Absence or inadmissibility of the evidence file
It is useful to recall briefly that what is commonly referred to as the “evidence file”, sometimes also described as an “evidence dossier” or, in the case of DocuSign, a “certificate of completion”, is a compilation of traces of the signing operation generated by the trust service provider on the basis of the data collected through its signing platform. The evidence file is generally produced in XML format and, in order to be intelligible to a non-technical reader, it usually contains a number of elements describing the signing operation in plain text, such as the names of the signatories, the title of the signed instrument, timestamps reflecting the chronology of the operation, and signature authentication identifiers such as the signatory’s email address and/or telephone number. The format of the evidence file is not standardised, and there are as many presentations as there are providers, ranging from a highly technical narrative to a bare, unadorned report.
Judicial scrutiny of this file operates on two levels.
First, judges tend to attribute to it qualities that it does not in fact possess when they state, for example¹:
“neither at first instance nor on appeal does she produce the electronic evidence file enabling the methods used to guarantee the identity of the signatory and the archiving method to be ascertained. She therefore fails to establish proof of recourse to a reliable electronic signature process intended to guarantee the integrity of the instrument and the identity of the signatory, in accordance with Articles 1366 and 1367 of the Civil Code.”
In practice, the evidence file does not guarantee the identity of the signatory, at least where non-qualified electronic signatures are concerned, as is generally the case today. At most, it refers to an email address or telephone number, or sometimes to the mention of an identity document, provided to the service provider during the signing operation by the professional implementing the signing workflow. The link between these elements and the purported signatory is not verified by the trust service provider, but by that professional. It is not the evidence file that will reveal the steps taken by the latter to ensure the reliability of those elements; such information would require a description of the signatory identification process, which is in practice rarely produced in litigation, not least because it often does not exist. By way of illustration, an email address or mobile phone number supplied on a declarative basis cannot establish a person’s identity unless additional verification measures have been implemented, and the evidence file will disclose nothing more than what the provider itself knows on the subject, namely nothing.
Similarly, the evidence file does not provide any information as to the archiving method, for the simple reason that, chronologically, it is generated before the document is transferred to an archival system. Finally, a number of recent decisions² suggest that the evidence file would make it possible to identify the “security rules” applied, which again attributes to it a scope it does not have.
These approximations highlight the benefit of harmonising the presentation of evidence files and clarifying the scope of their content, which remains widely misunderstood by legal practitioners. The issue is not addressed by any regulation, standard or consensus among trust service providers, which is regrettable given its significance in judicial proceedings.
Secondly, judges require the evidence file to be “readable”, a requirement that is readily understandable. A particularly strict decision delivered by the Fourth Chamber of the Paris Court of Appeal, which is usually receptive to the validity of electronic signatures, is illustrative³:
“The documents viewed and signed are not detailed, and the document contains four pages of coding that are completely unreadable.”
Finally, and this is not new, the failure to produce an evidence file almost automatically leads to the rejection of the electronic signature. This represents a significant weakness where the dispute concerns a signing infrastructure implemented without due regard to the importance of this document.
A statement on the document that it was electronically signed is not sufficient
In wording reproduced almost verbatim by very different courts (Riom, Douai, Paris, Rouen, Versailles), judges state that “proof of an electronic signature, even a simple one, requires more than the mere mention of an electronic signature at the foot of a document, even when accompanied by documents establishing the existence of a relationship between the parties”?.
This assertion is invariably made in cases where the electronic signature presentation file is clearly deficient: the evidence file is not produced, or is incomplete, or is unintelligible; or the reference to the electronic signature appears on one document but not on others forming part of the same set.
Several observations may be made in this respect.
The first is that what judges are stating here is self-evident. Indicating on a document that it has been electronically signed does not in itself prove that this is the case, since, unlike a handwritten signature, an electronic signature is invisible. It consists of a cryptographic processing of the digital file that leaves no visible trace on the document unless certain visible elements characterising the electronic signature are added. It should be noted that this is neither inevitable nor mandatory, but simply a feature of current implementations, which could be modified by introducing a non-forgeable encoding visible when printed, allowing the signing operation to be materialised. This is not yet the case, but in our view it is a line of development that should not be overlooked as digital contracting becomes more widespread.
The second observation is that these decisions should not be interpreted as dispensing with the need to display, in one form or another, a visible indication of the electronic signature on the document itself?. On the contrary, such an indication is essential, particularly where the signature block contains a signature artefact or a signature captured on a tablet, practices that frequently give rise in litigation to confusion between electronic and handwritten signatures.
Our final observation returns to a point we have repeatedly made in these pages. To present an electronic signature file in the best possible light, the materials required to satisfy a judge are now well identified: a readable evidence file containing a reference common to the signed instrument, relevant certifications, an explanatory guide to the process implemented describing the identification of signatories, and proof of reliable archiving.
It is becoming increasingly risky to disregard these requirements, as judges, notwithstanding certain approximations, are dealing with a growing number of cases involving electronically signed instruments and are becoming more demanding as to the quality of what is submitted to them, which (it must be acknowledged) is sometimes poor. This is a genuine issue, as judicial difficulties of this kind should not result in professionals losing more money than they gain from the digital transformation of their processes, particularly in transactions where the financial stakes will continue to increase.
References
[1] CA Rouen, 10 avril 2025, RG n° 24/01774, également CA Paris, 3 avril 2025, RG n° 23/19316
[2] CA Douai, 27 mars 2025, RG n° 22/05040 et RG n° 22/05041
[3] CA Paris, 3 avril 2025, RG n° 23/19316
[4] CA Riom 19 mars 2025 RG n° 24/00497 ; CA Douai 27 mars 2025, RG n° 22/05040 et RG n° 22/05041 ; CA Paris 3 avril 2025, RG n° 23/19316 ; CA Rouen 10 avril 2025, RG n° 24/01774, CA Versailles 29 avril 2025 RG n°24/04372, CA Paris 15 mai 2025 RG n°24/02375
[5] Par ex. : CA Amiens 24 avril 2025 RG N°23/04010
Disclaimer
The opinions, presentations, figures and estimates set forth on the website including in the blog are for informational purposes only and should not be construed as legal advice. For legal advice you should contact a legal professional in your jurisdiction.
The use of any content on this website, including in this blog, for any commercial purposes, including resale, is prohibited, unless permission is first obtained from Evidency. Request for permission should state the purpose and the extent of the reproduction. For non-commercial purposes, all material in this publication may be freely quoted or reprinted, but acknowledgement is required, together with a link to this website.
