Skip to content
DM11AI TRUST & IT RISK PROTECTION
AI TrustAI governance, ISO/IEC 42001 and AI risk
Governance, Risk & CompliancePCI DSS, ISO 27001, Central Bank, SOC 2 and LGPD
CybersecurityPentesting, SAST/DAST, phishing, cloud and forensics
Security OfficeCISO, DPO, Ethical Hacker and DevSecOps as a Service
Business ContinuityResilience, BIA, crisis management and Cyber Antifrágil®
ProductsStandardsServicesProposals
Self-assessmentsFree assessments with the result on screen, no sign-up required
ComparisonsISO 27001 vs SOC 2, LGPD vs GDPR, pentest vs vulnerability assessment
FAQA hundred answers on standards, deadlines and scope
Case Studies42 stories by segment and signed testimonials
About UsContact
PTES
Talk to an expert

IATA-accredited agencies

Without compliance, IATA restricts your customer's card.

It is not your accreditation at stake, it is the payment method behind most of your sales. DM11 prepares your agency to prove it meets PCI DSS, the security standard for card data, from designing the scope to the evidence that supports the declaration.

Take the self-assessmentTalk to a specialist

DM11 prepares your agency. The self-assessment and attestation are signed by the agency itself. And where the case calls for a formal assessment by a QSA (an accredited auditor), it is carried out by a partner.

Who runs the preparation

  • 17 years in information security and compliance
  • PCI DSS projects in retail, hospitality and payments
  • Data protection and privacy specialists
  • Experience with bank and Big Four audits

What IATA actually requires

The requirement is not about accreditation. It is about the card.

It is widely repeated that every accredited agency must be PCI DSS compliant. The rule is more precise than that, and the difference changes the size of the problem. IATA, the airlines' global association, only clears an agency to take payment with the customer's card if it meets the standard, through its Resolution 812. An agency that does not accept, transmit or store card data tells IATA exactly that and the matter ends there.

Where the rule comes from

Resolution 812, section 2.6.3.1: permission to use the Customer Card Payment Method (paying with the customer's card) depends on the agent's full compliance with PCI DSS. The BSP Manual for Agents, the manual that governs agents in IATA's billing and settlement system, reinforces this, adds Resolution 890, and states that IATA monitors compliance.

Does it apply to you?

It applies if your agency accepts cards against its own merchant agreement or issues tickets using the passenger's card through the BSP. If your operation never touches card data, there is a specific declaration for that case, and it replaces the compliance evidence.

Where you declare it

In the IATA Customer Portal, under Accreditation & Changes, in the PCI DSS compliance update service. Only contacts with administrator or authorised signatory profiles can submit. You attach the attestation, and the cycle is annual.

Without demonstrating compliance

  • Immediate restriction on using the customer's card to issue tickets
  • A Risk Event recorded against your history with IATA
  • Airlines notified that compliance was not evidenced
  • Risk of a change in Risk Status, which pulls in financial guarantee requirements
  • Exposure to card brand penalties after a breach, charged through your acquirer (the company that processes your card payments)

With compliance in place

  • The customer's card available as a payment method, without anxiety
  • A predictable annual cycle instead of a scramble at every request
  • Reduced scope: fewer systems and fewer people handling cards
  • Organized evidence, which also answers corporate client questionnaires
  • Less surface for fraud, the industry's silent cost

The standard

What PCI DSS asks of an agency

PCI DSS is the card industry's security standard, maintained by the PCI Security Standards Council. It does not ask how big your company is: it asks whether you accept, transmit or store card data. In an agency, that data comes in through more doors than most people expect.

The version that applies today

PCI DSS v4.0.1. Version 3.2.1 was retired in March 2024 and v4.0 in December 2024, so any current assessment is against 4.0.1. Documents and policies written against the old version no longer support a declaration.

The clock has already turned

Most of the new requirements in version 4 were best practice until 31 March 2025 and became mandatory on that date. Among them: multi-factor authentication for all access to the card environment, longer passwords, and training that covers phishing and social engineering.

