# PDF Naming Convention for Research: Step-by-Step Guide By the end of this guide you will have a working PDF naming convention for research, a folder structure to match, and the exact commands to rename an existing pile of downloaded papers on Windows, macOS, or Linux in a few minutes. You will also know how to hand off renaming work to reference management software so new papers organize themselves. You need nothing installed to start. To run the batch commands in the walkthrough you need one of the following: - Windows PowerShell, which is built into Windows 10 and 11 - A terminal on macOS or Linux (bash or zsh) - Python 3, if you want the most flexible renaming script If you plan to automate renaming with a citation manager, having Zotero, Mendeley, or EndNote installed helps, but the core convention works with plain files in plain folders. ## Why a PDF Naming Convention for Research Matters A research library grows one download at a time. Six months in, the folder contains files like `1-s2.0-S0140673623012345X-main.pdf`, `1661-Article Text-4702-1-10-20230214.pdf`, and `pmc8901234.pdf`. None of those names tells you what the paper is about. Worse, two journals can produce identical generated filenames for different papers, so one file silently overwrites another. A consistent filename solves three problems at once. 1. **Findability.** You can search your folder with the operating system's search box and find a paper by author, year, or topic without opening anything. 2. **Sorting.** Files sort chronologically when the date leads the name, which matches how literature reviews are usually written. 3. **Safety.** Unique names prevent accidental overwrites when two downloads collide. The convention also pays off outside your file manager. When you export a formatted bibliography, attach PDFs to a manuscript submission, or share a folder with a collaborator, names that already contain author, year, and title survive every transfer intact. ## The Anatomy of a Good Research PDF Filename ![Flow of filename fields from year through author and short title to optional type and version, ending in the pdf extension.](/article-images/illustrations/pdf-naming-convention-for-research-filename-anatomy.webp) Most workable conventions use the same small set of elements. You do not need all of them. Pick three to five and apply them to every single file. | Element | Purpose | Example | Recommendation | |--|--|--|--| | Date | Sorts files chronologically | `2023` or `2023-06-14` | Use the four-digit publication year at minimum. Use full ISO 8601 dates (YYYY-MM-DD) when publication month matters | | Author | Identifies the paper in citation lists | `Okafor` | Last name of the first author. Add a second author only when first-author collisions exist in your library | | Title | Makes the file readable without opening it | `CRISPRofftargetReview` | Shortened to the meaningful core, three to six words, capital letters at word boundaries | | Document type | Separates reviews from primary data | `review` or `RCT` | Optional. Useful for literature reviews that mix study designs | | Version | Tracks drafts, preprints, and final versions | `v2` or `preprint` | Required only when you hold multiple versions of the same paper | | Identifier | Guarantees global uniqueness | `10.1111-12345` | Optional. Use when you manage hundreds of papers with the same author and year | A practical template looks like this: ``` YYYY_FirstAuthor_ShortTitle[_type][_vN].pdf ``` Anything in brackets is optional. Here are three files named with the template: ``` 2023_Okafor_CRISPRofftargetReview.pdf 2023_Okafor_CRISPRofftargetReview_preprint.pdf 2019_Lindqvist_MicrobiomeRCT.pdf ``` ### Separators, Case, and Characters ![Decision path for choosing underscores between fields, hyphens within fields, and avoiding characters Windows forbids.](/article-images/illustrations/pdf-naming-convention-for-research-separator-choice.webp) Two conventions cover almost every situation researchers encounter. **Use underscores between fields and hyphens within fields.** Underscores act as column boundaries when you scan a folder listing. Hyphens join words inside one field, for example `2023-06-14` and `Lindqvist-Nilsson`. This makes a filename machine-parseable later, which matters when you write a rename script. **Use capital letters at word boundaries instead of spaces.** `CRISPRofftargetReview` reads better than `crisprofftargetreview` and avoids the quoting problems that spaces cause in command-line work. Spaces are legal in filenames on every modern operating system, but every shell command and script that touches your files has to handle them with quotes, and some older research software and pipelines still stumble on them. **Avoid the characters Windows forbids.** Windows filenames cannot contain the following nine characters: ``` \ / : * ? " < > | ``` A DOI copied raw into a filename contains slashes, so replace them with hyphens. macOS allows colons at the file system level but the Finder converts them to dashes, and colons are the path separator in classic Mac contexts, so skip colons there too. **Keep names under roughly 60 characters before the extension.