Incident response histories: sensitive context requires a careful handoff, so this boundary guide sorts eight layers of an archive into what a partner may mention, may ask about and must leave alone

Short answer

An incident response archive can point to a company worth introducing, but a partner should touch almost none of it. Mention only what the company has made public or approved, ask only metadata questions, such as how many closed cases, which years, which system and who wrote the reports, and never request, read or describe case files, indicators, privileged reports or breach notices. Find out whose each layer is: the company's own team, an outside response firm, counsel or a client. Then hand the conversation to the CEO, who decides whether to talk.

Visual briefFull-size diagram
Three nested zones: an outer zone of approved public facts a partner may mention, a middle zone of metadata a partner may ask about, and an inner core of case files, indicators, privileged reports and breach notices a partner never touches, with an arrow from the middle zone to a box where the CEO decides.
Illustrative boundary for an incident response archive; where each item sits is confirmed with the company and its counsel.

This guide is for people who sit near security work without owning a client's records: virtual CISOs, security consultants, managed service and managed security providers, response firm partners and former security leaders who now advise. You may know more about a company's incidents than almost anyone outside it. That knowledge is exactly what must stay out of a referral.

What an incident response history holds

A security incident leaves a bundle, not a single record. SourceX's guide to incident response report data lists the usual parts of a case file: engagement metadata, an analyst timeline, scoping notes, a root cause, a containment and eradication log, and a final report with an executive summary. It also notes that raw forensic images, memory dumps and full log exports are rarely licensable, because they are dense with credentials, personal data and other parties' content.

Archives also change shape over time. NIST finalized Revision 3 of SP 800-61 in April 2025, replacing Revision 2, the Computer Security Incident Handling Guide. In announcing it, NIST said the details of incident response change so often and vary so much across technologies, environments and organizations that a single static publication can no longer hold them. An archive built since 2016 may follow two or three generations of templates, so ask which framing each era used rather than assuming one structure.

Why the context is more sensitive than the record type

Many operational records become sensitive only when someone misreads them. Incident records start sensitive, for several independent reasons laid out on SourceX's guide:

  • Privilege. Reports written at the direction of counsel may be off limits entirely.
  • Split ownership. Response firms usually own their templates and methods while the client owns its incident facts, and both have to agree before anything is released.
  • Victim identity. Regulator filings, notification letters and attorney general correspondence identify who was harmed.
  • Other parties' intelligence. Indicators received under TLP:AMBER, TLP:RED or ISAC membership rules need the originator's permission before they flow anywhere.
  • Insurance. Cyber insurance panel terms may restrict who can see reports.

Public companies add one more. The SEC's 2023 rules require registrants to report any cybersecurity incident they determine to be material under Item 1.05 of Form 8-K, describing its material nature, scope, timing and impact, generally within four business days of that determination. A partner who speculates about an incident, even in a friendly note, is commenting on something a company's lawyers and board may be weighing. Never characterize an incident's severity or significance.

The sensitive-record referral boundary guide

Three zones govern what you do. Mention only facts the company has published or approved for you to repeat. Ask only for metadata: counts, years, systems, authorship and whether a layer exists. Leave alone everything else, which means not requesting it, not reading it if offered and not describing it from memory. Then apply the zones layer by layer.

Layer of the archiveWhose it usually isZoneWhat a partner may askWho answers
The company's own incident tickets and post-incident reviewsThe company'sAskWhich system, since which year, roughly how many closed cases, who writes themSecurity lead
A response provider's case files on its clients' incidentsFacts are each client's; templates and methods are the provider'sAsk, about the provider's own share onlyWhether any clients ever agreed to reuse; what share of the archive is the provider's own methodsProvider's leadership and counsel
Reports prepared at the direction of counselPossibly protectedLeave aloneOnly whether such reports exist, yes or noCounsel
Breach-notification letters and regulator correspondenceIdentify the victimLeave aloneOnly whether they are filed apart from case recordsCounsel
Shared threat intelligence received under TLP or ISAC rulesThe originator'sLeave aloneOnly whether it is stored apart from case recordsSecurity lead
Reports shaped by an insurer or brokerMay be limited by panel termsLeave aloneWhether any policy or panel agreement restricts reuseCounsel
Raw forensic images, memory captures and full log exportsThe company's, but dense with credentials and personal dataLeave aloneNothingNobody needs to
Anything tied to a public company's disclosure decisionsThe company's, under SEC rulesLeave aloneNothing; never offer a view on materialityThe company's counsel

