Blog/Online scams

How a pig butchering operation is actually run

Investment romance fraud is industrial work with shifts, scripts, quotas and software. Seen from the operator's side, the scheme's weak points become obvious.

CyberWatch AI21 November 2026 · 6 min read
Rows of desks with computer monitors in a large open office

Read enough accounts of long-form investment fraud and the same details keep appearing: the same opening messages, the same two-week pacing, the same screenshot of a trading dashboard, the same small successful withdrawal, the same tax demand at the end. Independent criminals do not converge on identical scripts. Employees working from the same manual do.

Almost everything written about this fraud is written for the person being targeted. This article looks at it from the other side, as a business with a staffing model, a software stack, a cost base and performance management. That view is useful for defense, because a victim sees a relationship while an operator sees a pipeline with measurable drop-off at each stage, and the stage boundaries are where the scheme shows itself.

The division of labor

The work splits into roles that rarely overlap.

Sourcing. Phone numbers, dating profiles, social media accounts and messaging handles arrive from purchased lists, data leaks, and sheer volume of cold contacts. Opening messages are deliberately banal, the wrong-number greeting or the misdirected friendly note, because the only job of the first message is to sort people into replies and non-replies. The consumer-facing version of this appears as a wrong number message that goes somewhere.

Conversation. Operators work shifts, each handling many targets at once from a shared record of who each target is, what they do, what they fear and what they want. Scripts cover the common turns of a conversation, and translation tools paper over language gaps. The persona is assembled from stolen photographs and a stable backstory: the same personas recur across cases, which is why romance scam personas read as familiar types rather than as people.

Conversion. Introducing the investment is a specialized step with its own staff, because this is where most targets are lost. The approach is almost never a pitch. It is a disclosure, offered reluctantly, about an uncle with inside knowledge or an arbitrage window the target happens to be trusted with. Group chats add social proof, the pattern covered in investment groups on WhatsApp and Telegram, where most of the enthusiastic members work for the operation.

Platform support. Once money is in, the target meets an account manager, a support agent, a compliance officer. These are the same operation, performing the institution that holds the money.

Payments and laundering. A separate function manages the receiving accounts, the recruited intermediaries described in money mule recruitment, and the movement of funds through exchanges and address chains.

Management. Quotas, commission, bonuses for a first deposit, penalties for idle operators, and a manager who steps in personally when a large balance stalls.

The labor behind the labor

Any honest account of this industry has to include how it is staffed. A substantial part of the workforce was recruited through fraudulent job advertisements for customer service or translation roles abroad, then held in compounds with documents confiscated, movement restricted, debts invented and violence used. United Nations bodies and labour organizations have documented this at a scale measured in hundreds of thousands of people across several countries.

Two things follow. The person on the other end of a scam conversation may be a victim of a different crime, which does not make the loss less real but does explain the mechanical, tired, slightly off quality of some exchanges. And the recruitment pipeline into those compounds runs on the same fake overseas job offers covered in fake overseas recruitment, which means the fraud recruits its own staff with the fraud next door.

The software

The fake platform is a product, not a craft project. Panel software is sold and rented to operations, which explains the uncanny similarity between apps in unrelated cases: the same chart library, the same deposit flow, the same support widget, the same wording in the withdrawal error.

What the administrator view controls is the whole trick. Displayed balances are editable. Price feeds can be real, simulated, or real with a thumb on the scale so that a target's first few trades win. Withdrawals can be allowed, delayed, or blocked with a message demanding a fee. Leverage can be offered to engineer a margin call that wipes an account and creates a reason to deposit more.

This is why the screenshots are so convincing and so worthless as evidence. A growing balance is a number in a database the fraudster owns. The same logic applies to the images discussed in fake profit screenshots: anything the platform displays is marketing.

The pipeline, stage by stage

  1. Contact. High volume, low cost, no investment mentioned.
  2. Warm-up. Days to weeks of ordinary conversation, usually moved quickly to a private messaging app where the operation controls the channel and no platform is monitoring it.
  3. Seeding. Wealth implied rather than described. Screenshots appear incidentally.
  4. Test deposit. A small amount, often a few hundred units of currency, treated internally as the single most important conversion event.
  5. Proof withdrawal. A real payout of a small sum. This is the budgeted expense that turns scepticism into trust.
  6. Scaling. Larger deposits, encouraged savings transfers, then borrowing, remortgaging and loans from family. Operators are measured on this stage.
  7. Harvest. Withdrawals blocked, followed by a sequence of fees, taxes and audits, each framed as the last obstacle. The mechanics are those of withdrawal fee scams, applied to a balance that never existed.
  8. Resale. The target's details, now marked as someone who paid, are passed on or sold to a recovery operation.