** Long names get truncated in file lists and preview panes. Windows also enforces a maximum total path length (folder plus filename), so deeply nested folders plus verbose filenames can hit that ceiling and refuse to save. ## Choosing Your Convention: Four Decisions Before renaming anything, write your convention down. A convention lives in a short text file at the top of your literature folder. Make these four decisions. **Decision 1: Which elements to include.** For most graduate students and individual researchers, `Year_FirstAuthor_ShortTitle` covers daily needs. Labs and systematic review teams usually add version and identifier fields because multiple people touch the same files. **Decision 2: Date precision.** Publication year alone sorts cleanly and keeps names short. Full dates matter when you track preprints that later get published, since the preprint and the final paper share a year but not a date. Pick one and stay with it. **Decision 3: What to do with papers with no clear first author.** Use the organization abbreviation, for example `WHO` or `NASEM`, or the word `Anon`. Decide now so you never improvise mid-folder. **Decision 4: Where renamed files live.** A convention without a folder structure still produces one giant scrolling list. A common pattern is one folder per project or review, sometimes split by year. Your naming convention handles sorting inside a folder, and the folder tree handles the project level. Record your decisions in a `README.txt` inside the top folder. Anyone who inherits your library, including your future self, reads that file first. ## Step-by-Step: Applying a PDF Naming Convention for Research This is the core walkthrough. Follow it in order with your own files. ### Step 1. Create the folder structure and README Make a top folder for your library and one subfolder per project. On any platform: ``` Literature/ README.txt project-bladder-cancer/ project-copd-cessation/ ``` Write your convention into the README in three lines: the template, the separator rules, and the no-author rule. ### Step 2. Rename files one by one the first time Before automating anything, rename five or ten files by hand. This exposes edge cases you will need to handle in bulk: supplementary files that share a parent paper's name, duplicated downloads, and publisher filenames with no useful metadata. Manual renaming is `F2` on Windows, `Enter` on macOS, or `F2` in most Linux file managers. Rename one supplementary file like this: ``` Before: 41598_2023_34567_MOESM1_ESM.pdf After: 2023_Okafor_CRISPRofftargetReview_supplement.pdf ``` The `_supplement` suffix keeps it adjacent to the main paper in a sorted listing. ### Step 3. Batch rename on Windows with PowerShell Open PowerShell in your downloads or literature folder (type `powershell` in the File Explorer address bar) and add a year prefix to every PDF, which is often the fastest first pass on a chaotic folder: ```powershell Get-ChildItem -Path "C:\Users\you\Downloads" -Filter "*.pdf" | Rename-Item -NewName { "2023_" + $_.Name } ``` Expected result, checked with `Get-ChildItem -Filter "*.pdf" | Select-Object Name`: ``` 2023_1-s2.0-S0140673623012345X-main.pdf 2023_1661-Article Text-4702-1-10-20230214.pdf 2023_pmc8901234.pdf ``` To strip a common publisher prefix instead of adding a prefix, use a regular expression with the `-replace` operator: ```powershell Get-ChildItem -Filter "*.pdf" | Rename-Item -NewName { $_.Name -replace "^1-s2\.0-S[0-9A-Z]+-", "" } ``` Expected result: ``` Before: 1-s2.0-S0140673623012345X-main.pdf After: 0140673623012345X-main.pdf ``` The `^` anchors the pattern to the start of the name so only the prefix is removed. The backslash before the period escapes it, because a bare period in a regular expression matches any character. ### Step 4. Batch rename on macOS and Linux with bash The equivalent loop on macOS or Linux: ```bash cd ~/Downloads for f in *.pdf; do mv "$f" "2023_$f" done ``` Note the double quotes around `$f`. They are required because your publisher filenames contain spaces and hyphens. Without the quotes the shell splits each name at spaces and `mv` receives the wrong number of arguments. To see what will happen before anything moves, replace `mv` with `echo mv` and run the loop. You get a dry-run listing: ``` echo mv 1-s2.0-S0140673623012345X-main.pdf 2023_1-s2.0-S0140673623012345X-main.pdf ``` Confirm the listing looks right, remove the `echo`, and run it for real. ### Step 5. Systematic renaming with a Python script Once files carry a consistent structure, a Python script can reorganize them into your template. The following script matches filenames of the form `Author_Year_Title.pdf` and rewrites them as `Year_Author_Title.pdf`, with a cleaned title field: ```python import os import re folder = "C:/Users/you/Literature/project-bladder-cancer" pattern = re.compile(r"^(.+?)_(\d{4})_(.+)\.pdf$") for name in os.listdir(folder): match = pattern.match(name) if not match: print("SKIP (no match):", name) continue author, year, title = match.groups() slug = re.sub(r"[^A-Za-z0-9]", "", title)[:60] new_name = f"{year}_{author}_{slug}.pdf" src = os.path.join(folder, name) dst = os.path.join(folder, new_name) if os.path.exists(dst): print("SKIP (exists):", name) continue os.rename(src, dst) print(f"{name} -> {new_name}") ``` Expected output: ``` Okafor_2023_CRISPR off-target review.pdf -> 2023_Okafor_CRISPROffTargetReview.pdf SKIP (no match): FullThesis_Chapter4.pdf SKIP (exists): Lindqvist_2019_MicrobiomeRCT.pdf ``` Three design choices in this script prevent classic disasters. Files that do not match the pattern are skipped and reported, never touched. Destination collisions are detected with `os.path.exists` before the rename. Titles are truncated at 60 characters so the total path stays manageable. ### Step 6. Hand routine renaming to reference management software Manual and scripted renaming cover the backlog. For papers you download from today onward, reference management software renames each PDF automatically the moment you import it, using metadata pulled from the publisher, PubMed, or the DOI. ### Step 7. Verify the result Run the same checks regardless of which method you used. 1. Sort the folder by name and confirm chronological order. Files should group by year with the year folder or year prefix. 