What SOC 2 CC6.7 asks for when data leaves your systems
CC6.7 asks you to limit the transmission, movement and removal of information to authorized internal and external users and processes, and to protect that information while it is in transit, being moved or being removed. If you are the compliance lead or engineer who must show a SOC 2 auditor that data transfers are controlled, the test is not whether your database is locked down. It is whether you can show who or what can send data out of the system, over which channels, to which destinations, and what protects it along the way. The main risks are customer data copied to unmanaged laptops, sent over unencrypted connections, exported to USB drives or personal cloud storage, or pushed to a third party through an API key nobody reviews. The difficulty is that transfers rarely follow one path. They happen through application APIs, file sync clients, backup jobs, email attachments, support exports, analytics pipelines and developer laptops, so an inventory of transfer paths is usually the first real piece of work.
CC6.7 sits next to two criteria that people often confuse with it. CC6.1 covers logical access security, including how users authenticate and how sensitive data is encrypted where it is stored. CC6.7 covers the data once it starts to move or leave. CC6.5 covers disposal of physical assets, meaning sanitizing or destroying disks and devices so their data cannot be recovered. CC6.7 covers removable media and endpoints while data is on them, but the final sanitization step belongs to CC6.5. Keep these boundaries clear in your control matrix so the auditor does not find the same control described twice under different criteria. The points of focus under CC6.7 cover restricting the ability to transmit data, using encryption technologies or secure communication channels, protecting removable media, and protecting endpoint devices. Points of focus are guidance that can help you design and evaluate controls, not separate requirements, and you do not need to address every one. What you do need is a documented decision for each transfer path you operate, with the control and evidence behind it.
Before you select tools or write policy, confirm three things with your auditor: which transfer paths fall inside the system description, which of them involve a subservice organization whose controls you rely on, and which transfers you have decided to permit, restrict or block. A short register that lists each path, its owner, the channel, the protection applied and the place where the record of transfers is kept will give you the material to answer later questions from the auditor. The rest of this page takes each area in turn.
Which transfers, systems and third parties fall under CC6.7
Scope decides how much of the data movement an assessor will test. Each of the items below changes the set of transfers, endpoints and parties that need controls and evidence, and each should be settled against the system description before fieldwork begins.
User uploads and downloads
Every path by which customers or staff send files into the service or pull them out is in scope, including web forms, APIs, sync clients and shared links. For CC6.7, check that each path uses encrypted channels, that link sharing is restricted to authorized recipients, and that download permissions are logged. Ask the team to list every upload endpoint, because file sharing features are often added after the original design.
Service-to-service calls
Traffic between internal services, databases, message queues and cluster components can carry the same sensitive data as external traffic, so assessors often ask whether it is encrypted. Document which internal links use TLS, how certificates are issued and renewed, and where service credentials are stored. Internal traffic that is left unencrypted needs a documented justification and a compensating control.
Exports to analytics and support tools
Data copied to analytics platforms, support desks, error trackers or spreadsheets leaves the system boundary as soon as it is exported. Inventory each integration and the fields it receives, and confirm that exports are limited to what the purpose requires. Also confirm who can run exports and whether those exports are logged, since a bulk export is one of the most common ways information is removed without an approved process.
Backups and replication
Backup jobs, snapshots and replicas copy the full data set to other locations, often in other accounts or regions. Confirm that copies are encrypted in transit and at rest, that the replication targets are inside the defined boundary or covered by a third party review, and that restore access follows the same rules as production. Backup media moved off site is a removal path that assessors expect to see described.
Staff endpoints and removable media
Laptops, workstations, phones and USB drives used by staff who handle production or customer data are the last step in many transfer paths. Document which devices are permitted, whether disk encryption and screen locking are enforced, and how removable media is restricted or logged. Staff who download data to personal devices create a transfer path that the system description must either include or exclude with a stated reason.
