The Best Reference Manager Setup
(Zotero + ZotFile + Cloud Storage)
I wrote this tutorial because after years of using Mendeley (and then Zotero with default
settings) I figured out a system for reference management that is extremely smooth, efficient,
and free. It has improved both my research and writing process so much that I wanted to share
it with others.
The key is to use Zotero with Zotfile (an amazing plugin), and a cloud-based storage service, in
addition to a few minor changes to Zotero’s settings. By the end of this tutorial you’ll be able to:
• Download, rename, and organize academic papers, news articles, web sites, and other
reference material with one click.
• Have virtually unlimited storage for your PDFs with automatic back up.
• Extract highlighted text and annotations from PDFs and associate them with references
• Search and insert references within a Word document and automatically insert a
reference list in the required formatting style of virtually any journal.
• Create multiple libraries that you can share with collaborators and/or the public
• Free and open source
• Powerful and flexible
• Active user community
Although this tutorial is written from a Windows perspective, all the essential steps will work on
a Mac. Some things in the screenshots and configuration options might be slightly different
from what is pictured in the tutorial, but it shouldn’t be too difficult to figure out.
A note for experienced Zotero users:
Although this tutorial is designed for people who are starting with Zotero from scratch, I
promise that you will get something out of it even if you have been using Zotero as your
reference manager for some time.
December 3, 2018
PART 0: Switching to Zotero
Many people may be in a situation where they have already invested a lot of time and money
into a proprietary reference manager such as EndNote or Mendeley. I used Mendeley for a
couple of years before it was bought by Elsevier, who have since moved toward commercializing
As far as I know, there are three options for migrating your existing reference library over to
1) Exporting your references from EndNote or Mendeley into Zotero is very easy. There are
instructions for exporting from EndNote (here and here) and Mendeley (here and here).
2) Exporting your references + PDFs might be a little more work, but shouldn’t be too
3) Exporting your references, PDFs, subfolders, and annotations is more challenging, but
definitely possible (I know because I did it myself!). I think Menotexport is probably the
best way to do this from Mendeley, but there are other options out there as well.
a. Note that as of June 16, 2018, the Beta version of Zotero includes the ability to
import directly from Mendeley. However, Mendeley Dekstop 1.19 has moved
toward encrypting their database and PDFs, so if you have already updated, you
will have to follow the workaround in the above link to revert to an older version.
Even then, your PDFs may be “locked” for editing outside of Mendeley.
4) When doing a large import, Zotero may try to automatically find the associated
metadata for each paper via Google Scholar. However, after a couple hundred of
requests, Google will suspect you are a bot and automatically put up a CAPTCHA, which
may slow or stop the import process. To avoid this, simply disable your internet
connection during the bulk import; you can always go back and tell Zotero to look up
citation information for specific files later.
Finally, although I’m confident in this setup, I recommend going through the rest of the tutorial
and trying out Zotero before doing any migrations, just to make sure you like it.
PART 1: Install Zotero Standalone and the Zotero Connector
1) Download and install Zotero Standalone AND the Zotero Connector
2) Turn off automatic screenshots (optional)
a. By default, Zotero will make a copy of the web page and attach it to the record in
your Zotero library. Some people like this feature, but I prefer to only attach
b. To turn this feature off, go to your add on manager and select:
Options Advanced Config Editor
c. Change the “Value” for automaticSnapshots to “false” (double click on it)
PART 2: Install a cloud storage service
1) Choose your service
a. This could be Google Drive or Dropbox that you already have an account with.
b. If you’re looking for a secure and private alternative to Google, I highly
c. If you want as much free space as possible, I recommend MEGA. It’s free,
encrypted, and gives you 25GB of space (which is a LOT of PDFs).
2) Download the relevant desktop sync app (e.g., MEGASync).
3) Create a new folder in the syncing folder on your deskotp. This is where your PDFs will
be stored, so you might want to give it an appropriate name like “Zotero” or
“Zotero_Library” (see below).
PART 3: Install and configure Zotfile
Zotfile is an excellent add-on for Zotero that helps to manage your PDFs. It can automatically
rename, organize, and extract annotations.
1) Visit the Zotfile site and download the xpi file (save as…in a folder you can find)
2) In Zotero Standalone, go to Tools Add-ons
3) Click the gear icon and click “Install add-on from file”
4) Navigate to and select the .xpi file for ZotFile that you downloaded
5) ZotFile should be installed, but you may need to restart Zotero
Setting up PDF syncing
Once Zotfile is installed, go into Zotero Standalone, select “Tools” from the top menu bar, and
click ZotFile preferences (you may need to restart Zotero first).
1) Go to the General Settings tab (see below)
2) Set Source Folder for Attaching New Files to a desired folder
a. When you add a reference to Zotero, Zotfile will look to this folder for new PDFs
to associate it with
b. Your “Downloads” folder is a good option (this is what I use)
c. Some people might prefer to use a separate folder as holding tank (e.g. a folder
on the desktop called “Papers for Zotero”)
3) Set Location of Files to “Custom Location” and point it to the folder you created in your
cloud syncing app in Part 2.
