We Built the Troll Farms: How the Philippines Engineered Disinformation

I’ve seen them online for years.

They were never hard to spot. The same phrases. The same insults. The same copy-pasted outrage jumping from one post to another. If you’ve been on social media in the Philippines any time after 2016, you’ve probably seen them too. Maybe you argued with them. Maybe you muted them. Maybe you even thought they were real people defending a cause.

So did I.

But I never really stopped to ask the bigger question—Who are they?
Not just the face behind the comment, but the system behind the pattern.
Who pays them? How do they organize? Why do they sound the same—yet always slightly different?

I guess I got used to them. Like background noise in a room you live in.
But the more I started reading, the more I realized—this wasn’t noise. It was machinery.
Messaging calendars. Payment schemes. Identity assignments. Political funding. Foreign money.
Real wages for fake identities.

We didn’t just get trolled.
We built the troll army. We trained it. We normalized it. And we taught the rest of the world how it’s done.

This isn’t a tell-all. I’m not an insider. I’m not even late to the story.
I just never thought to look at it this closely before.

Until now.

So here it is—the face of someone I’ve known online for almost a decade.
I just didn’t know their job description until today.

Where It Started: A Campaign That Couldn't Afford Billboards, So It Built a Digital Army

In 2016, Rodrigo Duterte’s presidential campaign faced a problem: they didn’t have the kind of money that bought prime-time ad spots or highway billboards. Traditional media was out of reach. So they turned to something cheaper, faster, and—at the time—less regulated.

They turned to social media.

It wasn’t an accident. It wasn’t a grassroots movement that just happened to go viral. It was strategy.

Nic Gabunada, a former ABS-CBN executive turned campaign mastermind, led the shift.
“When we realised we didn't have money for TV, radio, print, billboards etc,” he said in an interview,
“we made the decision to tap up the social media groups.”
(BBC)

And tap them up they did.

Duterte later admitted that his campaign had spent around ₱10 million—roughly $200,000—on social media operations. That figure matched what a University of Oxford study uncovered around the same time: a deliberate investment in digital manpower.
(Philstar, Mashable)

But this wasn’t just about running Facebook ads.

Gabunada assembled a network of 400 to 500 operators, assigned by region and demographics. It wasn’t unlike managing a brand campaign—except the product was a president, and the message was designed to dominate. There were content creators, sharers, fake account managers, and people in charge of stirring engagement. Trolls, but organized.

It ran on shifts.

During the day, the locals posted. At night, Filipino workers abroad—OFWs working across time zones—kept the engine running.
“Late at night, the people from abroad, the workers in a different time zone, took over,” Gabunada explained.
(BBC)

It was clever. Cheap. Effective. And very hard to stop once it worked.

By the time the campaign ended, it was clear:
They didn’t just win the online war.
They built a system that could outlive the campaign.

And it did.

What Happens When a Campaign Never Shuts Down

Most political campaigns end.
The rallies stop. The volunteers go home. The posters fade under new layers of dust and grime.

But this one didn’t.

When Duterte won in 2016, the machine that helped get him elected didn’t pack up and disappear. It evolved. What started as a campaign tool turned into a permanent feature of Philippine politics—quietly crossing over from election season tactic to full-time governance strategy.

It was subtle at first. The same pages and personalities that pushed for his candidacy simply kept going—now shifting their focus to defending policies, discrediting critics, and keeping the president’s popularity artificially inflated.

Then it became official.

Bloggers who had spent months attacking the opposition online were suddenly invited to press briefings. Some were granted media credentials, others were given positions in the government’s communications team.
(The Diplomat)

At some point, it was no longer clear where the troll farms ended and the government began.

The Presidential Communications Operations Office (PCOO), the official mouthpiece of the administration, began echoing the same messages the trolls were spreading—sometimes even quoting them directly.
It was propaganda, dressed as public relations.
And it didn’t bother hiding anymore.