External parties and the system boundary
Customers, processors and subservice organizations that receive data define where your responsibility ends. Confirm each recipient against the system description, the data processing agreements and any carve-out for a subservice organization's own SOC report. A transfer path left out of scope still needs a documented reason an assessor can accept, such as a separate system with no shared data or a clear contractual boundary.
How an auditor tests CC6.7 and what they ask to see
This table maps each likely auditor test for CC6.7 to the evidence the auditor requests and the findings that often appear. Read it as a preparation list: for each row, confirm the organization can produce the artifact before fieldwork starts, because a Type 2 test depends on records from the whole observation period.
| Test area | Evidence requested | Common findings |
|---|---|---|
| Data flow inventory against system description | Current data flow diagrams, transfer path inventory, system description sections on data movement, and sign-off dates | Transfer paths used by integrations, backups or support tooling that appear in no diagram or description |
| Encryption settings for external and internal channels | TLS configuration for public endpoints, internal service encryption settings, certificate inventory, VPN or private network configuration | Legacy protocols still enabled, internal services carried over plain HTTP, certificates that expired without an alert |
| Certificate management and renewal | Issuer configuration, renewal logs or automation records, and examples of certificate issuance and rotation | Manually installed certificates with no owner, or renewal that depends on one person remembering a date |
| Sampling of transfers and export events from logs | Auditor-selected samples of file transfers, bulk exports and API data pulls, with the log entries and approval records for each sample | Export events that cannot be tied to a ticket, approval or business purpose, or logs that do not capture destination |
| Removable media and USB restrictions | Endpoint management policy settings for removable storage, device compliance reports, and exception records | Restrictions configured on some device groups only, or unmanaged endpoints that never reported into the management console |
| Data loss prevention and egress rules | Egress rule sets, DLP policy configuration, alert records, and the list of approved exceptions with their approvers | Exceptions approved at the time of the request but never reviewed or removed, so they remain open indefinitely |
| Type 2 operating effectiveness over the observation period | Evidence that each control operated throughout the period, including recurring reviews, alert triage and exception reviews dated across the window | Controls documented as performed but with gaps in dated evidence, or a review that happened only at the start of the period |
How to implement CC6.7 in a sequence the auditor can follow
This sequence turns CC6.7 from a policy statement into inventoried transfer paths, approved channels, enforced controls and recorded exceptions, so the auditor can trace each step to evidence. The organization owns every control, while an implementation partner prepares the systems and evidence that show the controls work.
Inventory every transfer path and data class
Map each route by which information leaves or moves within the system boundary: user uploads and downloads, API calls, file sync, email attachments, backups, database replication, vendor and customer exchanges, and removable media. Record the data classification for each path and the owner who approves it. The output is a transfer register that lists source, destination, data class, owner and in-scope status, reviewed by management. Confirm the register with the system owners before moving on, because later steps depend on it being complete.
Define approved channels and encryption standards
For each path in the register, decide which channels are approved and which encryption standard applies, such as TLS for web and API traffic, encrypted sync for file storage, and encrypted transfer for backups. Write the standard as a short table that names protocols, minimum settings and the team that maintains them. Management approves the table, and it becomes the reference the auditor uses to test whether a transfer used an approved channel.
Deploy encryption in transit and certificate lifecycle
Configure the approved channels on self-hosted systems inside the client's infrastructure, and enable server-side encryption where the platform supports it. Issue certificates from an internal or public CA, and automate renewal so expired certificates do not break transfers. Tools such as FreeIPA's integrated certificate authority or cert-manager for Kubernetes workloads can issue and renew certificates, and OpenBao's transit engine can handle encryption of application data with key rotation. Confirm each platform's capabilities in its documentation before relying on them. The output is a configuration baseline and a certificate inventory with issuance and renewal records.
Restrict removable media and endpoint export
Set device policy so that removable media is blocked by default or permitted only for approved devices, and so that endpoints cannot copy regulated data to unmanaged storage or personal accounts. Apply disk encryption and remote wipe to laptops and mobile devices that hold in-scope data. Record which users have exceptions and why. The output is a device policy export, an approved-device list and a configuration report showing the settings are enforced, not merely documented.
