Researchers at the University of Southampton have published a white paper offering practical guidance on selecting, implementing and sustaining an Electronic Lab Notebook (ELN). ELNs in Chemistry: From Implementation to Sustainable Stewardship was launched in August by the Physical Sciences Data Infrastructure (PSDI) initiative.
The authors are Samantha Pearman-Kanza, Principal Enterprise Fellow, and Matthew Partridge, Senior Enterprise Fellow, both based in the university’s School of Chemistry and Chemical Engineering. The paper draws on the School’s ELN implementation programme and on wider engagement with the ELN, research data and digital research infrastructure communities.
That programme began with a trial of 50 licences and was later extended to all staff and undergraduates in the School. More than 90% of trial participants chose to continue using the ELN, as we reported in a previous interview.
The university is now exploring how the ELN could be extended to other scientific departments. In a follow-up interview, Pearman-Kanza and Partridge explain how to approach a new discipline, how to prioritise selection requirements, and what the white paper’s recommendations on interoperability, data capture and metadata mean in practice.
Following the publication of a white paper on their findings, the team behind the University of Southampton’s Electronic Lab Notebook rollout explains how to take an ELN into new disciplines, which requirements are non-negotiable, and why an exit strategy belongs in the selection process.
In our previous interview, you mentioned wanting to expand ELN use into different scientific disciplines. How do you start this process?
“The first step is understanding how researchers in that discipline currently work,” say Pearman-Kanza and Partridge. “Different fields generate different types of data, use different instruments and have different approaches to recording and reviewing research. Before configuring an ELN, we recommend mapping existing workflows, speaking to researchers and identifying where information is currently being captured, shared and stored.”
That process has already started closer to home. “At the University of Southampton, we are currently exploring how our ELN could be expanded into other scientific departments,” they say. “Our starting point in these instances is to assess whether Revvity Signals can be adapted to meet the requirements and workflows of those disciplines. If it can’t, then we plan to use what we have learnt from our experience in Chemistry and Chemical Engineering to help those departments identify their requirements, evaluate alternative platforms and run their own trials.”
An organisation with no ELN in place would start further back by identifying its requirements and selection criteria before considering the range of available platforms.
“The aim is not to recommend one particular system, but to help organisations identify the ELN that best fits their researchers, workflows and institutional context,” the pair say. Whatever the discipline, and whether or not the ELN is already used elsewhere in the organisation, they recommend that realistic trials with representative users come next. “The aim is not to force every discipline into a single way of working, but to identify where common structures can be introduced while retaining the flexibility researchers need,” they explain. “Successful implementation is usually driven by adapting the ELN to support existing research practices rather than asking researchers to completely redesign how they work.”
How should an organisation prioritise requirements when choosing an ELN?
The white paper sets out criteria for choosing an ELN, including compliance, data integrity, functionality, usability, accessibility, interoperability, scalability, and total cost of ownership.
“We would start with the non-negotiables,” say the authors. “Compliance, security and data integrity requirements need to be met before anything else because they underpin trust in the scientific record. If a platform cannot satisfy those requirements, it should not progress further in the evaluation process.”
Less obviously, they place the ability to leave a system in the same category. “Long-term access to and control of the scientific record should also be treated as a non-negotiable,” they say. “Our position is that unless we have created a mechanism to get data out of the ELN, we shouldn’t be putting it into the ELN.”
That means considering a data exit strategy during selection, including how records, metadata, and associated data can be exported, backed up, retained, and, ideally, migrated to another system. The pair also draw a distinction that is easy to miss during procurement. “Being able to retain a readable copy is important, but this is not necessarily the same as being able to reuse or migrate the structured scientific record."
Only then does the experience of the people using the system come into play, although it ranks highly. “After that, usability becomes one of the most important considerations,” they continue. “Researchers need to be willing and able to use the system as part of their everyday work. Even a highly capable platform can struggle if users find it difficult or essentially require more time or more ‘clicks’ to do the same job they were already doing.”
With those foundations in place, organisations can assess functionality, scalability, interoperability and total cost of ownership in the context of their long-term research strategy. On cost, the authors add a caution: the calculation, they note, “should include not only the licence cost, but also the hardware, infrastructure, training, administration, technical support and staff time needed to implement and sustain the system”.
How important is interoperability in ELN selection?
Interoperability may come after compliance and usability in that order, but the authors count it among the most important factors in the selection process. It affects both how well an ELN fits into the current research environment and whether records can be transferred in the future.
The first direction is towards the tools researchers already rely on. “Organisations should consider interoperability with existing software, instruments, inventories and other research systems,” they explain. “If researchers already depend on particular software or hardware, ELN selection should take this into account.”