If you leave Use subfolder defined by unchecked, you will get a single folder on the sync drive
with all of your PDFs in it.
Alternatively, you can check this box to instruct ZotFile to organize your PDFs into subfolders by
author name, year, journal name, etc. (see the ZotFile website for more info).
Personally, I find it more useful to leave it unchecked so all of my PDFs are put in one big folder
on the sync drive (below).
I then make subfolders within Zotero Standalone, since this is where I spend most of my time
navigating. Note: If you like subfolders, you may want to review the “Other Tips” part of this
tutorial for instructions on setting up Recursive Folders.
Set up renaming rules
I strongly recommend instructing ZotFile to automatically rename your PDFs. There are many
options for this, but here are the settings I use as an example:
Stopping automatic syncing to Zotero Cloud
When you make a Zotero account on Zotero.org and start to sync your Zotero Standalone
library with the Zotero cloud, it will default to uploading your PDFs along with your references.
However, we want to store the PDFs in “My Library” to a separate cloud service using ZotFile, so
we need to change some settings before starting to sync:
1) Open the Zotero Preferences
2) Navigate to the Sync tab
3) Under File Syncing, uncheck “Sync attachment files in My Library using Zotero”
This is the primary setting that you need to change in order to have your references sync to the
Zotero cloud, while your PDF attachments sync to your personal cloud service.
As you can see from the screenshot, there are a number of other sync options that you can
adjust. To learn more about syncing with Zotero, I suggest reading the documentation (here).
Relative paths (optional)
If you intend to sync across multiple devices, such as a work computer and a home laptop, you
will need to set up relative paths in the “Linked Attachment Base Directory”.
Go to Zotero preferences Advanced Files and Folders and set Base Directory: to the folder
you created in your cloud syncing app in Part 2. You will have to do this on each of the
computers you use.
Make sure to keep everything else in this window at default settings. In other words, keep the
“Data Directory Location” as the Default.
If you need it, you can find more information about this here and here.
PART 4: Install other add-ons to Zotero Standalone
I highly recommend installing a Zotero Word Processor plugin (either for MS Word or Libre
The following plugins are not required for the overarching setup to work but they will make your
life easier and improve the overall experience.
• Zotero Storage Scanner
• Google Scholar Citations for Zotero
• Zotero QuickLook
o This should work automatically with Mac
o On Windows, you will have to install QuickLook beforehand
For any of these, just visit the plugins page and follow the steps to download and install the
respective .xpi files like you did for ZotFile.
There are many other plugins that I have yet to try, including ones that enable integration with
Google Docs and LaTeX.
I also highly recommend Unpaywall. This is not a Zotero plugin, but is added to your browser
and will automatically search for open access PDFs on every journal web page you visit. If it
finds an open access PDF, Zotero will usually grab it when you click the button.
PART 5: Using Zotero
You now have everything you need to start managing your references with Zotero. Zotero has
many built-in features and there are many resources online for learning about these. I’m
therefore going to focus on some of the particularities of this specific setup. Let’s start with an
Suppose you come across a paper that you would like to download. Let’s use this PLOS ONE
paper on antimicrobial resistance in foxes as an example.
The first step is to open the Zotero Standalone application. Once Zotero is open, we can go
back to the browser and notice that the icon for the Zotero Connector (usually in the top-right
corner of the browser window) has changed to look like a manuscript (see below).
This indicates that Zotero recognizes that there is a reference on the web page, and that it is a
If we were to click this button, Zotero would go to work grabbing the reference information
from the page and looking for an open access PDF to download. It would then appear in our
However, I want to be more organized and direct this reference to a subfolder (or “Collection”),
not just my general library. So before I click the button, I’m going to go back to Zotero
Standalone and click the button to create a new Collection. I’ll name it “Foxes”.
Now go back to the browser and click the Zotero button.
Watch and wait in wonder as the Zotero Connector grabs the reference and the PDF, and saves
it to the “Foxes” collection in Zotero Standalone (below).
If you switch back to Zotero Standalone you should see the reference (red arrow) and the link
to the attached PDF (blue arrow). The column on the right shows the detailed metadata.
Meanwhile, in the background, ZotFile has automatically taken the PDF, renamed it, and moved
it into the sync folder for your cloud service:
References vs. Attachments
This is an important moment to note that Zotero makes a distinction between the reference (or
parent item) and the file(s) associated with it. If you wanted to, you could have a whole Zotero
Library composed of references, with no PDFs at all. You can also have PDFs in Zotero that have
no associated reference.
Continuing with our example, let’s say you find this book on foxes. The PDF is not available
from the website, but you download the reference with the Zotero Connector to your “Foxes”
As you can see below, Zotero has the Book reference, but no attachment. The PLOS ONE paper,
however, has a blue dot, showing it has an attachment.
Double-clicking on a reference will open the associated file. If the reference doesn’t have a PDF
attachment, Zotero will open the website where you found the reference (try it with the book on
However, if the reference does have a PDF attachment, Zotero will open it in the default PDF
viewer (I highly recommend PDF X-Change Editor for Windows users).