Set egress and sharing rules with documented exceptions
Define rules for outbound transfers, including external sharing links, API egress, and email to outside domains. Restrict who may create external shares and set expiry and approval requirements where the platform allows. Every exception needs a named approver, a business reason, a scope and an end date. The output is an egress rule set and an exception log that the auditor can sample against actual transfers.
Log transfers and removals
Enable logging for file transfers, share creation, administrative exports, API calls that move data, and removable media events, and forward the logs to a central store that users with operating rights cannot edit. Confirm that log retention matches the organization's policy and that someone reviews logs for unusual transfers. The output is a sample of logs for each transfer type, a documented review cadence and records of the reviews performed.
Write the policy and procedure
Document the transfer register, approved channels, encryption standards, device rules, egress rules and exception process in a data transmission policy and a supporting procedure. Assign an owner for each control and state how exceptions are approved and closed. Management approves the documents and records the approval date. The output is a signed policy set that matches what the systems actually do, which is the point the auditor will check first.
Run a first review of exceptions before observation starts
Before the Type 2 observation period begins, review open exceptions, confirm each one is still needed, close those that are not, and check that logs and certificate records cover the full period. Compare the transfer register with the current systems to catch new paths. The output is a dated review record with decisions and owners, showing management tested the control before the auditor arrived. Ask the auditor which evidence format and sample period they expect, since requirements vary by engagement.
Which open-source tools support encryption and certificate evidence
The four tools below each produce a different kind of CC6.7 evidence: encryption at rest for stored files, encryption and signing for application data, certificates issued by an internal authority, and TLS certificates for workloads. None of them makes an organization compliant on its own. Each one needs a documented configuration, a named owner and records that an assessor can sample.
Nextcloud server-side encryption
Nextcloud's server-side encryption protects file contents at rest on local and external storage, with keys held on the server. For CC6.7, the useful evidence is the encryption configuration, the key handling procedure and sample exports showing that stored files are encrypted. The caveat matters: file names and folder structure are not encrypted, so an assessor may ask how sensitive names are protected or whether they are excluded from the data classification. Confirm which storage backends are in scope and how key recovery is governed.
OpenBao transit secrets engine
The OpenBao transit secrets engine lets applications encrypt and decrypt data without the engine storing that data. It also signs and verifies data and supports versioned encryption keys with key rotation. Evidence includes policy files that limit which identities may call encrypt or decrypt, rotation records, and logs of key use. Confirm that application teams route sensitive data through the engine and that key version retention meets your own data retention rules.
FreeIPA certificate authority
FreeIPA includes an integrated certificate authority based on Dogtag for issuing and managing certificates to users, hosts and services. For CC6.7, this supports controlled issuance of the certificates that protect channels between internal systems. Evidence includes the certificate profiles, the issuance and revocation log, and the list of who may request or approve certificates. Confirm the CA hierarchy, key protection for the CA itself and the process for revoking certificates when a system is decommissioned.
cert-manager for Kubernetes workloads
cert-manager creates TLS certificates for workloads running in Kubernetes or OpenShift from issuers such as ACME (Let's Encrypt), a CA key pair, HashiCorp Vault, CyberArk or self-signed issuers. It stores certificates in Kubernetes Secrets and renews them before they expire. Evidence includes issuer configuration files kept in version control, certificate status records and renewal events. Confirm which issuer each namespace uses, who can change issuer resources and how a failed renewal alerts the on-call team.
What evidence passes for each transfer path
Assessors testing CC6.7 for data in motion want to see that every channel carrying customer or internal data is encrypted, that the configuration allows only approved protocols, and that certificates are issued, tracked and renewed under control. The evidence is mostly configuration state plus records showing that state held over time. Platform or security engineers generate it from load balancers, ingress controllers, service mesh settings, database connection settings and the certificate authority, and the auditor samples both current settings and historical changes.