2. Search the folder for the characters Windows forbids. If any survive, rename them. 3. Count files before and after. The numbers must match exactly. A rename loop that misses files usually reports them, so scan the script output for `SKIP` lines. 4. Open three random files and confirm the filename describes the content accurately. A clean but wrong name is worse than a messy right one. ## Renaming Methods Compared | Method | Best for | Speed | Reversibility | Skill needed | |--|--|--|--|--| | Manual rename (F2) | Under 50 files, edge cases | Slow | Easy | None | | PowerShell one-liners | Windows bulk prefixes and strip jobs | Fast | Medium (keep a log) | Low | | Bash loop | macOS and Linux bulk work | Fast | Medium (test with `echo` first) | Low | | Python script | Structural rewrites, repeated use | Fast | High if you script checks | Moderate | | Reference management software | All future downloads | Instant per import | High (manager stores originals) | Low once configured | ## Using Reference Management Software to Enforce the Convention A naming convention you apply by hand competes with your own deadlines. Reference management software applies it for you. Every major citation manager stores a library database, attaches PDFs, retrieves metadata from a DOI, PMID, or web page, and can rename imported files according to a pattern you set once. Here is a compact citation manager comparison focused on PDF handling, not full feature sets. | Software | Cost | Automatic rename on import | Metadata retrieval | Good fit for | |--|--|--|--|--| | Zotero | Free, open source | Yes, with customizable patterns | DOI, PMID, arXiv, web pages | Students and most researchers, especially those who value open formats | | Mendeley | Free (Elsevier account) | Yes, pattern-based | DOI, web importer | Researchers embedded in the Elsevier ecosystem | | EndNote | Paid license | Yes | DOI, PubMed, web of science records | Institutions with site licenses and heavy Word manuscript workflows | | Papers | Paid | Yes | DOI, publisher search | Researchers who prioritize a reading-first interface | Zotero deserves a specific note because of how its renaming works. Recent Zotero versions let you define a filename template using fields such as author, year, and title, with rules for how many authors to include and how to truncate long titles. Once set, every PDF you drag in, save from the browser connector, or import by DOI gets a consistent name without any action from you. The manager also keeps its own copy of the file inside its storage folder, which means the library database and the readable filename point at the same document even if you later move the attachment. Mendeley and EndNote follow the same pattern with pattern editors of their own. The practical guidance is identical across all three: configure the rename pattern before importing your first paper of the new era, then let the software handle the stream. If a manager names a file in a way you dislike, fix the pattern, then re-trigger renaming from the manager rather than renaming files behind its back. Citation managers track attachments by internal references rather than names in most cases, but files you rename externally can still desynchronize from the database entry, which shows up as broken links to attachments. A hybrid workflow covers both the backlog and the future: 1. Import the existing folder into your citation manager and let it match PDFs to metadata records. 2. Use the manager's own rename function to normalize names in bulk where possible. 3. Handle the files the manager cannot match, typically scanned book chapters and supplementary material, with the scripted approach from Step 5. 4. From then on, every new paper enters through the manager and inherits the convention automatically. This is where a naming convention and a citation manager reinforce each other. The convention gives you readable files you can email, drop into a shared drive, or attach to a manuscript. The manager gives you the bibliography, the notes, and the duplicate detection. Neither replaces the other. ## Common Mistakes and Limitations These are the errors that show up repeatedly when researchers build a PDF naming system. Each one has a concrete fix. **Renaming with a script before backing up.** A bad regular expression can mangle two hundred filenames in two seconds. Copy the folder, or commit it to version control if you use git, before any bulk operation. The cost is one `cp -r` command. The alternative is reconstructing names from a folder of unidentifiable PDFs. **Renaming the file while the citation manager holds it open.