLIMS Biorepository LIMS (accessed 2026-10-09)
- LabKey pricing (accessed 2026-10-09)
- Freezerworks 2 upgrade (accessed 2026-10-09)
- Freezerworks FAQs (accessed 2026-10-09)
- Agilent OpenLab Sync for Biobanks (accessed 2026-10-09)
- 21 CFR 11.1 Scope (eCFR) (accessed 2026-10-09)
- Part 11, Electronic Records; Electronic Signatures, Scope and Application (FDA guidance, 2003) (accessed 2026-10-09)
- 45 CFR 46.116 General requirements for informed consent (eCFR) (accessed 2026-10-09)
- Regulation (EU) 2016/679, General Data Protection Regulation (EUR-Lex) (accessed 2026-10-09)
