3D cyber incident reporting illustration with a silver aperture, cyan beam and glass timeline panels showing documented facts and decisions during an investigation.3D cyber incident reporting illustration with a silver aperture, cyan beam and glass timeline panels showing documented facts and decisions during an investigation.
S10 Group Newsletter · Identity & Containment Series

The Investigation Was Still Open.
Accountability Had Already Begun.

What you can tell customers and regulators while the facts are still changing
Newsletter #9
Published: 18 November 2026
By Stan van Gemert | S10 Group
Series #2: The Governable Organisation

Executive summary

In the previous issue, "Trust Has Become The Blast Radius", I looked at how trusted accounts, supplier connections and support tools shape how far a cyberattack can reach into your organisation.

Newsletter #9 picks up when someone outside your organisation needs an answer. What can you tell a customer, regulator or investor while the investigation is still open? And how will you show what supported that answer if the facts change?

I use two bank incidents to explore that question. Flagstar’s SEC case shows what happens when confirmed facts fail to make it into an investor filing. Sawyer Savings Bank’s updates this year show what customers can be told while an investigation is still open. A payroll example brings the question back to your team. What can you say today, and what evidence will you need if the answer changes tomorrow?

What Flagstar’s annual report left out

In December 2021, a cyberattack disrupted Flagstar’s mortgage business. The US bank had trouble closing loans. By the time it filed its annual report on 1 March 2022, Flagstar knew the attack had disrupted its business and that stolen files included customers’ personal information. The report repeated the previous year’s risk wording. It described business interruption and compromised customer data as things a cyberattack might cause.That’s the detail I keep coming back to. The people responsible for Flagstar’s disclosures were receiving regular updates. Yet what the bank already knew about the attack never made it into that part of the filing. The US Securities and Exchange Commission found the disclosure materially misleading. Flagstar later agreed to pay a $3.55 million civil penalty.

Flagstar knew customer data had been stolen before it filed

What happened?
Between 22 November and 25 December 2021, an intruder gained access to Flagstar Bancorp’s remote-access environment. The intruder obtained credentials and deployed ransomware that encrypted data on about 30% of the US bank’s workstations and servers. Data was stolen on 3 and 4 December.

The attack disrupted mortgage origination, servicing and loan closings. Flagstar took its network offline for several hours, restored or rebuilt hundreds of servers and reset thousands of passwords.

By 1 March 2022, Flagstar knew the stolen files included customer personal information. By early June, its review put the number of people whose personal information had been stolen at approximately 1.5 million. The SEC found the March annual report, a June customer notice and an August quarterly filing materially misleading. In December 2024, Flagstar agreed to a $3.55 million civil penalty without admitting or denying the findings.

Why it matters
The SEC found gaps in how Flagstar decided which facts to weigh for disclosure, who would assess materiality and how to document or pass that judgement to management. Regular technical updates were already reaching the people responsible.

‍Sources and further reading:

U.S. Securities and Exchange Commission - Flagstar Bancorp, Inc., now known as Flagstar Financial, Inc. (administrative order)
https://www.sec.gov/files/litigation/admin/2024/33-11343.pdf

U.S. Securities and Exchange Commission - SEC Charges Flagstar for Misleading Investors About Cyber Breach
https://www.sec.gov/enforcement-litigation/administrative-proceedings/33-11343-s

What happened?
CLOSE
View more details chevron A white downward-facing V-shaped chevron on a transparent background.

The branches could reopen while the investigation continued

On 3 August 2026, a network disruption forced Sawyer Savings Bank in New York to close its branches. In its 6 August update, the bank told customers they could still use online and mobile banking. Three days later, it announced that the branches would reopen on 10 August. It was still investigating whose data, if any, had been affected. Sawyer Savings Bank

On 9 October, Sawyer sent breach notices. The Indiana Attorney General’s register lists 16,069 people affected in total. The August updates helped customers decide how to access their accounts. The October notice addressed a different question about their personal information. In your own organisation, a customer may need a practical answer today while your team is still gathering the facts for a later notice.