The verification code can never be stored

The BSP Manual for Agents is explicit: agents must never store or write down the card verification value. The standard says the same: authentication data cannot be retained after authorisation, and encrypting it does not help. It is the most violated rule and the easiest to prove in an investigation.

Cards over messaging are prohibited

The standard prohibits sending an unprotected card number over end-user messaging technologies, which includes email, SMS and messaging apps. In practice this is an agency's most common and most costly habit, because it leaves data scattered across inboxes nobody controls.

Scope is the decision that saves most

Everything that stores, processes or transmits card data is in scope, and so is whatever connects to those systems. That is why the first job is shrinking the territory, before buying any control: taking data out of where it does not need to be reduces effort across every requirement at once.

Call recordings count

If your call center records calls and customers read out card numbers, that audio now contains card data. Where recordings can be searched, this conflicts with the prohibition on retaining authentication data. The usual answer is to pause recording during payment or strip that segment.

Which case is yours

The path depends on how the card enters your agency

The SAQ is the self-assessment: a questionnaire the agency answers and signs itself. There is more than one type, and the right one depends on the channel through which you receive the card, not on your revenue. Choosing the wrong type is the costliest mistake, because it produces a declaration that does not hold up.

How the card reaches youApplicable self-assessmentWhat that demands
Your own site redirecting payment to the gateway (the payment processor's page), never touching the dataSAQ AThe smallest set of controls. In return, you attest that your payment page is not susceptible to script attacks, a statement that only holds up with real monitoring of that page.
Your own site with a payment page that touches the data before sending itSAQ A-EPConsiderably broader than SAQ A: your web environment enters scope, with vulnerability management and change control.
Only the acquirer's virtual terminal, on an isolated computer, storing nothing digitallySAQ C-VTRequires genuine isolation of the workstation and no other electronic card channel. It does not apply if the agency also sells online.
Card typed into the form of payment field in the reservation systemSAQ DThe data passes through your systems and the booking record, so the environment is in scope. This is the most common scenario among Brazilian agencies, and the most underestimated.
Card received by phone, email or messaging appSAQ DBeyond the wider scope, there is an earlier correction to make: receiving cards over messaging already breaches the standard and needs an alternative channel before any declaration.
Card written down, kept in a spreadsheet, back-office system or call recordingSAQ DStorage is what most expands scope and risk. Here the work starts by finding and removing data from where it should not be.

It is often written that travel agencies fall under SAQ C-VT. In most Brazilian cases they do not: C-VT requires the acquirer's virtual terminal on an isolated machine and does not allow online sales. An agency typing cards into the reservation system is in a different scenario, and declaring the wrong type creates more exposure than not declaring at all.

What actually happens

IATA does not fine you. It restricts you.

Many people picture losing accreditation or paying a fine, and put the matter off. What Resolution 812 provides for is different, and in the short term worse for operations: the customer's card is removed as a payment method and the agency's record is marked.

  1. 01

    A Risk Event is recorded

    Non-compliance is a formally listed risk event in Resolution 812, in the same table that covers the agency's other risk events.

  2. 02

    Immediate card restriction

    The Resolution states that IATA will immediately restrict use of the Customer Card Payment Method. If the ticketing system provider cannot apply the restriction, the BSP airlines are informed so they can act.

  3. 03

    A mark on your risk history

    A single recorded occurrence already weighs on the agency's risk history assessment, in the same severity category as a payment default. A change in risk status can trigger financial guarantee requirements and more frequent remittance.

  4. 04

    You get back by evidencing

    The restriction stays in place until the agency demonstrates, to IATA's satisfaction, that it meets the requirements to use the method again. In other words, the way out is the same work that would have avoided the problem.

Who does what

Everyone's limits, stated before you hire anyone

This is a market where certificates get promised by people who cannot issue them. We would rather be clear from the start about who signs what, because it changes what you should demand from us and what you need to solve elsewhere.