According to Freedom House, the Philippines became one of the first democracies where online manipulation wasn’t just present—it was institutionalized.
(Freedom House via The Diplomat)
The same trolls who boosted Duterte during the campaign were now amplifying support for his most controversial policies—particularly the war on drugs, which human rights groups around the world had condemned.

But it didn’t stop there.

The messaging shifted with the moment. From “DDS loyalty” to “bash the opposition,” to “discredit mainstream media,” to “red-tag journalists.”
Whatever the threat, the machine adapted.
But the structure? That stayed the same.

Same networks. Same roles. Same tactics.
Only now, it wasn’t just a campaign.
It was government policy—run not through legislation, but through likes, shares, and digital intimidation.

And most of us kept scrolling, never quite realizing the campaign never really ended.

It Was Never Just Random Comments. It Was a Job.

The word “troll” makes you think of someone chaotic. Unhinged. Probably jobless, scrolling in the dark and picking fights for fun.

But that’s not what I found.

What I found was structure.

Every troll operation—at least the ones connected to Duterte’s campaign—followed a chain of command. There was a client, usually a politician or party. That client would work with a PR strategist, who then passed the plan to a content editor, who managed troll team leaders, who in turn handled dozens of workers.

It was clean. Functional. Almost corporate.
Like a call center—but instead of handling irate customers, they handled political enemies.

During the 2016 campaign, Nic Gabunada ran it like a digital ad agency. He assigned coordinators by region. One group focused on Luzon, another on Mindanao, and a special team was set up just for OFWs. They weren’t voters, but they were vocal—and they worked in time zones that let the operation run 24/7.
(BBC)

That’s when I realized something:
The internet never sleeps. And neither did this machine.

Each troll managed anywhere from 50 to 200 fake accounts.
And these weren’t just burner profiles—they had identities.

You didn’t just create a fake name and spam posts. You were assigned a character.

Someone played a teacher—so their tone had to sound educated and rational.
Someone else played a mother—so their posts focused on safety, emotion, family.
Another played a student—sharp, reactive, always defending “real change.”
(GMA News)

Trolls weren’t told to go wild.
They were told to stay in character.

Some were “white trolls”—the polite ones. They'd act reasonable, ask loaded questions, sound like they were just trying to understand.
Others were “black trolls”—the attack dogs. They flooded comment sections with insults, slurs, and threats.
Same troll farm. Different mask.

And they almost never knew who they were working for.

One troll, interviewed by Channel News Asia, said she didn’t know who her handler reported to, and she didn’t care. She got her talking points. She hit her daily quota. That was the job.
(CNA)

It was designed to be compartmentalized—on purpose.
So that if the whole thing got exposed, no one could point up the chain with certainty.

The thing is, once you understand how it works, you start seeing it everywhere.
You scroll past a “concerned mother” talking about how much safer the country feels now, and you wonder—is that a person? Or a persona?

And you start to realize:
It was never just noise.
It was a job.

Trolling Wasn’t Just a Tactic. It Was an Industry.

At some point, you have to ask the question nobody wants to ask out loud:

How much does a troll actually make?

It turns out—more than most honest jobs in this country.

An entry-level troll could earn around ₱30,000 a month. That’s nearly double the national average wage in many provinces. For team leaders or those handling dozens of accounts, the number could go up to ₱50,000 or even ₱70,000.
(GMA News)

Some were paid per engagement.
One dollar per post. Sometimes more, depending on the assignment.
(LA Times)

And if you had influence—if you were a vlogger, a TikTok ranter, or a Facebook live regular—you didn’t even need to hide it.
₱30,000 per post wasn’t unheard of for creators who could move a narrative.

The money wasn’t coming from thin air.

Rodrigo Duterte himself admitted that his campaign in 2016 spent ₱10 million on social media ops. That figure was echoed by an Oxford University study analyzing the rise of organized trolling in the Philippines.
(Philstar, Mashable)

But the budget didn’t stop with the campaign.