Service restored. Investigation still open.

Reopening branches answered an immediate service question. It did not resolve whether personal information had been affected. Customers needed practical guidance while that separate investigation continued.

Sources and further reading:
‍

‍Sawyer Savings Bank - Technical Disruption Update (6 August 2026)
https://www.sawyersavings.bank/technical-disruption-update/

Sawyer Savings Bank - Branch Service Disruption Update (9 August 2026)
https://www.sawyersavings.bank/branch-service-disruption-update-branches-to-open-monday-august-10-2026/

Indiana Attorney General - Security Breaches: 2026 report, page 29, Sawyer Savings Bank
https://www.in.gov/attorneygeneral/consumer-protection-division/id-theft-prevention/security-breaches/

Why it matters
CLOSE
View more details chevron A white downward-facing V-shaped chevron on a transparent background.

When a customer needs an answer

Imagine you’re the CISO at a payroll provider. Your team detects unauthorised activity and suspends a file transfer connection. A customer sees the connection has gone down and asks whether to send its payroll file again. Your team knows it accepted the file before the suspension. Operations is still checking whether those wages have already been paid. If they have, sending the file again risks paying people twice.

You already have enough to give the customer useful instructions. At 11:00, you tell them you received the file, ask them to hold off sending it again and promise an update by 14:00. Operations checks the payment records and finds that the file was processed before the suspension. At 14:00, you confirm the file was processed and there’s no need to resend it.

The people assessing reporting obligations need a different view of the same facts. The confirmed payment tells them what happened to this customer’s payroll. To judge the wider impact, they still need to know which other payroll runs were affected and what the investigation has established about access to customer data.

Different reporting clocks ask different questions

A customer’s payroll deadline, a regulator’s notification deadline and an investor disclosure each call for a different judgement. Under NIS2, awareness of a significant incident starts the reporting process. DORA uses major-incident classification, with a time limit also linked to awareness. Under the current US SEC rule, the filing deadline generally runs from the determination that an incident is material. Record when the facts behind each judgement became available. The time your team first saw unusual activity may be different from the time it confirmed the impact on a service.

When each reporting clock starts

EU NIS2
For in-scope entities, Article 23 generally requires an early warning within 24 hours and an incident notification within 72 hours of becoming aware of a significant incident. Both are due without undue delay. The directive sets a shorter incident-notification deadline for certain incidents affecting trust services.

EU DORA
For in-scope financial entities, the initial notification is due as early as possible, generally within four hours of classifying an incident as major and no later than 24 hours from awareness. If major classification occurs more than 24 hours after awareness, the initial notification is due within four hours of that later classification.

US SEC Form 8-K
For domestic registrants subject to Item 1.05, filing is generally due within four business days of determining the incident is material. The determination must be made without unreasonable delay after discovery. The Flagstar case concerned reports and disclosure controls that predated this rule.

Sources and further reading:

European Union - Directive (EU) 2022/2555 (NIS2), Article 23
https://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=CELEX%3A02022L2555-20221227+

European Commission - Delegated Regulation (EU) 2025/301, Article 5
https://eur-lex.europa.eu/eli/reg_del/2025/301/oj/eng/pdf

U.S. Securities and Exchange Commission - Form 8-K, Item 1.05
https://www.sec.gov/about/forms/form8-k.pdf

Why it matters
CLOSE
View more details chevron A white downward-facing V-shaped chevron on a transparent background.

Keeping the facts behind each update

In the payroll example, I’d keep the 11:00 customer message beside the file receipt, the confirmed suspension and the question Operations was still checking. At 14:00, I’d add the payment confirmation, who verified it and the updated instruction. If the reporting team is still assessing the wider impact, the record should show the judgement so far, who made it and when they’ll revisit it.

That is what I mean by a record you can stand behind externally. Someone who missed the incident call can pick up the customer update or reporting decision and trace it back to the facts available when it was made. Keeping both versions also tells them which customers may need a correction when a later finding changes the advice.

Keep the evidence behind each update, especially when the next update changes the answer.