Your agency
Signs the self-assessment and the attestation, and submits them in the IATA portal. The word “self” is in the name: the declaration is yours, which is why it has to rest on real evidence.
DM11
Prepares. We design and reduce the scope, implement the controls, write the policies, train the team, organize the evidence and guide you through completing the questionnaire correctly. We do not issue certificates and we are not a QSA.
Your acquirer
Defines what to require from you and where to send it. Together with the card brand, they also determine whether your case calls for a formal assessment instead of a self-assessment.
An ASV
A vendor approved to scan for vulnerabilities. It runs the external scan when your environment has something published on the internet within scope. Only companies approved by the PCI SSC (the body that maintains the standard) may run that validation scan.
A partner QSA
Conducts the formal assessment when your volume or your acquirer requires one. In that case we work alongside an accredited partner, and the assessment is theirs, not ours.

Self-assessment

Where your agency stands today

Fourteen questions about how cards move through your operation. The full result appears on screen, with a score per area and an honest read on what has to change first. We do not ask for your email to show it.

Scope and flowQuestion 1 of 14

Can you list every point where a card number enters your agency?

Phone, email, messaging app, website, counter, reservation system, acquirer terminal.

What each result means

The reading for each area, across the three result ranges. It is the same text emailed to those who identify themselves, published here for anyone who wants to understand what the score measures before answering.

Scope and flow

Below 50

Without knowing where cards come in and go out, you cannot answer the self-assessment or size the work. Map the flow first: it is what most reduces cost later, because removing a channel from scope is worth more than protecting it.

50 to 79

The flow is partly mapped. What usually goes missing is the informal path: the card that arrives by messaging app, the one an agent writes down, the customer who calls after hours. That is exactly where compliance breaks.

80 or above

A well-drawn flow. The next step is freezing it in a document reviewed periodically, because new channels appear unannounced whenever the agency changes system or acquirer.

Storage

Below 50

Storing card numbers is what most inflates the cost of compliance, and it is almost always unnecessary. Find where they are, including email attachments and old spreadsheets, and remove them before protecting anything.

50 to 79

Primary storage has been dealt with and residue remains. Old backups, mailboxes and legacy systems hold cards for years without anyone remembering, and the self-assessment asks about all of them.

80 or above

No stored data, which is the best possible position. Keep scanning periodically: one new process is enough for numbers to start being stored again without anyone deciding it.

Channels

Below 50

Phone, email, counter and website carry different risk and need different measures. Treating them all with one rule is the mistake that makes an agency spend on the wrong channel.

50 to 79

Digital channels have been handled and the phone has not. It is the most common channel in an agency and the most overlooked: a call recording with a card number is stored data, whether anyone intended it or not.

80 or above

Channels under control. Check what happens after hours and in exceptions, which is where the agreed procedure stops being followed.

Access

Below 50

Without named users there is no way to know who saw customer data. It is the minimum the self-assessment asks, and a shared account undoes the whole answer.

50 to 79

There is a user per person and the rest is missing: multi-factor where the standard requires it, and periodic review of who still needs access.

80 or above

Access handled well. Look at provider access and the reservation system separately: they usually fall outside review because nobody thinks of them as users.

Systems

Below 50

An unpatched workstation with no antivirus is the most immediate finding. Start there: it is cheap and covers a good share of the configuration questions.

50 to 79

Workstations are looked after and the reservation system was not counted. It is part of the environment and the self-assessment asks about it.

80 or above

A sound technical base. What adds value now is a remediation cycle with deadlines, and a record that it happens: the question is not whether you fix, it is how fast.

Suppliers

Below 50

The agency depends on the GDS, consolidators and the reservation system, and their compliance is not automatic. Start with the list of who touches your customer's card.

50 to 79

The list exists and confirmation is missing. The self-assessment asks about each provider's compliance, and the acceptable answer is their document, not your assumption.

