# File Naming Conventions for Research Notes: A Step-by-Step Guide By the end of this guide you will be able to design a file naming convention for your research notes, apply it to new files as you create them, rename an existing pile of notes in bulk with a few commands, and avoid the mistakes that make note collections unusable six months later. You need nothing installed beyond a plain text editor. The command examples use the macOS or Linux terminal, with a PowerShell alternative for Windows. ## Why File Naming Conventions Matter in Research Research note taking generates files faster than most people expect. A single project can produce meeting notes, lab observations, interview transcripts, data extracts, manuscript drafts, and hundreds of literature notes. Modern evidence synthesis shows how quickly this scales. A bibliometric analysis of primary insomnia treatment identified 2,826 relevant publications across two decades of the Web of Science Core Collection [1]. A Cochrane methods project is currently evaluating artificial intelligence tools across three distinct review tasks, title and abstract screening, full-text screening, and data extraction [2]. Each of those stages produces working files that someone will need to find again during auditing or revision. Findability is the whole point of a naming convention. In one machine learning study, researchers built an expert knowledge base that recorded model outputs and the clinician revisions made to them, then used a retrieval algorithm to pull relevant historical cases for future runs [3]. That kind of workflow only works when every stored record can be located by its name or metadata. Your notes deserve the same discipline, even if your knowledge base is a folder on a laptop. A naming convention is a fixed pattern that every file name follows. It turns file names from free text into structured, sortable, machine-friendly records. This guide walks through building one from scratch. ## Core Principles of Research File Naming Seven principles cover almost every situation. They come from long-standing practice in [data management](/blog/guides/data-management-basics-principles-processes-and-best-practices) and from how operating systems actually handle file names. **1. Order elements from general to specific.** The most significant element goes leftmost. That is the date, because dates sort chronologically as text. A file starting with `2026-11-11` sorts before one starting with `2026-11-12` in every file browser, with no extra work. **2. Use ISO 8601 dates.** ISO 8601 is the international standard for date representation, and the basic format is `YYYY-MM-DD`. So November 11, 2026 is `2026-11-11`. This format sorts correctly, is unambiguous about month and day, and is readable in every country. **3. Avoid spaces.** Spaces force you or your scripts to quote every file name, and some older tools misinterpret them. Use underscores `_` to separate elements and hyphens `-` to join words inside an element. **4. Avoid special characters.** Windows forbids these characters in file names: `\ / : * ? " < > |`. The forward slash is also the directory separator on macOS and Linux, so it breaks paths there too. Stick to letters, numbers, underscores, and hyphens. **5. Use zero-padded numbers.** Version 2 of ten versions should sort after version 9, so write `v02` through `v10` rather than `v2` through `v10`. Zero padding keeps ascending order in alphabetical sorts. **6. Keep the descriptive part short.** Windows historically limits a full path to 260 characters. Deep folder structures plus long file names can hit that ceiling. Aim for descriptive elements under about 40 characters. **7. Be consistent or be findable.** A mediocre convention applied to every file beats a perfect convention applied to half of them. ## File Naming Conventions for Research Notes: Step-by-Step This walkthrough builds a working convention from nothing. Follow it with your own project details substituted in. ### Step 1: Choose your elements Pick the elements your notes actually need. For most research projects, five are enough. 1. **Date** in ISO 8601 format, when the note was created or last revised. 2. **Project code**, a short uppercase abbreviation, for example `INS` for an insomnia project or `PGX` for a pharmacogenetics project. 3. **Note type**, a fixed vocabulary of three or four types. Common choices are `raw` for capture-stage notes, `lit` for literature notes, `data` for dataset documentation, and `out` for outputs like drafts. 