What the operator's view tells a defender

Reading the pipeline backwards gives better advice than any list of warning signs about the person's story.

The structure is the signal, not the content. An operation needs the conversation to move to a private channel, needs the investment to arrive as a confidence rather than an offer, and needs the platform to be one it controls. Those three steps are mandatory for them and all three are visible to you.

Verification has to happen outside the relationship. Every reassurance a target seeks from inside the conversation is answered by the operation. Checking whether a platform is licensed with a regulator, as in how to check an investment licence, is the one question they cannot answer for you.

A successful small withdrawal is evidence of nothing. It is the strongest sales tool in the pipeline and it is paid for out of expected returns.

Any fee to release funds marks the end, not a delay. By the time a tax or unlock payment is requested, the operation has decided no further deposits are coming.

After a loss, expect a second approach. A caller who knows the amount lost and the platform name is almost certainly reading an operator's own notes, which is the economics behind recovery scams.

Been pointed at a trading platform by someone you met online? Paste the link into CyberWatch AI Scan for a free check before you deposit anything.

Why the industrial view matters

Treating this as a scam committed by an individual leads to advice about judging character, which fails because the character is a product with a quality assurance process behind it. Treating it as an industry leads to advice about structure, which holds regardless of how convincing the person is.

If money has already moved, speed matters more than anything else, and the steps in recovering stolen crypto and reporting an investment scheme to the regulator are worth starting the same day. For the wider pattern of how these organizations operate across different fraud types, the online scams guide sets out the common machinery, and how organized scam operations work covers the same industrial logic in other product lines.

Frequently asked questions

What does pig butchering mean?

It is a translation of a phrase used inside the trade for a long-form investment fraud: build the relationship over weeks, get the target to invest on a platform the operation controls, let the balance grow on screen, then take everything. The food metaphor refers to fattening before slaughter.

Is one person doing all the messaging?

Usually not. The work is split by function, so a target may be handled by one person for the early conversation, another for the investment stage, and a third who appears as a platform agent or manager. Handovers are managed from shared notes about the target.

Who actually works in these operations?

A mix of willing employees and people trafficked into compounds under false job offers, held through confiscated documents, debt and violence. United Nations and labour bodies have documented hundreds of thousands of people in such conditions across parts of Southeast Asia.

Where does the fake trading platform come from?

Usually bought rather than built. Panel software with an administrator view that controls displayed balances, fake price feeds, withdrawal blocking and tax prompts is sold as a product to multiple operations, which is why the apps look alike across unrelated cases.

Why do small withdrawals sometimes work?

Because one successful withdrawal is the strongest sales tool the operation has. Paying out a small amount early converts a cautious target into a confident one and is budgeted as a cost of acquisition.

What are the tax and fee demands at the end?

The harvest phase. Once no further deposits are expected, invented fees, taxes, audits or unlock payments extract whatever remains, including borrowed money. Each demand is presented as the last one.

Does the money go straight to the operators?

No. It moves through mule accounts, over-the-counter brokers and chains of crypto addresses, with different teams handling conversion and laundering. That is why recovery depends on speed rather than negotiation.

What is the most reliable way to recognize one of these approaches?

The structure rather than the content. An unsolicited contact that becomes warm quickly, moves to a private app, introduces earnings as a personal secret, and ends on a platform only your contact could have recommended. Nothing about the person's story needs to be examined.

Sources

  1. UNODC Regional Office for Southeast Asia and the Pacific, United Nations Office on Drugs and Crime
  2. International Labour Organization, International Labour Organization
  3. Online fraud schemes, Europol
  4. 2024 Internet Crime Report, FBI Internet Crime Complaint Center
  5. Types of Fraud, US Securities and Exchange Commission, Investor.gov
  6. Investment scams, Moneysmart, Australian Securities and Investments Commission
  7. Investor Alerts Portal, International Organization of Securities Commissions
  8. Internet Crime Complaint Center, FBI
  9. Advisories and Articles, US Commodity Futures Trading Commission
Share this article
LinkedIn WhatsApp X
For organizations

Your people get these messages at work too.

CyberWatch AI sends your staff realistic practice attacks, trains the gaps it finds, and shows management exactly where the organization stands.