In the years that followed, government departments like the Department of Finance and the Department of Foreign Affairs awarded contracts to “public relations consultants” who had already been flagged by platforms like Facebook for running fake account networks.
(Inquirer)

And then there’s the foreign money.

In 2025, Senator Francis Tolentino presented evidence showing that the Chinese Embassy paid ₱930,000 to a local agency called Infinitus Marketing for what was basically a troll farm service agreement.
(Senate, ABS-CBN)

And here’s the part that makes it even more slippery:
These deals didn’t come with the label “troll operation.”
They came dressed as digital strategy.
Communications consulting.
Reputation management.

Which brings us to the quiet question lurking behind all of this:
Is any of this taxed?

The short answer is: no one knows.
Some of these contracts are formal—filed, documented, probably taxed like any legal service. But many others? Cash transfers, GCash, prepaid SIMs, burner phones, anonymous handlers. There’s no paper trail. No receipts. Just a post quota and a paycheck.
(Inquirer Opinion)

So yes, trolling was a tactic. But for thousands of people across the country, it was also a job.

A system.
An industry.
And for some, a career that paid more than decency ever could.

Flood the Feed, Follow the Script

If you ever thought trolls just made it up as they went along, you’d be wrong.

They weren’t guessing.
They were following a script.

Every week, troll teams received a “message of the week.” It came from above—from the campaign strategist, the PR head, or whoever was running the operation. Sometimes it was as simple as: “Focus on discrediting this journalist.” Other times, it was a narrative push: “Frame the war on drugs as necessary.”
(BBC)

And once the message was handed down, the troll network went to work.

But they didn’t all post the same thing.
That’s what made it effective.

Each troll had a persona, and that persona determined how the message would sound.

The “teacher” accounts used academic tone—polished grammar, subtle persuasion.
The “mother” accounts leaned on emotion—talking about safety, protection, and love for country.
The “student” accounts were all energy—buzzwords, slang, and optimism about “real change.”

Same message.
Different voices.
All scripted.

They weren’t just told what to say—they were told how to say it.

And they had targets.

One troll interviewed in 2022 said she was expected to share at least 150 posts a day. Her job wasn’t to convince people one by one—it was to flood the feed, overwhelm the timeline, and make the message impossible to ignore.
(CNA)

These weren’t loose guidelines.
Every troll had to report their work—through spreadsheets, screenshots, or engagement tallies. They were monitored like employees, evaluated on visibility, not nuance.

And the tactics?

• Flooding hashtags to dominate algorithmic space
  
  

• Spamming comments to derail critical discussions
  
  

• Disinformation: fabricated quotes, fake news articles, doctored screenshots
  
  

• Personal attacks: mocking journalists, threatening activists, swarming anyone who dared question the official line
  
  

• Red-tagging: labeling dissenters as communists or terrorists—an accusation that, in the Philippines, can put someone in danger
  (D+C Journal, The Diplomat)
  
  

But the most dangerous part wasn’t the volume.

It was the goal:
To distort public reality.

This wasn’t debate. It was psychological warfare.

Trolls were taught how to gaslight on a national scale. They made people feel outnumbered. They took real-world criticism and buried it under manufactured consensus. They erased nuance until there were only two sides: the loyal and the enemy.
(Scientific Journal)

And if you weren’t sure which one you belonged to, the feed would make that decision for you.

All They Needed Was a SIM Card and a Spreadsheet

I’d already read about the fake accounts, the personas, the weekly talking points.
I knew about the quotas, the handlers, the comment flooding.

But what struck me most—once the structure became clear—was how simple it all was.

No fancy software. No backend dashboard. No AI.

Just prepaid SIMs, cheap phones, a spreadsheet, and a quiet place to post.

Before the SIM card registration law, you could walk into any store and buy a number—no ID, no questions asked. That number became your gateway to a new persona. Multiply that by fifty, sometimes two hundred, and suddenly one person could be an entire comment section.

And this wasn’t random posting.
There were trackers—Google Sheets, Excel files—listing what needed to be posted, where, and when. Screenshots were submitted. Quotas were enforced. The handler wasn’t just watching what you said—they were checking if you said it enough.

