- Online databases are digitised selectively, often cover only recent years, and can omit the records that matter most.
- Original documents carry information that a scan or database entry loses, including stamps, marginalia and the order in which papers were filed.
- We combine online research with archive visits, lawful records requests and interviews, and we plan for the time this takes.
A great deal of investigative research now begins with a search box. Corporate registers, court portals, sanctions lists and news archives can be queried in minutes, and the results feel complete. They rarely are. Much of what matters about a person, a company or an asset still sits in paper files, court registries, local and national archives, land offices and libraries, and in the memories of people who were present at the time. Research that stops at the screen tends to describe only the part of the record that someone chose to put online.
What online sources leave out
Digitisation is selective. Archives and registries put material online according to budgets, priorities and legal constraints, and the result is uneven. Recent records are more likely to be searchable than older ones. Some databases begin at a fixed point and say nothing about what came before. Others hold an index without the underlying documents, or a summary without the filing itself.
Gaps are not always visible. A search that returns nothing may mean that no record exists, or only that the relevant record was never digitised. Local newspapers, planning files, older court judgments and company filings from before electronic registration are common examples. Treating an empty result as a negative finding is one of the easier ways to reach a wrong conclusion.
The problem is sharper in some regions. In post-Soviet countries, some historical records remain in regional archives, and many are in Russian, which was widely used in Soviet administration, even where another language is now the state language. Records from the 1990s, when many companies and assets changed hands, can be partial, handwritten or held only locally. Access rules vary by country and by archive, and they change, so they need to be checked at the time of research.
The value of the original document
A scanned image or a database entry carries the text of a document. The original carries more. A receipt stamp shows when a filing arrived and who received it. Marginal notes show that someone read it and what they made of it. Corrections, signatures and the type of form used can indicate when a document was prepared, or whether it was later replaced.
Sequence matters too. The order of papers in a file, the gaps in a numbered series and the documents filed alongside the one being sought often say as much as the document itself. A share transfer recorded on a date that does not fit the surrounding correspondence is a reason to ask further questions. Very little of this survives the conversion of a file into a single search result.
An empty search result is a fact about the database before it is a fact about the world.
How we combine the two
We start online because it is efficient and because it shows where the gaps are. The online work produces a map of what is known, what is missing and where the missing material is likely to be held. From that map we decide which archives, registries or libraries to visit, which records to request and which people might remember the period in question.
Records requests are made only where the law allows and through the proper channels, whether that is a court registry, a land office, a public archive or a freedom of information regime. Some records are closed for set periods, restricted under data protection rules or available only to parties with a defined interest. We respect those limits. Where local researchers visit archives on our behalf, we brief them on what is lawful and safe in that jurisdiction and we do not ask them to take risks for us.
People are part of the archive. Former employees, retired officials, local journalists and long-standing residents often know which records exist, where they were sent and why a file is thinner than it should be. Their recollections need corroboration like any other source, and they frequently point to documents that no search would find.
Time and patience
Archive work is slow. Requests can take weeks to process, reading rooms keep limited hours, and catalogues can be incomplete or organised in ways that only make sense after several visits. A document may sit in an uncatalogued box or under a former name of the company concerned. Patience is a research tool in its own right, and a realistic timetable is part of a sound research plan.
Some enquiries need answers faster than an archive can supply them. In those cases we set out what the online record supports, what it cannot settle and what a records search would add, so that any decision about time and cost is an informed one.
Our approach is to treat online sources as a starting point and the documentary record as the standard. Where the original exists and can be reached lawfully, we try to see it. Where it cannot, we say so and grade our findings accordingly.
Published 16 July 2024. Last reviewed October 2026. This note describes public records and general research practice, and it does not discuss any client matter. Laws, registers and access to records change, so check the current position before relying on it. It is general information, not legal, financial or investment advice.