The table is lopsided on purpose. Only the first two rows can carry an introduction, and the second only for the provider's own methods unless clients have agreed. Most layers exist on the table so that you recognize them and keep your distance. SourceX's guide says realistic supply comes mostly from older closed matters where the client consented, from internal security teams documenting their own incidents and from response firms willing to release redacted timelines without narrative, which is the same pattern seen from the buyer's side.

Original work, mixed client content and an unclear archive (Illustrative)

Illustrative. Three generic companies with invented details; nothing here implies an outcome.

A. The company's own incidents. A 450-person software company has an internal security team that has logged every incident, from phishing to cloud misconfigurations, in its own tracker since 2017, with a post-incident review on each closed case. No client cases are mixed in. The head of security can answer every metadata question and agrees to raise the idea with the CEO. Counsel-directed reports from one 2021 investigation are filed separately and stay out of any description.

B. Mixed client content. A 120-person response and managed security firm holds nine years of case files on clients' incidents, written with its own templates and playbooks. Some engagements came through clients' outside counsel. The firm's own methods, templates and anonymized playbooks are its own; each client's incident facts need that client's agreement, and the counsel-directed reports may be off limits. The firm's leadership describes these as separate parts, and the realistic starting point is its own methods.

C. An unclear archive. A 260-person managed service provider handled clients' incidents for years under master service agreements that say nothing about reusing records. In 2024 it acquired a smaller security firm, and that firm's case archive came with the deal. Nobody knows whether the acquired firm ever obtained client consents. The MSP's internal incident tickets are plainly its own; the inherited archive is an open question. If the CEO wants to explore it, the introduction goes ahead with that question written down, and counsel and SourceX's rights review answer it. MSPs meet a version of this whenever client work and internal work share one platform.

Who owns which layer, and who decides

The security lead knows the systems, the years and how cases are filed. Counsel knows which reports are privileged, what engagement letters and client agreements say, and whether any matter is still in litigation or under regulatory inquiry. For a provider, each client owns its incident facts. The CEO decides whether the company discusses licensing and signs for it.

SourceX's who-qualifies page frames rights in three directions at once: a license must leave intact what the company owes its clients, its employees and the other parties to its contracts. Incident archives press on all three, which is why the boundary guide routes most layers to counsel. None of that matters if the company fails the basic screen: a peak of at least 50 full-time staff with contractors excluded, an operating history on record, the rights to license and a CEO open to the conversation. Companies in the U.S., Canada and the EU qualify, as do companies in the UK, Australia, New Zealand, India, Singapore, South Africa, Latin America, the UAE and Saudi Arabia when at least 70% of the data is in English. If the company says its records are simply too sensitive, the honest answers are collected in is my company's data too sensitive to license.

The handoff, step by step

The careful part of this referral is how little passes through you.

  1. Talk to the likely sponsor, usually the head of security, in general terms. No incident names, dates or clients.
  2. Ask the metadata questions below, and nothing else.
  3. Let the sponsor brief the CEO. An introduction made without the decision maker's knowledge is usually not a fit.
  4. If the CEO wants the conversation, share your partner link so the company applies itself, after you have signed the Channel Partner Agreement.
  5. Step back from the details. Security and privacy questions go between the company and SourceX directly; you neither relay nor summarize the answers.

Questions to send the sponsor, none of which needs a file:

  • Which system holds closed incident records, and since which year?
  • Roughly how many cases close in a year, and what share involve clients rather than the company's own systems?
  • Who writes post-incident reviews: your team, outside responders or both?
  • Are counsel-directed reports, breach notices and shared intelligence filed separately from routine cases?
  • Is any matter still in litigation or under regulatory inquiry?
  • Have any of these records been licensed for AI training before?