Coordinating all this didn’t take a war room.
It took Telegram, Viber, Facebook Messenger, and sometimes just a private Facebook group.
Everything dropped there: the message of the week, sample captions, hashtags, attack targets.

And when it was done? The payment came through.

GCash. Bank transfers. Or straight-up cash, for those with no digital access.
No contracts. No receipts. No company name. Just proof of work and a handler’s go-ahead.

Some troll farms used post schedulers to mimic consistency. Others had engagement pods—circles of trolls liking and sharing each other’s posts to boost visibility.

It was efficient. Affordable. Almost boring in its design.

That’s what makes it hard to dismiss.
This wasn’t some dark, mysterious operation.
It looked like any other group chat, any other freelancer tracking deliverables, any other gig.

But what they were delivering wasn’t design or writing.
It was distortion.

And for many, that distortion paid better than the truth ever did.

The Platforms Saw It, Too. They Just Didn’t Stop It in Time.

It’s tempting to believe trolls worked in the shadows—quiet, invisible, slipping past the system.

But that’s not what happened.

The platforms saw them.
They just didn’t stop them.
Or maybe they just didn’t care enough to do it when it would’ve actually mattered.

Facebook, for instance, eventually took down a network of fake accounts linked to Philippine police and military personnel. They called it “coordinated inauthentic behavior,” which is just a polite way of saying it was all fake from the start.
(CNA)

These pages were used to defend government policies, harass critics, and manufacture the illusion of public support.

By the time Facebook removed them in 2020, they had already done their job.

They planted confusion.
They attacked journalists.
They made people question what was real.

And Facebook? They published a transparency report. No apology. No accountability. Just a log of what they took down after it had already shaped public perception.

YouTube wasn’t much better.
Troll-produced content did well—not because it was well-made, but because it was angry. Outragey titles, clickbait thumbnails, threats disguised as “commentary.” These videos got recommended over and over again.

Why?
Because the algorithm rewards what people engage with.
And trolls know exactly what to post to get reactions.

Twitter—now X—let it all run wild.
Trolls hijacked hashtags, flooded replies, baited arguments, and got boosted for being divisive. No one cared if the post was true—only if it got traction.

None of these platforms had to hire the trolls.
They just had to reward the content.
And that was enough.

Because the algorithm doesn’t care if something is honest.
It cares if people stop scrolling.
And if outrage does the job, then outrage gets to stay.

Even now, the trolling hasn’t stopped. It just evolved.
It hides behind meme pages.
It dresses up as satire.
Sometimes it even looks like a grassroots community.

The only major platform that’s slightly harder for trolls to exploit is Threads—but not because of moderation.
It’s because Threads is tied to an Instagram account, which makes tracing a fake identity easier.
Trolls can’t disappear as easily when their IG profile is just a few taps away. You can follow the trail. You can see when something feels off.

That kind of visibility makes them nervous.
Which is probably why most of them are still camping out on Facebook, X, and YouTube—platforms that know what’s happening and keep the doors open anyway.

And maybe, that’s the most honest thing about all of this.
The platforms didn’t build the troll farms.
But they built the stage—and handed them the microphone.

We Didn’t Just Fall for It. We Were the Perfect Test Market.

People like to say Filipinos are resilient.
But online? We’re reactive.

We laugh fast. We argue fast. We share fast.
And most of all, we click fast—faster than we verify.

That’s not an insult. That’s just the internet we grew up with.

We’ve always been one of the world’s most active social media users.
Facebook isn’t just an app here—it’s the comment section of the entire country.
It’s where we get our news, our fights, our gossip, our grief.
(YouTube)

So when trolls started flooding the feed, they didn’t need to force their way in.
We were already online.
We were already scrolling.

And for a lot of people, what they saw wasn’t manipulation.
It was just another post. Another meme. Another hot take that seemed to say what they already felt.

