Why Zotero in Google Docs Made Me Build My Own Citation System

We used Google Docs with Zotero for my review article. And I want to be clear upfront: this isn't really Zotero's fault.

Zotero is great software. It's free. It's open source. It has a huge community. The desktop application works well. I'm sure the Zotero team has done their best to make the Google Docs integration work as well as possible.

But Google Docs has limitations with extensions and add-ons. That's just the way the world works. You don't control the platform. You work within its constraints. And sometimes those constraints create problems.

The Setup Hurdle

Before you can even start, you need to set up Zotero and make it work with Google Docs. I honestly don't remember the exact process anymore, but I remember it wasn't trivial.

You have to enable add-ons in Google Docs. You have to connect Zotero. You have to log in to your Zotero account. You need the desktop application installed. The add-on communicates with the desktop app somehow.

And if you ever need to work from a different computer? Good luck if Zotero isn't installed there with all the right configuration. Or at least you need to be logged into your account. I think it works through the account, but I'm not entirely sure even now.

It's not impossible to set up. But it's friction before you even start writing.

The Lag Creeps Up

Here's the insidious part: in the beginning, when you add only a few citations, it works fine. It's actually great. You add a citation, it appears, everything is smooth.

The problem is that as you continue to add publications and articles, the lag starts to show up. It creeps up on you. You're not aware that this problem will exist when you start.

By the time you figure it out, you're already too deep. You've already written big parts of your draft. You've already moved all your citations into Zotero's database. And then it just makes sense to continue, even though things are not great anymore.

Every time I added a citation, the whole Google Docs interface would slow down. Sometimes it would freeze entirely. Sometimes it just stopped working and I had to refresh and do it all over again.

Why Not LaTeX?

Yes, I was aware of LaTeX. It would have been better for citations. More reliable. More stable.

But the thing is, we used Google Docs because of the real-time collaboration features. I think real-time collaboration is very important when writing with others. You can see what your co-authors are doing. You can work on the same document simultaneously. It's a huge advantage.

At some point I thought about switching to LaTeX. And honestly, I probably could have made it work. I could have assigned each co-author a small part of the article to work on separately. I could have just copied the text back and forth. There are ways to make LaTeX work with collaboration.

So I continued using Zotero's citations in Google Docs, thinking to myself "why didn't I use LaTeX?" And yes, collaboration was part of the answer. But the real answer? I was already too deep. Too committed to the current solution. I had already invested so much time and effort into this workflow that starting over felt impossible, even though the workflow was frustrating.

Random Unlinking

Sometimes Zotero just unlinks all your citations. For no apparent reason. You don't know when it's going to happen. There's no way to prevent it.

Maybe there's an error message when it happens - I'm not sure. But even if there is, it doesn't help. The damage is done. Your citations are just plain text now. They're no longer connected to Zotero. You can't update them. You can't change styles. They're just dead text.

Living in Backup Fear

Because of this, I was backing up constantly. Before adding new citations - backup. Before major edits - backup. Every few hours - backup. I was backing up like crazy and restoring when needed.

I was pretty tech-savvy, so I found ways to mitigate the damage when things went wrong. I don't think I ever had to relink everything from scratch. But you end up with your Google Docs version history looking like it has 50,000 backups. I'm exaggerating, but you get the point.

You don't want your work to go to waste. And you don't want to risk having to relink everything again.

The Relinking Process Is Tedious

Because the relinking process itself is already very tedious even under normal circumstances.

You have to look at the author, the title, the year. You have to always be sure you're selecting the right article. You start typing or copy-pasting the title into the Zotero citation search thing. Then you hope it shows the article synced to your account.

Doing this for one citation is fine. Doing this for an entire document's worth of citations because something broke? That's a nightmare.

My Friends' Horror Story

Some friends of mine were writing an article. They're all physicians. One day before their publication deadline, all the citations got unlinked.

They had to stay up all night to relink them. Every single citation. Manually going through the document, finding each reference, relinking it to the correct Zotero entry.

And they all had to work the next day. As physicians. Seeing patients. After an all-nighter. Not because they were finishing the intellectual work, but because a tool broke.

That was a really bad experience for them. I'm glad I never had it that bad. But I knew the danger. I lived with that fear throughout my writing process.

You Shouldn't Have to Fear Your Tools

This is what really bothered me. The constant background worry that your work might be destroyed by a tool malfunction.

Tools are supposed to help you. They're supposed to make your work easier. You shouldn't have to fear your tools.

Time Lost on Repetitive Tasks

When you write a scientific publication, you're doing it for free. You're doing it for science, for humanity. You don't get paid for this work.

And here I was, losing time to manual labor. Repetitive tasks that don't offer anything intellectually. Work that a monkey could do, but I had to do it because I chose to write an article for science.

Yes, reliability matters. I need my citations to be correct. But the cost of achieving that reliability through manual backup-and-restore workflows? The cost of constantly working around tool limitations? That's time and energy I should have spent on the actual science.

So I Built Something Better

When I built LitRevs, citations were built-in from the start. Not a plugin. Not an extension that has to communicate with a desktop application through some complex setup.

Citations that don't randomly unlink. Citations that work reliably every time. And real-time collaboration too - because I learned from my Google Docs experience that collaboration matters.

No more fear. No more constant backups. No more lag that creeps up on you.

If This Sounds Familiar

If you've ever watched the lag get worse as your document grew, or lived in fear of citations unlinking, or spent hours setting up integrations that could break at any moment, or continued with a bad workflow because you were already too deep to switch - you're not alone.

This was one of the major frustrations that made me build LitRevs. You deserve tools that work.