80 or above

The chain is mapped. What most agencies still lack is the responsibility matrix: which control is yours, which is the provider's, written down and agreed.

People

Below 50

With no written rule, the agent decides alone what to do with a card number, and each one decides differently. A one-page procedure and a short training session cover most of it.

50 to 79

The rule exists and has not been trained. A record of who received it and when is what turns the procedure into acceptable evidence.

80 or above

A trained team with recorded acceptance. Check new joiners and high season, when extra staff arrive without going through the same process.

Evidence

Below 50

With no organised evidence, filling in the annual self-assessment becomes guesswork, and the agency signs a statement it cannot support. Start by defining where each proof lives.

50 to 79

The evidence exists, scattered. Organising it by questionnaire item, with an owner, is what makes next year's filing take hours instead of weeks.

80 or above

Organised evidence, and it is what supports the statement. The next gain is reviewing ahead of the deadline rather than in its final week.

How we run it

From the card map to the evidence behind the declaration

Every phase ends with a deliverable. You know what you get before you start.

  1. 01

    Map the card

    We chart every point where data enters, travels and stops in your operation, including the informal ones nobody documents. This step defines the size of everything that follows.

    You get

    • Card data flow map
    • Inventory of the systems and people involved
    • Indication of the applicable self-assessment
  2. 02

    Reduce the scope

    Before implementing controls, we take data out of where it does not need to be. Changing the intake channel and eliminating unnecessary storage usually shrinks the project dramatically.

    You get

    • Scope reduction plan
    • Redesigned payment intake
    • Removal of improper storage
  3. 03

    Implement the controls

    We work with your team on what the standard requires for the scope that remains: access, authentication, network separation, patching, logging and monitoring, and management of the suppliers that touch cards.

    You get

    • Technical controls in place
    • Approved policies and procedures
    • Documented supplier management
  4. 04

    Train the front line

    The travel consultant is the person who receives the card. We train the team on the new procedure and on recognizing email and phone scams, which the current version of the standard explicitly requires.

    You get

    • Training delivered and recorded
    • Payment handling script
    • Quick reference material
  5. 05

    Organize the evidence

    We build the file that supports every answer in the self-assessment and guide the correct completion. The agency signs, and signs knowing what sits behind each item.

    You get

    • Evidence file per requirement
    • Self-assessment reviewed with the agency
    • Guidance for submission in the IATA portal
  6. 06

    Keep the cycle

    The declaration is annual and the operation changes along the way. We track the changes that affect scope and prepare the next cycle before it becomes urgent.

    You get

    • Annual cycle calendar
    • Scope review at each relevant change
    • Renewal preparation

Stories

Four situations we have already solved

We anonymise our clients with the same confidentiality that will protect your agency later. The names change, and the pattern of the problems repeats.

Corporate travel agency

The card that arrived by message

Situation
The service desk received corporate customers' cards by email and messaging app, because that was what customers preferred. Years of history sat in inboxes, and nobody could say how many numbers were in there.
What we did
We changed the intake channel to a payment link before anything else, which stopped new data entering within the first month. Then we located and removed the history, and rewrote the service script so that declining the old channel was polite and automatic.
Outcome
The scope shrank before any technology investment. Corporate clients accepted the change better than the board expected, and two of them asked for the same procedure from their other suppliers.
Agency with a call center

The recording that kept what it could not

Situation
The call center recorded every call for quality purposes, and customers read out card details over the phone. The recordings were searchable and archived for years, verification code included.
What we did
We introduced a recording pause at the payment moment, built into the operator's script so it did not depend on memory. We dealt with the old archive and adjusted retention to what the operation actually needs.
Outcome
The quality requirement stayed satisfied, without the liability. The point that caused most internal debate, supervisors resisting the loss of part of a recording, settled once it was clear the lost part was precisely what could not exist.
Agency selling online