- Configuration exports: Snapshots of TLS settings from reverse proxies, ingress controllers and internal services, showing enabled protocol versions and cipher suites, taken from the running configuration rather than the template.
- Certificate inventory: A list of every certificate in use, with issuer, subject, expiry date and the owning system. For Kubernetes workloads, cert-manager resources and their Secrets are the usual source.
- Renewal logs: Records showing automatic or manual renewals before expiry, including failed attempts and how they were resolved.
- Change records: Tickets or pull requests for any change to TLS policy, linked to approval.
In a Type 2 sample, the auditor usually selects several sample dates across the period and asks for the configuration and certificate inventory as of each date, then checks a handful of renewals. Confirm with your auditor how many dates and renewals they expect for your population, and capture these records on a fixed schedule so no gap needs explaining.
Where applications encrypt data themselves, or call a service to do so, the auditor needs proof that keys are controlled, rotated and used only by authorized people and processes. Evidence here centers on the key management system, so the platform team typically owns it, with the security team reviewing it. Key version history and key operation logs are the core artifacts, and they must be kept for the whole period, not reconstructed at audit time.
- Key version history: Export of key versions, creation dates and state changes from the key management system, such as the OpenBao transit secrets engine, which supports versioned keys and rotation.
- Rotation records: Evidence that rotation ran on the schedule your policy sets, including the operator or automation identity that performed it.
- Key operation audit logs: Logs of encrypt, decrypt, sign and key administration calls, showing which identities made them.
- Access to key operations: Current and historical policies that define who may administer keys and who may use them, with periodic access reviews.
In a Type 2 sample, the auditor often picks a key, traces its version history across the period, and then checks that rotations happened when the policy said they should. They may also sample key operation logs to confirm that only expected service identities decrypted data. Ask your auditor whether they want logs from the key system or from the application, and confirm that log retention covers the whole period.
Self-hosted file platforms move data between users, devices and external parties, so CC6.7 evidence must show both how files are protected at rest and in transit and how sharing is restricted. Nextcloud, for example, offers server-side encryption that protects file contents on local and external storage with keys held on the server, but it does not encrypt file names or folder structure, so your evidence must describe that limit. Administrators generate the configuration evidence, and the security team reviews sharing policy.
- Encryption settings: Snapshots of the server-side encryption state, showing whether it is enabled, which storage is covered and who may enable or disable it.
- Admin configuration snapshots: Exports of the configuration files that control sharing, link expiry, password requirements for public links and external user access, captured on the dates the auditor selects.
- Sharing and external link policy: The written rule set, plus records of any exceptions granted.
- Sync client controls: Settings that govern which devices may sync and whether remote wipe is available, where the platform supports it.
In a Type 2 sample, the auditor usually asks for the configuration as of several dates, then selects a few external shares and checks that they followed policy and were approved where required. Confirm with your auditor how they want to see the limit on file name encryption addressed, since an unexplained gap is more likely to draw questions than a documented one.
Removing information from the system through a USB drive, an external disk or a laptop is where CC6.7 meets endpoint management. Evidence has to show that device controls exist, that exceptions are approved and bounded, and that every in-scope endpoint is inventoried with its encryption status. The endpoint or IT team generates most of this from device management tools, and the security team approves exceptions. Because device state changes constantly, auditors rely heavily on dated exports.
- Device policy exports: Exports of the mobile or endpoint management policy showing removable media restrictions, disk encryption requirements and screen lock rules, with the policy version and assignment groups.
- USB restriction settings: Configuration showing which device classes are blocked or allowed, and whether restrictions apply to all users or only some.
- Exception approvals: Signed or ticketed approvals for any device or user granted an exception, including the business reason and an end date.
- Device inventory with encryption status: A list of devices with owner, operating system, last check-in and disk encryption state, exported for each sample date.
In a Type 2 sample, the auditor typically takes the inventory as of several dates, checks that each sampled device shows encryption enabled, and then reviews exceptions to confirm they were approved and expired on time. Ask your auditor how they will treat devices that are offline at the sample date, and decide in advance how your team records them.