4. **Description**, a few hyphenated words identifying the specific note. 5. **Version**, zero-padded, starting at `v01`. ### Step 2: Write the pattern ![Five-step flow showing how to assemble a research note file name from date, project code, note type, description and version.](/article-images/illustrations/file-naming-conventions-research-notes-build-pattern.webp) Assemble the elements in order, separated by underscores: ``` YYYY-MM-DD_PROJ_TYPE_description_v##.ext ``` A concrete example for a literature note on cognitive behavioral therapy for insomnia: ``` 2026-11-11_INS_lit_morin2015-cbt-review_v01.md ``` Read left to right, that name answers when, which project, what kind of note, what it covers, and which revision it is. ### Step 3: Create a test file and confirm the pattern works Open a terminal, move to a scratch folder, and create a file with your new pattern: ``` touch 2026-11-11_INS_lit_test-note_v01.md ls -lh ``` Expected output: ``` -rw-r--r-- 1 you staff 0B Nov 11 10:04 2026-11-11_INS_lit_test-note_v01.md ``` If the file appears exactly as typed, your pattern contains no characters your operating system objects to. ### Step 4: Add a timestamp option for rapid capture For fast research note taking, where you create many short notes in one session, a timestamp identifier is a common addition. It guarantees every note gets a unique name without any thought. ``` date +%Y%m%d%H%M ``` Expected output: ``` 202611111042 ``` Append that identifier to atomic notes, one idea per note, like `202611111042_ins-sleep-restriction-rationale.md`. This is the convention used in Zettelkasten-style note systems, and it works because no two notes created minutes apart can collide. ### Step 5: Name your data and reference files the same way Notes rarely live alone. Published studies give a sense of the data files that accumulate alongside them. A national online survey study analyzed data from 281 pharmacist respondents collected over eight months [4]. A multicenter [retrospective cohort study](/blog/guides/retrospective-cohort-studies-design-analysis-and-interpretation) analyzed arterial blood gas results from 798 critically ill obstetric patients across six hospitals spanning more than a decade [5]. Datasets like these, plus the scripts that process them, belong in the same naming system. A typical pattern: ``` YYYY-MM-DD_PROJ_data_source-description_v##.csv 2026-11-11_INS_data_redcap-export-baseline_v01.csv 2026-11-11_INS_data_clean-baseline-script_v01.py ``` ### Step 6: Document the convention in a README Write the pattern, the project codes, and the note-type vocabulary into a plain text file called `README.txt` or `README.md`, and place it at the top level of your notes folder. Anyone joining the project, including you in two years, reads that file first. Structured checklists like this are standard practice in research reproducibility. An anatomical validation study of a skull base surgical approach, for example, developed an explicit checklist from its dissections to ensure every case was executed the same way across 19 cadaveric specimens [6]. A naming README serves the identical function for your files. ### Step 7: Apply the convention to new files immediately From this point, every new note gets the pattern at creation. Build it into your workflow. If you keep daily capture notes, preface each with the date. If you write one literature note per paper, use the first author and year in the description element, for example `lit_carlson2019-microbiome-virome_v01.md`. When you revise a note, save as a new version rather than overwriting, so the sequence `v01`, `v02`, `v03` becomes an audit trail of your thinking. ## Summary Table of Naming Elements | Element | Purpose | Format | Example | |--|--|--|--| | Date | Chronological sorting, context | ISO 8601, `YYYY-MM-DD` | `2026-11-11` | | Timestamp ID | Unique names for rapid capture | `YYYYMMDDHHMM` | `202611111042` | | Project code | Grouping across folders | 2 to 5 uppercase letters | `INS` | | Note type | Fixed controlled vocabulary | lowercase, 3 to 5 letters | `lit`, `raw`, `data`, `out` | | Description | What the note covers | lowercase words joined by hyphens | `cbt-i-review` | | Version | Revision tracking | zero-padded, `v##` | `v01`, `v02` | | Separator | Between elements | underscore | `_` | | Word joiner | Inside descriptions | hyphen | `-` | ## Renaming Existing Files in Bulk Most people adopting a convention inherit a folder of badly named files. Bulk renaming solves this in minutes. ### Bash (macOS or Linux): prefix every file with a date ``` for f in *.docx; do mv "$f" "2026-11-11_INS_raw_${f}" done ls ``` Expected output: ``` 2026-11-11_INS_raw_meeting-notes-oct.docx 2026-11-11_INS_raw_phone-call-dr-lee.docx ``` Add `mv -i` instead of plain `mv` if you want the shell to prompt before overwriting any existing file. Plain `mv` overwrites silently, which is the fastest way to lose a note you forgot you had. ### PowerShell (Windows): the same operation ``` Get-ChildItem *.docx | Rename-Item -NewName { "2026-11-11_INS_raw_" + $_.Name } Get-ChildItem ``` Expected output: ``` 2026-11-11_INS_raw_meeting-notes-oct.docx 2026-11-11_INS_raw_phone-call-dr-lee.docx ``` ### Bash: replace spaces with underscores ``` for f in *" "*; do mv "$f" "$(echo "$f" | tr ' ' '_')" done ``` Run `ls` afterward to confirm every name is now space-free. Preview first with `echo "$f"` in place of the `mv` line if you want to see the planned changes before committing. ## Common Errors and Their Fixes **Windows illegal character error.