Our eDisclosure service
Jandr is an Everlaw partner and reseller. We hold the highest level of Everlaw certification available, Litigation Support Manager, and with four certified project managers and over 20 years of combined Everlaw experience in the team, we are one of the most experienced Everlaw practices in the UK. Everlaw regularly refers clients to us when ongoing project management support is needed. We are Cyber Essentials certified and all work is performed on secure Microsoft 365 cloud infrastructure. Everlaw is ISO 27001 certified and SOC 2 Type II compliant.
Where a matter requires a larger review team, we can mobilise fully autonomous review teams of experienced legal freelancers at short notice. Every team member is provisioned with a secure virtual machine and works within Everlaw under the supervision of a Jandr project manager. The instructing lawyers have a single point of contact throughout.
We work on matters of all sizes, from single-custodian reviews to large multi-party disclosure exercises. Our pricing is flexible: time and materials with detailed estimates, fixed fees, volume discounts and retainer arrangements are all available depending on what the matter requires. Whatever structure we agree, we track and report against estimates so there are no surprises at invoice time.
How we work
Every eDisclosure matter starts with a scoping conversation. We want to understand the issues in the case, the likely document population and the disclosure order or protocol that will govern the exercise. That conversation shapes everything that follows.
From there, we take ownership of the end-to-end process:
Disclosure scope
If engaged early enough, we can advise ahead of the disclosure case management conference, or its equivalent in arbitration, to help arrive at a reasonable and proportionate disclosure requirement. Getting the scope right before the order is made is considerably easier than revisiting it afterwards
Data collection
We work with your IT team or third-party forensic providers to collect data from the relevant custodians and systems. We advise on collection scope and ensure the process is documented and defensible.
Processing and early case assessment
Once data is loaded into Everlaw, we analyse the document population before the review begins. Volume, date distribution, key custodians and concepts are all visible at this stage. Early case assessment often surfaces useful information before a single document has been reviewed.
Review design
Because our services are lawyer-led, we understand the end requirement and bear it in mind at every stage. A failure to anticipate a requirement, such as the redaction of sensitive third-party details, adds cost and last-minute effort compared to designing it into the workflow at the outset. We design the review methodology, document it and ensure it can be explained and defended if challenged.
Review management
We manage the review from start to finish, whether that means running the exercise ourselves, supervising a larger team, or a combination of both. Quality control is built in throughout, not applied at the end.
Production
We produce documents in the format required by the protocol or agreed with the other side, with a full audit trail of the review decisions taken.
Beyond disclosure: strategic insight
A disclosure review done well is more than a defensibility exercise. The document population in a commercial dispute contains the factual record of the events in issue. A well-run review does not just identify what must be disclosed; it finds the evidence that shapes the case strategy.
Everlaw’s analytical and AI functionality allows us to identify concepts, patterns and sentiment across large document sets without relying on search terms alone. Relevant documents, good or bad, surface earlier in the review process, giving the legal team more time to assess their significance.
Everlaw’s collaboration and work product tools allow legal teams to build chronologies, draft arguments and prepare witness statements directly within the platform, drawing on the underlying documents in the matter. The review and the legal work happen in the same environment, reducing the risk of things falling through the gap.
We work closely with the legal team throughout the review to ensure that what we find is understood in context. The output of a Jandr review is not just a disclosure list. It is a clearer picture of the case.
The language of eDisclosure
Clustering
A technique that groups documents by similarity of content, allowing reviewers to make consistent decisions across large numbers of related documents efficiently.
Concept search
A search methodology that finds documents based on meaning rather than specific words. Useful for identifying documents that are relevant to a theme or issue without knowing in advance which words the authors used.
Continuous active learning (CAL)
The current standard approach in technology assisted review. Unlike earlier predictive coding models, CAL does not require a fixed training phase. The system learns continuously as reviewers code documents, improving its predictions throughout the review. This means the most relevant documents rise to the top of the review queue early, and the exercise can be stopped when the marginal return on further review becomes negligible.
Custodian
The grouping of emails into complete conversation chains, so that reviewers can read an exchange in context rather than reviewing each message individually. Reduces review time and improves consistency.
De-duplication and near-duplicate detection
De-duplication removes exact copies of the same document from the review set. Near-duplicate detection identifies documents that are substantially similar but not identical, for example different versions of the same contract. Both reduce review volume without risk of missing relevant material.
eDisclosure and eDiscovery: the same thing
eDisclosure is the term used in England and Wales. eDiscovery is used in the United States and some other common law jurisdictions. They describe the same process: using technology to collect, process, review and produce electronically stored information in the context of legal proceedings or investigations. You will see both terms used by technology providers; on this page we use eDisclosure throughout.
Email threading
The individual whose data is being collected for review: typically an employee or former employee whose emails, files and messages are within scope.
Keyword search
An umbrella term for any approach that uses technology to prioritise or classify documents for review, reducing the volume that requires human attention. Keyword search is technically a form of TAR, but the term is most commonly used to describe machine learning approaches.