Where CC6.7 programs usually fall short
Most CC6.7 weaknesses are not missing tools. They are gaps in what the organization knows about its data paths and whether its settings are enforced and reviewed. The cards below cover the gaps assessors most often find, the fix for each, and the evidence that closes it.
Transfer paths missing from the data flow inventory
Teams often document the main application flow but miss secondary paths: nightly exports to a reporting tool, file syncs to a partner, log shipments to a third-party analytics service, or backups replicated to another region. An assessor who finds a transfer that is absent from the data flow diagram will question every control built around it. Fix this by walking each system boundary with engineers and recording every inbound and outbound path, its protocol, its endpoints and its owner. The evidence that closes the gap is a dated data flow inventory reviewed against infrastructure configuration and network rules.
Certificates renewed by hand
Manually tracked TLS certificates are a common cause of outages and of gaps in encryption in transit, because expiry is discovered when a service fails. Certificates for workloads on Kubernetes or OpenShift can be issued and renewed automatically by cert-manager, which stores them in Kubernetes Secrets and renews them before they expire. Check which certificates exist outside that automation, who holds their private keys, and how renewals are logged. The evidence that closes the gap is the issuer configuration, a certificate inventory and a renewal record for a sample of certificates.
Encryption at rest assumed to cover names and structure
Teams often assume that server-side encryption protects everything stored, including file names, folder structure and metadata. In Nextcloud, server-side encryption protects file contents on local and external storage with keys kept on the server, but it does not encrypt file names or folder structure. Assessors test what is actually protected, so the system description and the control narrative must state this boundary plainly. Compensate where names are sensitive by controlling who can see directory listings or by naming files with opaque identifiers. The evidence that closes the gap is the encryption configuration, a written statement of what is and is not encrypted, and a sample of storage paths showing the stated behavior.
Removable media rules that are not enforced
A policy that prohibits copying data to USB drives means little if endpoints allow mass storage devices without restriction. Assessors look for the technical control behind the policy: device control settings, disk encryption on laptops and workstations, and a record of exceptions. Confirm which endpoint management tooling applies the rule to each device class and whether it covers remote staff. The evidence that closes the gap is the endpoint policy export, a device compliance report, and a log of approved exceptions with their approvers.
Egress exceptions with no end date
Temporary exceptions that allow outbound transfers to a vendor, a test environment or a new destination often remain open for years. Each exception should name the data, the destination, the business reason, the approver and an end date, and it should be reviewed on a schedule. Without this, the firewall rule set drifts away from the approved design and the organization cannot show that transfers are limited to authorized destinations. The evidence that closes the gap is the exception register with dated approvals, expiry records and the results of periodic egress rule reviews.
No evidence for transfers to subservice organizations
When a cloud platform, email provider or managed database receives or moves customer data, the organization still needs evidence that those transfers are protected. Many teams rely on a vendor's SOC report without checking which of its controls cover encryption in transit and which are the organization's own responsibility. Confirm the complementary user entity controls in each vendor report and assign an owner for each one. The evidence that closes the gap is a subservice organization register, the relevant sections of each vendor report, and the organization's own configuration records for those integrations.
What meeting CC6.7 costs in money and effort
This table compares the main cost components of meeting CC6.7, who is paid for each, whether the cost is one-time or recurring, what drives it up and how to keep it down. No prices or durations are given because none are supplied for this topic, so use the table to structure a written quote and to decide which parts to keep in-house.