What containment gives you to verify

Stopping an intruder on a protected server closes a route they could use to reach more systems or data. Where deployed, Server Intrusion Protection detects reconnaissance and interrupts malicious activity, including lateral movement and data theft.

When your responders verify that interruption, they can give you a specific finding for the incident record. It should show which route was interrupted, when the team confirmed the effect and what remains under investigation. If you later need to describe the actions your team took, you can point to the result it verified.

Try it with an update you would actually send

In your next exercise, give your team the payroll customer’s question and leave the status of the file unresolved. Ask them to write the update they would send. Then give the original update and the new payment evidence to a colleague who missed the exercise. Can that person explain why the original advice was sound at 11:00, what changed by 14:00 and who still needs to hear about it?

That’s the question I’d take back to your next incident review. If a customer forwards yesterday’s update tomorrow, could you show what your team knew when it sent it and what it did when the answer changed?

The next horizon

In December, "What Good Looks Like Before The Next Incident" closes The Governable Organisation. I’ll look at what a realistic exercise can tell your executive team about its readiness, and which findings should lead to a change before the next incident.

In 2027, The Half Life of Control picks up where that test ends. How much confidence should you still place in the result when attackers change their methods, your technology evolves or a supplier changes how it operates?

Hear first when a new S10 Group release is published

Would you like to receive a short note when S10 Group publishes a new article or newsletter?

Leave your name and email address below. We will send a short update when a new release is available, with a brief summary and a direct link to the publication.
Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.

Coming soon

Friday 20 Nov 2026, Operational Attack-series #11: "The Extortion Trap Begins Before the Ransom Note"

Monday 23 Nov 2026, Resilience-series #12: "When Ransomware Becomes a Life-Critical Decision"

Wednesday 25 Nov 2026, Leadership-series #12: "The Board Sees the Outcome. Teams Carry the Pressure."

FOCUSED RANSOMWARE RESILIENCE ASSESSMENT

Compare controlled file-encryption scenarios with your existing security controls active, first without Ransomware Containment and then with it. The assessment itself is limited to encryption, but alongside the assessment results, we can explain how our platform and its additional Server Intrusion Protection and Virtual Server Protection features, helps detect and interrupt lateral movement, reduce the risk of data theft, protect remote server access and defend virtual environments such as VMware and Hyper-V. That makes the session useful not only as a ransomware-containment test, but also as a practical discussion about what your organisation can still control when identity, privilege and infrastructure trust come under pressure.‍
Bright conceptual visual of frosted-glass blocks where the first block is stopped by a titanium pillar, representing ransomware containment before wider operational disruption spreads.

Further readings

Previous S10 Group newsletter arc

17 March 2026 - #01: The Odido lesson:
The Odido Lesson →
Return to the identity-trust lesson: access may continue to work even after the organisation can no longer trust what that access represents.
08 April 2026 - #02:
When the Vendor Is Non-Negotiable →
Explore how dependency on critical suppliers can reduce room to manoeuvre when trust becomes uncertain.
14 May 2026 - #03:
The First Hour: Who Is Allowed to Act? →
Revisit the decision-rights question that determines whether the first hour produces action or delay.
17 June 2026 - #04:
Detection Is Not Control →
Understand why detection only matters when the organisation can safely turn signals into action.
15 July 2026 - #05:
Retention Drift →
Retention drift: the breach multiplier nobody decided on, using incident evidence and strategic analysis to explain trust, containment, governance and operational resilience under cyber pressure.
15 July 2026 - #06:
Resilience Has Become a Governability Question →
Read how NIS2 and governability turn resilience from a technical ambition into an executive control question.
16 September 2026 - #07:
The Organisation Moves Slower Than the Attack →
See why escalation alone cannot contain an attack, and how pre-agreed, bounded authority turns a credible signal into action before the containment window closes.
14 October 2026 - #08:
Trust Has Become The Blast Radius →
Explore how compromised accounts, supplier connections and unsupported equipment can widen a cyberattack’s reach, and what your teams need to restrict access and contain further progression.