** Typing a colon or slash into a file name in File Explorer produces the message `The filename, directory name, or volume label syntax is incorrect`. Fix by replacing the character with a hyphen. Colons appear most often when people paste timestamps formatted as `10:42` into names. **mv: cannot stat.** The full message reads `mv: cannot stat '2026-11-11_INS_lit_note.md': No such file or directory`. This means the file name you typed does not exactly match what is on disk. Run `ls` in the folder, copy the exact name, and check for a trailing space you cannot see on screen. **zsh: no matches found.** Running the bulk rename loop in zsh when the folder contains no `.docx` files gives `zsh: no matches found: *.docx`. Move to the correct folder with `cd`, or confirm the extension is lowercase and not `.DOCX`. **File name too long.** Linux refuses names beyond 255 bytes with the error `File name too long`. Windows fails a full path beyond 260 characters with a message about the specified path being too long. Fix by trimming the description element and moving depth out of the folder structure instead. **Silent overwrite.** There is no error message, which is the problem. A `mv` command whose destination already exists replaces it without warning. Use `mv -i` interactively, or `mv -n` to never overwrite, when renaming anything you care about. ## Applying the Convention to a Literature Notes System A [literature notes system](/blog/research-skills/the-ultimate-guide-to-building-a-literature-notes-system-for-research) is the organized collection of notes you write while reading papers. It benefits most from strict naming, because the volume grows fast. The bibliometric analysis cited earlier screened thousands of records over 20 years [1], and even a fraction of that reading flow produces a large notes folder. Use this pattern for literature notes: ``` YYYY-MM-DD_PROJ_lit_firstauthoryear-topic_v01.md ``` Two practical tips improve a literature notes system beyond the file names themselves. First, keep the description tied to the source, not your evolving opinion of it. `lit_bastien2014-cbt-components_v01.md` stays meaningful even after your interpretation changes. Version numbers carry the interpretation history. Second, record who produced a note and at what stage of understanding. The value of this shows in clinical education research. A survey of 526 hospital clinicians found a clear pattern across training levels, in which staff with no family violence training demonstrated their knowledge less often and with less specific detail, while trained clinical champions gave the most complete responses [7]. Notes written early in your training are simply less reliable than notes written later. Version numbers and dates let you see which understanding a given note reflects. ## How to Check That Your Convention Is Working Run these three checks monthly. 1. **Sort test.** Open the folder and sort by name. Files should appear in the exact order you would want to read them, chronological within each note type. Any file out of place reveals a formatting slip, usually a missing zero pad or a two-digit year. 2. **Search test.** Search the folder for your project code. Every project file should appear. Then search for a description keyword. It should surface the note you want within one page of results. 3. **Stranger test.** Hand the folder, without explanation, to a colleague or open it yourself after a week away. If you cannot tell what any file contains from its name alone, the description element is too terse. Structured verification of this kind mirrors practice in other fields. The endoscopic skull base study mentioned earlier confirmed its approach by checking that key landmarks, including the uncinate process, ethmoid bulla, and sphenoid sinus, were identified consistently across all 104 analyzed sides [6]. Your landmark equivalents are the date, the code, and the version. They must be present and correctly formatted on every single file. ## Common Mistakes and Limitations ![Six warning signs of poor file naming: ambiguous dates, word versions, mixed separators, special characters, broken references and no README.](/article-images/illustrations/file-naming-conventions-research-notes-naming-mistakes.webp) **Ambiguous dates.** `11-05-26` could be November 5 or May 11 depending on the reader. Always use `YYYY-MM-DD`. **Version words instead of version numbers.** Files named `draft`, `final`, `final2`, `final_REALLY` are the most common failure pattern in research folders. Numbers with zero padding remove the judgment call entirely. **Mixed separators.** Some files named with spaces, some with underscores, some with camelCase produces a folder that sorts unpredictably and breaks automation. Choose one style per element and enforce it. **Special characters that survive on one system and break on another.** A file named perfectly on a Mac can fail when copied to a Windows server if it contains a character such as `#` or `&`. Letters, digits, underscores, and hyphens only. **Renaming files that other things depend on.