An earlier AI training license is the one answer that closes the file on the spot: SourceX takes only records its suppliers confirm were never licensed for that purpose. For handoffs between response providers, the incident containment handoff guide covers the records a provider change splits.

When the boundary holds and the referral still fails

  • Screening removes most of the supply. Privilege, client consent and victim identity take out large parts of most archives. A company may have years of incidents and little it can offer.
  • Open matters wait. The illustrative request template on SourceX's incident guide asks for cases that are closed and outside active litigation or regulatory inquiry.
  • Demand is a review question. Whether a current SourceX program wants incident records from that sector is checked in review.
  • Your knowledge is a liability here. The more you know about a client's incidents, the more careful the referral has to be, and the more tempting it is to say too much.
  • Unknown odds. How incident archives fare once introduced is not yet published. The best gauge available is a private log of the layers each sponsor confirmed and how each CEO answered.

Next step

Refer a company. No cap on rewards. Default reward on a company's first qualifying transaction: the greater of $50,000 plus 5% and 15% of Actual Supplier Payout, with no cap; further qualifying transactions first agreed within 12 months of the first Closing pay 15%. Eligibility and payment conditions apply, including the USD 100,000 Minimum Transaction, and a signed Schedule A or addendum can differ. A lead, meeting or signature alone does not trigger payment.

When the sponsor confirms the company's own incident records exist, the sensitive layers are filed apart and the CEO wants to talk, apply for a partner account, check the company against who qualifies and read the program terms before your link goes anywhere. Related guides on incident and on-call records, MSP client security questions and assessing material without seeing privileged work sit in the guides hub.

  1. Step 1Share your linkSend your personal link to a company you know.
  2. Step 2Company appliesThe company applies itself at /apply.
  3. Step 3Company signs and is paidThe company signs, delivers its data and is paid by SourceX under its agreement.
  4. Step 4You get your rewardthe greater of $50,000 plus 5% and 15% of the actual supplier payout, with no cap. A reward becomes payable after SourceX has paid the referred company under its signed agreement and the payout conditions in the program terms are met, including finance approval and your tax and payment details.

Common questions

Can I mention a client's well-known breach as a reason they might have useful records?

No. Even a breach that made the news is the company's to discuss, and linking it to a licensing idea in an email can land in the wrong inbox. Raise the company's security operations in general terms, and only with the person who would sponsor the idea.

Is an IT outage postmortem the same as an incident response record?

Not for this page. SourceX's incident response guide covers security incidents handled by response teams, managed security providers and managed service providers, and points to separate material for outage and reliability postmortems. Ask which kind of archive the company means before using the boundary guide.

What do I do if the security lead sends me a redacted report anyway?

Do not open or forward it. Tell the sender it is not needed, ask them to keep it inside the company and describe the archive in words instead. A redaction you have not checked can still carry hostnames, client names or indicators.

Sources

  1. refer.sourcex.si/terms
  2. refer.sourcex.si/who-qualifies
  3. sourcex.si/ai-data/industries/incident-response-report-…
  4. csrc.nist.gov/news/2025/nist-revises-sp-800-61
  5. sec.gov/newsroom/press-releases/2023-139
  6. sourcex.si/how-it-works

Free resources

Know a company with valuable proprietary data?

Become a referral partner, get your link and introduce the company’s CEO (in an owner-run company, usually the owner). Referral partners can join from most countries. We can't work with partners in countries or regions under U.S. sanctions or on U.S. restricted-party lists. The companies you introduce should be based in the U.S., Canada and the EU, or in the UK, Australia, New Zealand, India, Singapore, South Africa, Latin America, the UAE and Saudi Arabia with at least 70% of their data in English.

Refer a company →

I own a business

Explore licensing your company's data directly to SourceX for AI training and evaluation. Start a short assessment; no uploads needed.

Start an assessment