** If Zotero, Mendeley, or EndNote is running, renaming its storage files externally can desynchronize the database. Close the manager, or better, rename from inside the manager. **Assuming the publication year equals the download year.** Stamp files with the publication year, not the year you found them. A folder full of `2024_` prefixed papers from across two decades destroys chronological sorting, the main reason to lead with a date. **Putting spaces and special characters in the convention itself.** A convention that includes `&`, `(`, or spaces works fine on your desktop and breaks the day a collaborator syncs the folder to a lab server or a data repository that sanitizes names. Stick to letters, digits, underscores, and hyphens. **Encoding meaning in capitalization alone.** Two fields separated only by a case change, like `OkaforCRISPRofftargetReview`, are unparseable by script and easy to misread. Keep an explicit separator between every field. **Ignoring supplementary files.** Journals deliver supplementary material with generated names that collide with the main paper. Add a role suffix such as `_supplement`, `_SI`, or `_protocol` so the pair sorts together and never collides. **Skipping the README.** Conventions drift. The single highest-value habit is writing the template down and updating it when you change it. Six months from now you will not remember whether you decided that no-author papers get `Anon` or the publisher abbreviation unless the README says so. **Expecting the filename to carry the full citation.** A filename is an index, not a reference. Do not try to squeeze journal names, volume numbers, and page ranges into sixty characters. Store the full citation in the citation manager and keep the filename lean. If you need the identifier, add the DOI with slashes replaced by hyphens, but recognize the tradeoff in name length. **One limitation to state plainly:** no naming convention replaces reading and judging each paper. Systematic review teams use standardized screening, risk-of-bias tools such as ROBINS-I and GRADE, and [structured data](/blog/data-analysis/what-is-structured-data) extraction to evaluate the evidence they collect, and individual study questions always require reading the paper itself rather than trusting its label. ## Frequently Asked Questions ### What is a PDF naming convention for research? A PDF naming convention for research is a fixed template, such as `YYYY_FirstAuthor_ShortTitle.pdf`, that you apply to every paper you save. It makes your library sortable, searchable, and safe from filename collisions without opening any file. ### Should I use underscores or hyphens in PDF filenames? Use underscores between fields and hyphens within fields. Underscores mark the boundaries between author, year, and title, while hyphens join words or date parts inside a single field. ### How do I rename hundreds of PDFs at once? Use a batch command. On Windows, a PowerShell loop with `Get-ChildItem` and `Rename-Item` handles the whole folder. On macOS or Linux, a bash `for` loop with `mv` does the same. For structural rewrites, the Python script in this guide matches a pattern and skips anything it does not recognize. ### Does Zotero rename PDFs automatically? Yes. Zotero renames imported attachments according to a filename template you configure, drawing author, year, and title from the metadata record it retrieves for the paper. Mendeley and EndNote offer equivalent pattern-based renaming. ### What date format should I use in research filenames? Use the four-digit ISO 8601 year, `2024`, at minimum, and the full `2024-06-14` form when you need to distinguish preprints from their published versions. Four-digit years are nonnegotiable because two-digit years sort incorrectly. ### How long should a research PDF filename be? Keep the part before `.pdf` under roughly 60 characters. Longer names truncate in file listings and, combined with deep folder paths, can approach the maximum path length Windows enforces. ### Should I include the DOI in the filename? Usually no. The DOI is long, contains slashes you must replace, and duplicates information your citation manager already stores. Add it only when you manage very large libraries where author and year collide often, and replace the slashes with hyphens. ### What characters are not allowed in PDF filenames? On Windows you cannot use `\ / : * ? " < > |`. macOS and Linux are more permissive, but sticking to letters, digits, underscores, and hyphens keeps files portable across every system and server you will ever use. ## Further Reading - [A Systematic Review of the Efficacy and Safety of Outpatient Bladder Tumour Ablation.](https://pubmed.ncbi.nlm.nih.gov/33602641/) - [Smoking cessation for patients with chronic obstructive pulmonary disease (COPD): an evidence-based analysis.](https://pubmed.ncbi.nlm.nih.gov/23074432/) - [Internet-based device-assisted remote monitoring of cardiovascular implantable electronic devices: an evidence-based analysis.](https://pubmed.ncbi.nlm.nih.gov/23074419/) - [Pharmacist-Led Self-management Interventions to Improve Diabetes Outcomes. A Systematic Literature Review and Meta-Analysis.](https://pubmed.ncbi.nlm.nih.gov/29311916/) - [Community-based care for the management of type 2 diabetes: an evidence-based analysis.](https://pubmed.ncbi.nlm.nih.gov/23074528/) - [Behavioural interventions for type 2 diabetes: an evidence-based analysis.](https://pubmed.ncbi.nlm.nih.gov/23074526/)