** If a manuscript, script, or collaborator references a file by its current name, renaming it breaks that reference. Update references in the same pass as the rename, or use versioned additions rather than replacements for shared files. **Treating naming as a substitute for real infrastructure.** A naming convention organizes files, but it does not back them up, track changes the way version control software does, or search inside file contents. Projects with serious data needs need backups and version control in addition to naming discipline. **Adopting a convention mid-project without a written record.** The README is not optional. Undocumented conventions decay, because each collaborator invents a private variation. Qualitative research shows why shared, explicit resources matter. A study of backyard poultry owners found that owners relied heavily on informal knowledge and had uneven engagement with available management resources [8]. Research teams fall into the same pattern when their file conventions live only in one person's head. Finally, conventions themselves have limits. No single pattern fits every field. Qualitative researchers managing interview transcripts need note types that a computational modeling project does not. Start with the five-element pattern here, adjust the vocabulary to your discipline, and record whatever you change. ## Frequently Asked Questions ### What is the best file naming convention for research notes? Use `YYYY-MM-DD_PROJ_TYPE_description_v##.ext`, for example `2026-11-11_INS_lit_morin2015-cbt-review_v01.md`. This pattern sorts chronologically, groups by project, identifies note type, and tracks versions, and it avoids every character that operating systems forbid. ### Should I use underscores or hyphens in file names? Use underscores between elements and hyphens between words within an element. That distinction lets scripts split a file name into its parts while keeping multi-word descriptions readable. ### How do I name different versions of a research note? Append a zero-padded version element, starting at `v01` and incrementing with each revision, for example `v01`, `v02`, `v03`. Never overwrite an earlier version, and never use words like `final`, which invite ambiguity. ### What characters should I avoid in file names? Avoid spaces and the characters `\ / : * ? " < > |`. These are forbidden on Windows or break paths on macOS and Linux. Stick to letters, numbers, underscores, and hyphens. ### How do I rename hundreds of existing research notes at once? Use a shell loop. On macOS or Linux, `for f in *.docx; do mv "$f" "2026-11-11_INS_raw_${f}"; done` prefixes every Word document with the new elements. On Windows, `Get-ChildItem *.docx | Rename-Item -NewName { "2026-11-11_INS_raw_" + $_.Name }` does the same in PowerShell. ### How long can a file name be? Individual file names can reach 255 bytes on most modern systems, but Windows has historically limited the full path to 260 characters. Keep your descriptive elements short and your folder structure shallow so deep paths stay under the limit. ### Why use YYYY-MM-DD dates instead of other date formats? The ISO 8601 format `YYYY-MM-DD` sorts correctly as plain text, so files arrange themselves chronologically in any file browser without extra tools. It is also unambiguous, since day-first and month-first formats vary between countries. ### Do file naming conventions help with reference managers and citation tools? Yes, indirectly. Reference managers store bibliographic metadata in their own databases, but the PDFs and reading notes you attach to them still need names you can recognize. Naming PDFs and literature notes with author, year, and topic keeps attachments matched to the correct reference records. ## Sources 1. [Global research landscape and emerging frontiers in primary insomnia treatment (2005-2025): A bibliometric analysis.](https://pubmed.ncbi.nlm.nih.gov/42994835/) 2. [Cochrane evaluation of (semi-)automated review methods: protocol for an adaptive platform study within reviews.](https://pubmed.ncbi.nlm.nih.gov/42320766/) 3. [Learning from experts: A self-improving LLM framework for study population generation in clinical research.](https://pubmed.ncbi.nlm.nih.gov/41197329/) 4. [Knowledge and Attitudes Regarding Pharmacogenetics of Pharmacists in the Netherlands: A National Online Survey Study.](https://pubmed.ncbi.nlm.nih.gov/41764771/) 5. [Acid-base imbalances in critically ill obstetric patients: A multicentre retrospective cohort study.](https://pubmed.ncbi.nlm.nih.gov/40639298/) 6. [Systematic anatomical validation of the endoscopic mononostril transethmoid-paraseptal approach to the central skull base.](https://pubmed.ncbi.nlm.nih.gov/42211323/) 7. [Impact of a 3-year transformational change project in family violence clinical response: Qualitative evaluation of the depth and breadth of knowledge in hospital clinicians by training level.](https://pubmed.ncbi.nlm.nih.gov/40014661/) 8. [Practices, Management and Resource Preferences of Backyard (Hobby) Poultry Owners in Central North Carolina: A Qualitative Study.](https://pubmed.ncbi.nlm.nih.gov/41677397/)