It thought it qualified for the simplest questionnaire

Situation
The agency sold through its website and believed it qualified for the shortest self-assessment, because payment finished at the gateway. But the payment page belonged to the agency and touched the data before sending it, which changes the classification entirely.
What we did
We reclassified the case and adjusted the architecture so the data stopped passing through the agency's server. With the flow redesigned, the simpler classification became legitimate rather than declared by mistake.
Outcome
The agency moved from a declaration that did not hold up to one that does, with fewer controls to maintain. The technical change was smaller than the effort of continuing to declare it wrongly.
Consolidator

The scope that covered the whole company

Situation
The system handling cards shared a network with the entire operation, from sales to the visitor wifi. In practice every computer in the company was in scope, and the cost of treating all of it was prohibitive.
What we did
We separated the network to isolate the payment environment and cut the number of workstations touching cards to a minimum. The rest of the company left the scope, and most of the work left with it.
Outcome
The project became affordable because the territory shrank before any control was bought. The separation also made it easier to answer corporate clients' security questionnaires, which had been holding up contracts.

Frequently asked

What people ask before deciding

Answers anchored in IATA's resolutions and the PCI Security Standards Council's standard. Where no official figure exists, we say so.

Not exactly, and the difference matters. Resolution 812 makes authorisation to use the Customer Card Payment Method conditional on PCI DSS compliance. If your agency accepts cards against its own merchant agreement or issues tickets with the passenger's card, the requirement applies. If your operation never accepts, transmits or stores card data, there is a specific declaration for that case which replaces the compliance evidence.

Resolution 812 treats non-compliance as a risk event and provides that IATA will immediately restrict use of the Customer Card Payment Method. The agency stays in the BSP but stops issuing with the customer's card. If the ticketing system provider cannot apply the restriction, the BSP airlines are informed. The restriction remains until the agency demonstrates it meets the requirements. A single recorded occurrence also weighs on the risk history, which can lead to financial guarantee requirements.

We found no provision for an IATA fine in the text of Resolution 812. What IATA does is restrict the payment method and record the risk event. Fines do exist in the card ecosystem, but they belong to the card brands, sit in contracts and are charged through the acquirer, usually after a breach. These are different things, and it pays not to confuse them, because the response differs depending on where the charge comes from.

No. DM11 is not a QSA and does not issue certificates. We prepare your agency: we design and reduce the scope, implement the controls, write the policies, train the team and organize the evidence. The self-assessment and attestation are signed by the agency itself, which is how the mechanism works. When a case calls for a formal QSA assessment, we work alongside an accredited partner, and the assessment is conducted by them.

In most cases, no. For most merchants, including practically every travel agency, the path is a self-assessment signed by the merchant. A formal QSA assessment is usually required at very high transaction volumes, or when the acquirer specifically asks for it, which can happen after an incident. That is decided by your acquirer together with the card brand, not by IATA and not by us.

It depends on how the card comes in and whether it gets stored, not on your revenue. A warning is worth making: it is often written that travel agencies fall under SAQ C-VT, and in most Brazilian cases they do not. That type requires the acquirer's virtual terminal on an isolated machine and does not allow online sales. An agency typing cards into the reservation system's form of payment field is in a different, usually broader scenario. Declaring the wrong type creates more exposure than not declaring, because the attestation then asserts something the operation cannot support.

It depends on your environment, not your size. An external scan by an approved vendor is required when there are internet-facing components in scope, such as your own booking site or an exposed remote access. An agency operating only by phone and reservation system, with no public asset in scope, may legitimately not have that requirement. IATA does not define this: the standard does, according to your environment, together with your acquirer.

The verification code, never. The BSP Manual for Agents is explicit that agents must not store or write down that code, and the standard prohibits retaining authentication data after authorisation, with no exception and with encryption offering no way around it. The card number itself may be retained where there is a business need, with proper protection and a defined disposal deadline. In practice most agencies discover they do not need to keep it, and that single decision cuts both risk and effort at once.