Predictive coding
A machine learning approach to document review in which a reviewer codes a sample of documents as relevant or irrelevant, and the system uses those decisions to predict the relevance of the remaining document set. The approach has been accepted by the English courts since Pyrrho Investments v MWB Property in 2016.
Technology assisted review (TAR)
The traditional approach to document review: searching for documents that contain specific words or phrases. Straightforward to explain to a court, but limited in practice. Keywords miss documents that use different language to express the same concept, and they return large volumes of irrelevant material that still require human review.
Redaction
Redaction is the process of obscuring parts of a document before it is produced to another party. It is used to protect privileged material, confidential third-party information, or data outside the scope of the disclosure order. In an eDisclosure context, redactions are applied electronically within the review platform and burned into the document at the production stage, so the underlying text cannot be recovered by the recipient.
Sentiment analysis
Sentiment analysis uses AI to assess the tone or emotional content of documents within a dataset. In a litigation or investigations context, it can identify communications that express frustration, hostility or anxiety, which may indicate that the author knew something was wrong or was under pressure at the time of writing. It is a useful triage tool in large datasets where reading everything is not feasible.
Production
Production is the formal step in which documents are disclosed to another party. A production typically consists of documents in an agreed format, accompanied by a load file containing metadata and a schedule identifying what has been produced and on what basis. Getting the production format right matters: errors can require reproductions, which are costly and draw unwanted attention.
Native files
A native file is a document in its original format, as created by the application that produced it. A native Excel spreadsheet, for example, retains its formulae, hidden rows and cell formatting in a way that a PDF version does not. Whether to produce in native format or as image files is a common point of disclosure negotiation. Native files contain more information but can reveal more than intended; image productions are more controlled but may strip out relevant content.
Produced documents
Produced documents are those formally disclosed to another party following the review. Once produced, a document forms part of the disclosure record for the matter and can be relied upon in correspondence, witness statements and at trial. The production set is distinct from the full review set: not every document reviewed will be produced, and the reasons for withholding documents must be recorded and, where required, communicated to the other side.
Bates numbers
A Bates number is a unique reference applied to each page of a document production, used to identify and track individual pages throughout the litigation. In a modern eDisclosure production, Bates numbers are applied automatically by the review platform, typically formatted as a prefix followed by a sequential number, for example JNR000001. Any reference to a specific document in correspondence or court submissions will use its Bates number.
Metadata
Metadata is data about a document rather than its content. It includes information such as who created a file, when it was created and last modified, and in the case of emails, the full send and receive information. Metadata is often invisible when opening a document normally but is captured and searchable within a review platform. It can be significant in litigation: the metadata trail may reveal, for example, that a document was amended after a key date, or that its stated author did not in fact create it.
The disclosure obligation and cost management
Disclosure in litigation and arbitration can be one of the most significant cost items in a dispute. Those costs are scrutinised: by the court in costs budgeting, by arbitral tribunals, and by clients who need to understand what they are committing to before the exercise begins.
We are regularly engaged at or before the cost estimation stage. Early involvement allows us to assess the document population, identify the most cost-effective approach and produce realistic estimates. A well-scoped exercise with the right technology costs considerably less than one that starts without a clear plan.
AI in disclosure: the legal position
There is a perception in some quarters that the use of AI in disclosure workflows is legally uncertain or that courts require special permission before it can be used. That perception is mistaken.
The Civil Procedure Rules are technology agnostic. They require disclosure to be conducted properly and proportionately; they do not prescribe how. Litigants are entitled to use AI tools in their disclosure workflows now, without a specific court order permitting it. The same principle applies in most arbitral frameworks.
What the rules do require is that the approach taken is reasonable, proportionate and defensible. A party that uses AI must be able to explain and justify its methodology if challenged. That is no different from the obligation that applies to any other disclosure approach.
AI does not have to be the primary basis for a review, or used in isolation. It has a role to play in early case assessment, prioritisation and quality assurance regardless of the overall methodology. The optimal approach to a review will often combine keyword searching, continuous active learning, AI coding and human review in whatever proportion best serves the matter.
Far from prohibiting AI use, the CPR’s requirement for proportionality arguably makes a blanket refusal to consider technology-assisted review difficult to justify, particularly in high-volume matters where manual review alone would be disproportionately expensive. Jandr’s role is to design the right combination of tools and methodology, justify it and execute it to a standard the court or tribunal will accept.
Talk to us about eDisclosure
Whether you have a live matter, an upcoming case or just want to understand what we can do, we are happy to have a no-obligation conversation. We can also arrange a demonstration of Everlaw’s platform or an AI in eDisclosure demo so you can see the technology in action before committing to anything.
Schedule a conversation
Schedule a conversation
Book an Everlaw demo
Book an Everlaw demo
Book an AI in eDisclosure demo
Book an AI in eDisclosure demo
Contact Ryan Hockley at ryan@jandr.legal or call 0800 772 3386.