Why a filename still matters
Search has not replaced the filename. In two studies, users estimated that they reached 56–68% of their files by navigating folders and only 4–15% by searching. Better search engines barely changed that [1].
HOW USERS REACH THEIR FILES
[1]56–68%
by navigating folders
4–15%
by searching
Brain imaging offers a reason. Navigating folders activates the same regions we use to find our way in the physical world, while search draws on language areas [2]. Browsing feels easier because, for the brain, it is the more automatic task.
“A filename should help you recognize a document before you open it. [3]
If people browse, the name is what they read. A scanner's scan001.pdf gives them nothing to read.
What memory keeps
Psychology tells us which parts of a name actually help.
Recognition is easier than recall. Recognizing something in front of you takes fewer memory cues than retrieving it from nothing, which is why good interfaces show options rather than ask you to remember them [4]. A name that describes its content lets you recognize the document in a list.
Time and people are strong cues. In a study of a personal search tool used by more than 230 employees, time and people proved to be important retrieval cues [5].
Finding is a mental task, not a technical one. Managing personal documents rests on categorizing, recognizing and recalling [6]. A convention should support all three.
A helpful name therefore answers four questions: when, what kind, what about, and from whom.
Four schools of thought
Most conventions in use today follow one of four schools. Each is right about something.
School
Chronological [7–9]
Leads with
The date
Strength
A clean timeline; sorts the same in every language
Weakness
A date alone says nothing about content
School
Classification [10,11]
Leads with
Document type or project
Strength
Keeps like with like
Weakness
Needs a fixed vocabulary; many documents fit several categories [2]
School
Descriptive [12]
Leads with
A readable title
Strength
Recognizable at a glance
Weakness
Grows long and drifts without rules
School
Coded [11,12]
Leads with
Reference numbers and abbreviations
Strength
Short and unique
Weakness
Opaque without a key
Our ISO-inspired naming structure, field by field
Our ISO-inspired naming structure combines the first three schools in a fixed order and leaves the codes out.
YYYY-MM-DocumentType-Subject-AuthorOrSource2026-08-Contract-OfficeLeaseRenewal-GlobalTech.pdf
Contract
One word for the kind of document. This is the classification school.
OfficeLeaseRenewal
A short, readable title. This is the descriptive school.
2026-08-Contract-OfficeLeaseRenewal-GlobalTech.pdf.Hyphens separate the fields. Capital letters separate the words inside a field. The name stays in the language of the document, so a French invoice becomes 2026-03-Facture-AbonnementInternet-TelecomNord.pdf.
One folder, sorted by name
Here is a small mixed archive as any file manager would list it.
2026-03-Facture-AbonnementInternet-TelecomNord.pdf2026-06-Judgment-RentReductionClaim-DistrictCourtLyon.pdf2026-08-Contract-OfficeLeaseRenewal-GlobalTech.pdf2026-08-Invoice-QuarterlyMaintenance-AcmeIndustries.pdf2026-09-Invoice-AnnualServerHosting-AcmeIndustries.pdfThe folder reads as a timeline. Within a month, documents line up by type. Every line says what the document is about and who sent it, so you can pick the right one without opening any of them.
Why the mix works
Our ISO-inspired naming structure is a moderate position: it takes what each school does well and stops before each one's weakness.
It sorts by the cue people rely on.
Time is one of the strongest retrieval cues [5], and most published guides put the date first for that reason [7–9].
It can be recognized, not decoded.
The subject is written in plain words, so you recognize the document instead of recalling what a code stood for [4].
It classifies without a rigid scheme.
The type word groups documents inside each month and is easy to search for. Nothing has to be forced into a single folder category [2].
It keeps the sender in view.
The author or source is the “who” that people remember [5].
It needs no key.
A new colleague or a family member can read the name on the first day.
It costs no effort when automated.
OCR and AI can read the four fields from the document and check the name before saving [3].
How it compares
Other conventions are good answers to their own use cases. The differences are small but they add up.
Convention
Filently [8]
Pattern
YYYY-MM-DD_Type_Details
Where it fits best
Personal and freelance files
What our ISO-inspired naming structure adds
A defined subject and sender instead of a free “details” field; one separator instead of two
Convention
NameQuick [9]
Pattern
YYYY-MM-DD__Entity__DocType
Where it fits best
Receipts and scans sorted by sender
What our ISO-inspired naming structure adds
A subject, so two letters of the same type from the same sender stay distinguishable
Convention
Docupile [10]
Pattern
Invoice_YYYY-MM-DD_ClientName
Where it fits best
Folders that hold one document type
What our ISO-inspired naming structure adds
A timeline across all document types
Convention
ISO 19650 [11]
Pattern
Project-Originator-…-Number
Where it fits best
Large construction projects with many firms
What our ISO-inspired naming structure adds
Names that need no code list
We agree with these guides more than we differ. The three document tools all use an ISO 8601 date and give the type its own field. Our ISO-inspired naming structure adds the readable subject and a fixed place for the author or source.
Honest limits
No convention is free, and this one has costs too.
The names are long. Our example has 46 characters before the extension, above the 32 that one research-data guide suggests [12]. We trade brevity for recognition.