But let’s not pretend this was just about belief.
For thousands of people, trolling wasn’t a cause.
It was a job.

₱30,000 a month to copy-paste opinions and follow a script.
That’s twice what many honest jobs pay—and a hundred times easier than any work that requires thinking for yourself.

When you have bills to pay, loyalty becomes a currency.
And trolling pays on time.

But the real fuel wasn’t money. It was culture.

We’ve been trained to avoid conflict.
To choose peace over questions.
To defer to people in power, even if they lie.
Pakikisama taught us to keep the harmony.
Utang na loob taught us to be loyal to those who helped us, even if they’re wrong.
And smart shaming made sure we never listened too long to anyone who tried to explain why it all felt off.

So when the trolls came—loud, certain, aggressive—it was easier to just laugh, nod, and scroll.

It didn’t help that many of us had already stopped trusting traditional media.
After years of political division, fact-checking started to feel like side-taking.
Even when the facts were right.

And while journalism slowed down to confirm, troll content flew.
Anger travels faster than truth.
So does a joke.
So does a perfectly timed post that confirms what someone wants to believe.

We didn’t just fall for it.
We were primed for it.

Social media made it visible.
Our conditions made it work.

Even Trolls Get Tired

We talk a lot about what trolls do to people.
But rarely about what trolling does to them.

Not everyone stayed in the system.

Some quietly left after the campaign.
No grand exit. No apology post. Just silence. The kind that feels like someone slowly backing out of a room they were never supposed to be in.

They didn’t all leave because they found a conscience.
Some were burned out.
Some couldn’t take the stress.
Some just couldn’t keep up with the quotas anymore.

One troll, interviewed in 2022, said she felt numb by the end of it. She couldn’t tell which opinions were hers and which ones were assigned. Another admitted to regretting posts that may have led to real-world threats—“I just posted what they told me to. But sometimes I wonder what happened to the people we targeted.”
(CNA)

Others weren’t worried about guilt.
They were worried about getting caught.

A few changed sides politically.
Some started working for opposition accounts.
And that’s when the paranoia kicked in.

On Reddit, you’ll find anonymous posts from former trolls asking how to scrub old content. How to wipe digital fingerprints. How to disappear before someone traces them back to a post they were paid ₱1 to share.
(Reddit – Buhay Digital)

Some were doxxed by rival troll groups. Others were quietly blacklisted or cut off. A few received real threats—screenshots shared behind closed doors, names leaked in private groups, numbers passed around like warnings.

Most of them didn’t speak out—not out of loyalty, but out of fear. Explaining what you did and why you did it is harder than just moving on. Forgetting was easier.

But leaving the job doesn’t mean leaving what it taught you.

Some still scroll with suspicion. They watch comment sections with old reflexes. They trust less, post less, and wonder if they’re seeing the same machinery they used to be part of—only this time, from the outside.

Logging out is easy.
But learning to feel clean again? That takes longer.

Not Everyone Scrolled Past

It’s easy to feel like the trolls won.
They were louder. Faster. Funded.
And for a while, it seemed like they controlled every thread, every trending topic, every reaction under every post that dared to ask better questions.

But not everyone stayed quiet.

While trolls were flooding timelines, a handful of journalists, researchers, and ordinary users tried to keep something grounded online. They didn’t have quotas or handlers. They didn’t get paid by the post. But they showed up anyway.

Rappler, Vera Files, Tsek.ph—they made fact-checking a daily job.
Sometimes hourly.
Sometimes thankless.

They verified fake quotes. Debunked viral lies. Posted clarifications no one wanted to hear.
And while their corrections didn’t go half as viral as the lie that came before, they stayed.
Quiet. Consistent. Unmoved.
(Rappler)

In 2022, Rappler helped launch #FactsFirstPH, a coalition of newsrooms, NGOs, and advocacy groups working to clean up the digital space before the elections. It was one of the most coordinated fact-checking efforts we’ve seen in the country. Most people never heard of it.
That doesn’t make it any less important.