Southampton’s own choice is a case in point. “At Southampton, compatibility with existing departmental software, including ChemDraw and ChemInventory, was an important factor in selecting Revvity Signals for the trial,” the pair say. “This type of interoperability can reduce duplication and allow the ELN to support, rather than disrupt, existing research workflows.”
The second direction is between ELNs themselves. “Organisations must also consider interoperability between ELNs, because very few organisations will use the same platform indefinitely,” they say. “An ELN should allow records, metadata and associated data to be exported and transferred in ways that can be understood and reused by other ELNs. Without this, institutions risk becoming locked into a particular supplier, making future migration difficult, costly and potentially resulting in the loss of valuable scientific context.”
The benefit, they argue, reaches well beyond procurement. “Strong ELN-to-ELN interoperability helps ensure that research records remain accessible, interpretable and under institutional control over the long term, while reducing dependence on individual vendors and future-proofing the scientific record.”
Your previous work has emphasised that FAIR data practices need to begin at the point of data creation. Do you have any best practices for other organisations regarding data collection and storage?
“One of the most important lessons from our work at Southampton is that good data practices were adopted better when embedded into day-to-day research activities rather than applied retrospectively,” explain Pearman-Kanza and Partridge. “Much of the context that makes data understandable and reusable is easiest to capture while experiments are being performed, not months or years later.”
In practice, that means keeping experimental records, instrument outputs, samples and supporting files connected from the outset. “Templates, structured workflows, naming conventions, tagging and automatically populated fields can all help researchers capture useful context consistently without relying on retrospective data curation,” they explain.
It also means making sure researchers can reach the ELN where the work is actually being done, a practical point that a software evaluation can overlook. “Our use of dedicated in-lab hardware improved researchers’ ability to capture information at the point of creation, rather than creating a disconnect between the experiment and the digital record,” the authors say. “It also reduces the need to record information elsewhere before transcribing it into the ELN. Reliable Wi-Fi, access to power and suitable laboratory devices therefore need to be considered as part of ELN implementation.”
The same thinking guided the design of the processes themselves. “We also focused on creating approaches that fitted naturally into researchers’ workflows rather than adding additional administrative tasks,” they say. “The more seamless the process is, the more sustainable it becomes.”
Asked which guidelines shaped the project, the pair point to a combination of sources. The School of Chemistry and Chemical Engineering project, they say, “was informed by established FAIR principles, research data management guidance and wider engagement with the ELN and digital research infrastructure communities. The goal was to create records that remain understandable, reusable and trustworthy over the long term.”
The white paper discusses the importance of sensible metadata. How much metadata is enough?
“‘Enough metadata’ means capturing sufficient information for someone to understand, find and reuse a research record in the future,” say Pearman-Kanza and Partridge. “That usually includes basic details such as who performed the work, when it was carried out, what materials or samples were involved, and how the record relates to the underlying data.”
Getting the level right matters in both directions. “The challenge is finding the right balance. Too little metadata limits discoverability and reuse, while too much can make systems burdensome and discourage adoption. Our recommendation is to focus on metadata that delivers clear value to researchers and supports searching, reviewing and future interpretation of records. We also looked for ways of automating as much of the metadata capture as possible.”
Their advice to organisations worried about over-engineering a standard is to start small. “A practical approach is to begin with a relatively small set of high-value metadata fields and refine requirements over time based on feedback from users. A shared core template can be supplemented with additional fields where particular disciplines or research groups have a clear need, rather than attempting to create a single standard that anticipates every possible requirement from the outset.”
As with every other part of the ELN journey they describe, adoption depends on what researchers receive in return. “Researchers are far more likely to engage with metadata standards when they can see a direct benefit to their own work, rather than viewing them as an administrative exercise.”
Treating the ELN as infrastructure
The white paper concludes that an ELN should be treated as long-term research infrastructure. Sustaining one, the authors write, requires ongoing resourcing, maintained guidance, regular review, storage planning and a clear exit strategy, so that research records remain accessible, interpretable and under institutional control.
The white paper contains more guidance than this article covers. It examines implementation in practice, with an emphasis on clear ownership, process mapping, training and continued support. It also considers change management, including governance, stakeholder engagement, communication, and the embedding of ELN use in everyday research culture.
Download the white paper: ELNs in Chemistry: From Implementation to Sustainable Stewardship.
Dr Samantha Pearman-Kanza is a Principal Enterprise Fellow at the University of Southampton and PSDI’s Pathfinder Lead for Process Recording. Doctor Matthew Partridge is a Senior Enterprise Fellow (PSDI) at the University of Southampton.