I like to annotate PDFs with highlights or comments. For example, open the “Mo et al.” paper
and highlight the first sentence of the abstract:
Now save the changes to the PDF, and go back to Zotero Standalone.
Right-click on the reference, and go to Manage Attachments
You should now see a note attached to the reference, and the extracted annotation in the right-
Clicking the blue link will take you to the relevant page in the PDF.
Click the green arrow in the top right corner to sync your references to your Zotero Cloud
(you will have to make an account first). Remember, this will only sync the references, not the
PDFs. The latter are being synced by your personal cloud service (e.g., MEGA).
PART 6: Deleting
There are a number of different options for deleting, and they’re a little tricky with this setup.
Usually, to delete an entry, you just right click and select “Move Item to Trash”.
However, because we are storing the PDFs outside of Zotero, moving items to the trash will only
delete the reference, not the PDF. Therefore, I recommend deleting the PDF first, then deleting
the reference in Zotero.
1) To delete the PDF, right-click on the link to the attachment and choose “Show File”. This
will open up your storage folder in a new Window where you can delete the PDF as you
would any other file on your computer.
2) Now that the PDF is deleted, go to Zotero, right-click on the reference, and select “Move
Item to Trash”. Alternatively, you can type Ctrl+Delete.
PART 7: Lookup Engines
This isn’t required for the setup to work, but it’s so useful that I’ve given it its own section.
If you look to the top right of Zotero Standalone, you’ll see a “Locate” button . Clicking this
gives you different options for opening whichever reference is currently selected in your library.
This can be useful for locating the original website or URL or the associated record in your
institution’s library catalogue. However, these defaults are just the tip of the iceberg.
1) Follow the instructions under "Managing Search Engines" on this webpage,
2) Navigate to your Zotero Data Directory. Mine looks like this:
3) Delete the default "engines.json" file and replace it with the one that I’ve provided here.
4) Close and restart Zotero.
5) Select a reference from your library and then click the “Lookup” button.
6) You should see a variety of new options to choose from, including Google Scholar, Web
of Science, and Sci-Hub. These can be extremely useful for quickly locating reference
information and/or PDFs from the internet.
Note: Not all search options will always be available. It will depend on what type of
reference it is, and whether it has a DOI listed in your Zotero library.
PART 8: Other tips
You now have the basics of downloading and managing references in Zotero! There are many
other neat functions available in Zotero, and the best way to learn is to start exploring the
program. The Zotero documentation and forum are also great places to start looking for tips
and tricks. Here are a few functions/options that I find particularly useful:
Other ways to bring a reference into Zotero
1) Drag and drop a PDF into Zotero Standalone. Zotero will attempt to find the correct
metadata and create a reference for it.
2) Sometimes this doesn’t work, and you can right-click and say Retrieve metadata for PDF.
3) Perhaps you have a DOI for an article but don’t have a PDF. You can click the tool at
the top of the Zotero Standalone window, and paste a
DOI or other type of identifier into it. If Zotero can find a
matching record online, it will create a new reference in
If you happen to acquire the relevant PDF at a later date:
a) Put the PDF in the “Downloads” folder on your computer (or whichever folder you
selected as the Source Folder for Attaching New Files in ZotFile settings in PART 3).
b) Go to Zotero Standalone and right-click on the reference.
c) Select Add attachment from source folder (right).
d) ZotFile will grab the PDF, rename it, move it to your sync
folder, and associate the reference with it in Zotero, just as if
you had clicked the Zotero Connector button in your
If you like it so that selecting a Collection shows all the references in that Collection and its Sub-
collections then you need to switch on “recursive collections.” This is the default behavior in
Mendeley, but is not the default for Zotero, and can be easily changed by the following steps:
1. Open Zotero Preferences
2. Go to the “Advanced” tab
3. Click Config Editor (it may give you a warning, just ignore it)
4. Search for “recursive” in the search bar
5. Change extensions.zotero.recursiveCollections so that Value = true (double-click on
the word “false” to change it)
There are many ways of searching your Zotero library. It’s worth reading through the search
documentation to get a sense of what’s possible.
Consider setting up an OpenURL resolver to help identify references with the “Lookup” button.
This can be very useful, especially if your institution or organization has a library that gives you
access to journals. Here are instructions.
Deleting PubMed entries
Some journals, such as Science automatically download a PubMed metadata attachment to the
reference. This can be annoying because it will seem like the reference has a PDF attached, when
in fact it’s just the PubMed entry.
I suggest creating a saved search which will identify any references with a PubMed attachment.
Once it’s set up, simply check the search once in a while, and delete the PubMed entries. Here
If you sign up for table of contents alerts from your favorite journals, you might consider
sending them straight to Zotero via an RSS feed. Here is how to do it.
Groups are incredibly useful when working with collaborators. A couple of notes about groups:
• Each of your group libraries is distinct from each other and from your personal library
• You can easily drag references from one group to the other
• If you followed the setup in this tutorial, your library PDFs will be stored and synced
in your cloud service. However, your group PDFs will be synced on Zotero’s cloud,
which gives you 300 MB for free, and larger storage plans for an annual fee.