Meanwhile, in classrooms and community halls, some teachers and non-profits started teaching digital literacy.
How to check sources.
How to spot fake engagement.
How to tell the difference between an opinion and a manipulation.

It wasn’t part of any national curriculum.
It didn’t have a big budget.
But it reached students, barangay workers, and parents who wanted to understand what was happening to their feeds—and to their families.

Not everyone fighting back was part of a newsroom.
Some were just people in the comment sections who refused to be drowned out.
They dropped links. Posted receipts. Called out the trolls by name, even when it meant getting dogpiled in return.

But pushing back had its cost.

Maria Ressa didn’t just get arrested for journalism—she got harassed daily by troll accounts.
Fake memes. Edited videos. Coordinated attacks.
Sometimes from real people.
Sometimes from someone paid to pretend.

The resistance was never equal in size.
It was underfunded.
Understaffed.
Often ignored.
But it kept going.

And maybe that’s the most stubborn, Filipino thing of all:
Trolls had budgets.
But the truth still had people.

We Thought It Was Just Us. It Was Never Just Us.

For years, it felt like we were alone in this.

Like the trolls were our burden.
Our problem.
Our weird, broken little corner of the internet.

But we weren’t just the target.

We were the model.

Disinformation researchers around the world have been studying the Philippines—not just as a case of digital chaos, but as a working prototype.
Not a fluke. Not an accident.
A playbook.
(GIGA Focus, The Diplomat)

What happened here didn’t look like what Russia did with its state-run troll farms.
It wasn’t as ideological as India’s IT Cell or as militarized as Myanmar’s information war.

What we had was something else entirely:

• Low-budget but high-impact

• Decentralized, informal, and believable

• Run by people who looked like us, sounded like us, and spoke like they care

It worked because it didn’t look like propaganda.
It looked like the feed.

And when it worked here, others paid attention.

The tactics used in our comment sections—paid influencers, fake personas, red-tagging, hashtag flooding—eventually showed up in other countries too.
Brazil. Indonesia. Myanmar. Even the U.S.

It turns out you don’t need bots when you have a country full of people trying to earn a living online.

What started as a campaign tool turned into an exportable framework.
And while we were busy surviving the arguments and memes, the rest of the world was quietly watching what we built—taking notes, testing the same methods elsewhere.

We didn’t know we were early.
We didn’t know we were building the rough draft of something global.

All we knew was that our feed was noisy, angry, and hard to trust.

But now that the noise has spread, the silence around it feels even heavier.

Because this wasn’t just about us anymore.

It never was.

The Face Behind the Comment

I didn’t write this to break a story.
Trolling in the Philippines has been around since 2016. Most of us knew. We saw it, scrolled past it, argued with it, maybe even shared a few posts without thinking twice.

But for almost a decade, I never bothered to ask the bigger question:
How did this become so normal?

What I found wasn’t chaos. It was coordination.
Budgets, roles, schedules, talking points, handlers, performance reports.
This wasn’t some spontaneous army of passionate supporters.
It was a machine.

A machine that was built here—scaled here—and, in many ways, perfected here before anyone else realized it was even possible.

We didn’t just get manipulated.
We created a system that showed the rest of the world how cheap, fast, and effective digital control could be.
It didn’t require bots or dark money.
Just SIM cards. Messaging apps. A Google Sheet. And someone willing to play a part.

And maybe what’s more absurd is that some people did it for free.
While thousands were being paid monthly salaries to post fake support and harass critics, there were also the laughably loyal DDS trolls who joined the pile-on without ever seeing a peso.
They were volunteers in a propaganda machine that didn’t even need them—because the professionals were already hitting quota.
They called it loyalty.
The industry called it free engagement.

So no—this blog isn’t a takedown.
It’s a record. A map. A quiet reminder that when a comment sounds off, it probably is.

And that the person posting it may not be some random stranger with strong opinions.
They might just be working.
Filling a quota.
Saying what they were paid to say.

Or worse—saying it for free.