The month is coarse. Two documents from the same month sort by type, not by day. Where the day matters, such as deadlines or hearings, extend the date to 2026-08-14. It is still ISO 8601 [13].
CamelCase is not proven faster. In studies on program code, beginners read camel case more accurately and with less effort [14], while an eye-tracking study found underscore names were recognized faster at equal accuracy [15]. Neither tested filenames.
Search can stumble on joined words. Some tools treat OfficeLeaseRenewal as one word. Keep the subject to two to four words.
Hyphens do two jobs. They sit inside the date and between fields. People read this easily; scripts must know that the first two parts are the date.
No head-to-head trial exists. The case rests on research into memory and navigation [1,2,5], not on a test of this exact pattern. It is reasoned, not proven.
The verdict
For the mixed archive of a household, a freelancer or a small office, we consider our ISO-inspired naming structure the best general-purpose choice. It gives you the timeline of the chronological school, the grouping of the classification school and the readability of the descriptive school in one name. A folder with a single document type can lead with the type, and a large building project should stay with its industry standard. For most other cases, date, document type, subject and author or source are enough.
The naming checklist
- 01Start with year and month as
YYYY-MM - 02Add one word for the document type
- 03Describe the subject in two to four words, each capitalized
- 04End with the author or source
- 05Separate the four fields with hyphens
- 06Keep the language of the document
- 07Add the day when the exact date matters
- 08Leave out account numbers and other sensitive data
- 09Apply the same order to every file
CONTINUE READING
PAPERARCHIVER
Let your documents name themselves.
PaperArchiver reads each document, finds its date, document type, subject and author or source, and applies our ISO-inspired naming structure automatically.
Explore AI PDF Organization2026-08-Contract-OfficeLeaseRenewal-GlobalTech.pdf
Disclosure: PaperArchiver develops software that applies the naming structure described in this article.
References
- 1.Bergman O, Beyth-Marom R, Nachmias R, Gradovitch N, Whittaker S. Improved search engines and navigation preference in personal information management. ACM Trans Inf Syst. 2008;26(4):20.
- 2.Benn Y, Bergman O, Glazer L, Arent P, Wilkinson ID, Varley R, et al. Navigating through digital folders uses the same brain structures as real world navigation. Sci Rep. 2015;5:14719.
- 3.PaperArchiver. How to automatically rename PDF files [Internet]. PaperArchiver; [date unknown] [cited 2026 Oct 6]. Available from: paperarchiver.com/blog/how-to-automatically-rename-pdf-files
- 4.Budiu R. Memory recognition and recall in user interfaces [Internet]. Nielsen Norman Group; 2014 [cited 2026 Oct 6]. Available from: nngroup.com/articles/recognition-and-recall
- 5.Dumais S, Cutrell E, Cadiz JJ, Jancke G, Sarin R, Robbins DC. Stuff I've Seen: a system for personal information retrieval and re-use. In: Proceedings of the 26th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval; 2003 Jul 28–Aug 1; Toronto, Canada. New York: ACM; 2003. p. 72–9.
- 6.Lansdale MW. The psychology of personal information management. Appl Ergon. 1988;19(1):55–66.
- 7.Digital Preservation Coalition. Why are file naming and file formats important for digital preservation? DP Note 4 [Internet]. DPC; [date unknown] [cited 2026 Oct 6]. Available from: dpconline.org/docs/1865-dp-note-4-file-naming-and-formats
- 8.Uhrmeister V. How to name files: five rules and worked examples [Internet]. Filently; 2025 Dec 24 [updated 2026 Jun 26; cited 2026 Oct 6]. Available from: filently.com/blog/file-naming-conventions
- 9.NameQuick Team. Automatically name scanned documents from the To File pile [Internet]. Synthropy; 2026 May 4 [updated 2026 Sep 23; cited 2026 Oct 6]. Available from: namequick.app/blog/scan-and-organize-documents-software
- 10.Patel U. AI auto file naming: a game-changer for businesses [Internet]. Docupile; 2025 Jan 3 [updated 2026 Jul 13; cited 2026 Oct 6]. Available from: docupile.com/ai-auto-file-naming-is-essential-for-businesses
- 11.Pasalski J, Owerko T, Kasznia D, Wrzosek M, Rochel M, Czerw K. Efficient file-naming convention for data exchange in construction industry. Geomat Environ Eng. 2025;19(2):99–129.
- 12.Briney KA. File naming convention worksheet [Internet]. Pasadena (CA): Caltech Library; 2020 [cited 2026 Oct 6]. doi:10.7907/894q-zr22
- 13.International Organization for Standardization. ISO 8601-1:2019. Date and time — Representations for information interchange — Part 1: Basic rules. Geneva: ISO; 2019.
- 14.Binkley D, Davis M, Lawrie D, Maletic JI, Morrell C, Sharif B. The impact of identifier style on effort and comprehension. Empir Softw Eng. 2013;18(2):219–76.
- 15.Sharif B, Maletic JI. An eye tracking study on camelCase and under_score identifier styles. In: 2010 IEEE 18th International Conference on Program Comprehension; 2010 Jun 30–Jul 2; Braga, Portugal. IEEE; 2010. doi:10.1109/ICPC.2010.41