Sending an unprotected card number over end-user messaging technologies breaches the standard, and that includes email, SMS and messaging apps. We know it is the industry's most entrenched habit, and the answer is not to ban it without an alternative: it is to offer a channel customers accept just as easily, usually a payment link. Changing the channel is usually the project's first step, because it stops new data entering while everything else is still being handled.

No. A virtual card is still a card number: if your agency receives it, types it and transmits it, it stays in scope. The real benefit of a virtual card is different and not small, because single use and spending limits cut the damage from fraud or a breach. But that is risk reduction, not scope exemption, and it is worth being wary of anyone presenting the two as the same thing.

Version 4.0.1. Version 3.2.1 was retired in March 2024 and 4.0 in December 2024. One date catches a lot of people out: most of the new requirements in version 4 were best practice until 31 March 2025 and became mandatory then, among them multi-factor authentication for all access to the card environment and training that covers phishing. Anyone who prepared documentation before that and has not revisited it is probably declaring against a version that no longer exists.

It depends how far cards have spread today and how much scope can be cut at the start. An agency that only needs to change its intake channel and organize evidence moves quickly. One that stores cards across several systems and records calls has earlier work to do. We would rather not publish a standard timeline because it would be guesswork: after the diagnosis we can put a date on your specific case.

No. Remediation happens alongside operations, and the first step usually improves sales rather than hindering them, because a payment link tends to be quicker for the customer than reading out a card over the phone. What needs planning is your team's time: consultants have to learn the new procedure, and evidence does not produce itself.

It is, and solving both together saves effort. Card data is personal data, so the same breach that triggers contractual penalties from the card brands also triggers privacy obligations, including notifying the authority and the individuals. Much of what PCI DSS asks for, such as knowing where data lives, limiting access and defining disposal, is exactly what privacy law requires. We handle both in the same project when the agency wants it.

More questions? Talk to DM11

Start by knowing where the card comes in

A thirty minute conversation is usually enough to say which self-assessment is likely to apply to your agency and what can be resolved quickly. No obligation.

Talk to a specialistTake the self-assessment

Comparisons on this subject

  • Pentest vs Vulnerability Assessment
  • CIS Controls vs ISO 27001
See all 13 comparisons

Sources and references

  • Payment Card Industry Data Security Standard

    PCI Security Standards Council · PCI DSS v4.0.1 · accessed on July 28, 2026

Published on
July 24, 2026
Last reviewed on
August 01, 2026
Editorial review
DM11 Consultoria em Segurança da Informação Ltda

This page is informational and describes how DM11 reads and applies the sources above. It does not reproduce the text of any standard, does not replace reading the official document, and does not replace an audit, a certification, an independent assessment or legal advice. Where a standard requires formal assessment, it is carried out by an accredited body, auditor or assessor, always separate from whoever did the preparation.

DM11AI TRUST & IT RISK PROTECTION

ouvir. entender. resolver.

Trust to grow in the AI era. AI governance, IT risk and compliance, cybersecurity and business continuity for companies that cannot stop.

Solutions

  • AI Trust
  • Governance, Risk & Compliance
  • Cybersecurity
  • Security Office
  • Business Continuity

Products

  • oitenta20®
  • Jigphish®
  • Ethical Hacker as a Service
  • DPO Backoffice®
  • SastAction®
  • NosConformes®
  • Cyber Antifrágil®
  • All products

Company

  • About Us
  • Case Studies
  • FAQ
  • Contact

Contact

  • contato@dm11.com.br
  • +55 (11) 4837-5758
  • Av. Eng. Luís Carlos Berrini, 1140, 7º andar, Brooklin, São Paulo/SP, CEP 04571-000
  • Standards and certifications
  • Comparisons between standards
  • Self-assessments
  • Services
  • Request a proposal

DM11 © 2026 · All rights reserved.

  • Privacy Policy
  • Cookies
  • Terms of use
  • Ethics and conduct
  • Anti-corruption
  • Social responsibility