| Cost component | One-time or recurring | What drives it | How to keep it down |
|---|---|---|---|
| Internal engineering time to map transfer paths and configure encryption and certificates | One-time, with recurring changes | Number of data flows, external integrations, legacy protocols and systems that need reconfiguration | Inventory transfer paths first, prioritize flows that carry customer or sensitive data, and reuse one encryption standard across services |
| Endpoint management and device control tooling | Recurring license or support spend, or one-time setup for open-source tools | Device count, operating systems, whether removable media must be blocked or logged, and the number of managed endpoints | Confirm what the current device management licenses already cover, then compare with self-hosted open-source options where they meet the same control objective |
| Certificate and key management operations | Recurring | Number of TLS endpoints, issuers in use, rotation frequency and whether keys are held in a secrets platform or a certificate authority | Automate issuance and renewal, centralize issuers, and document key ownership so rotation does not depend on one person |
| Evidence upkeep: configuration exports and exception reviews | Recurring each review period | Number of controls that produce records, how often configurations change, and how exceptions are logged and approved | Export configurations on a schedule from infrastructure as code, store them in one evidence location, and assign a named reviewer |
| External review or testing of channels and endpoints | Recurring or per engagement | Scope of external-facing services, depth of testing, and whether testing covers encryption settings, endpoint controls and removable media handling | Agree scope in writing, test the highest-risk paths first, and retest only after material changes |
| Auditor sampling time during the Type 2 observation period | Recurring per audit cycle | Number of samples the auditor selects, population size, and how quickly evidence can be produced on request | Keep evidence indexed by date and control, so samples can be produced without rebuilding records during fieldwork |
Who owns CC6.7 and what an implementation partner can take on
Your organization owns CC6.7 in full. The criterion asks you to limit how information is transmitted, moved and removed, and to protect it along the way, so the decisions about what is allowed belong to you. No outside party can decide which data may leave the environment, which channels are approved, or whether an exception is acceptable. Those choices must be documented, approved by someone with authority over the risk, and reviewed when the system changes.
- Data classification: Decide which information is sensitive enough to require encryption, approval for transfer, or blocking from removable media.
- Approved channels: Name the transfer methods permitted for each data class, such as managed file sharing, secure APIs or email with encryption, and retire the rest.
- Policy and standards: Write the transmission, removable media and endpoint policies, and assign an owner for each.
- Exception sign-off: Record each exception, its business reason, its compensating control, its expiry date and the approver's name.
- Ongoing review: Confirm the channel list and exceptions still match the system description after every material change.
Before engaging anyone, check that these decisions already exist in writing, because an auditor will test whether the approvals were actually followed.
An implementation partner such as DeployOpen works on the technical controls that carry the decisions above. Under a fixed-scope written agreement, a partner can plan and configure self-hosted platforms inside your own cloud accounts, data centers or air-gapped networks, set up encryption and certificate issuance, and prepare the evidence the auditor will request. The partner prepares systems and records. It does not own the policy, approve exceptions, or attest that the controls work.
- Encryption setup: Configure server-side or key-managed encryption for stored and transmitted data on the platforms in scope.
- Certificate issuance: Build internal certificate authorities or automated issuance for internal services, with renewal and revocation procedures.
- Transfer and endpoint evidence: Produce configuration exports, logs and reports showing which channels are enabled and which devices are protected.
- Documentation: Write runbooks and a control matrix mapping each CC6.7 point of focus to the system that implements it.
- Scope limits: Confirm in writing which systems, environments and evidence types are covered, and who approves changes to that scope.
Ask for the deliverables, the named evidence and the acceptance criteria in the written scope before work begins.
The auditor is an independent licensed CPA firm. It tests whether CC6.7 controls are designed appropriately and, in a Type 2 examination, whether they operated effectively over the period. It then issues the report. Independence is the core requirement: the firm cannot design, implement or operate the controls it tests, because that would mean testing its own work. The line between preparing evidence and performing the examination is where most engagement questions arise.
- Design testing: Evaluate whether the channel rules, encryption settings and removable media controls would meet the criterion if they worked as described.
- Operating testing: Sample transfers, device records and certificate renewals across the period to confirm the controls ran as designed.
- Independence: Expect the firm to decline work that would make it responsible for your controls, such as writing your policies or configuring your systems.
- Evidence requests: Expect a request list covering approvals, exception logs and configuration exports, and check that each item exists before fieldwork.
- Report: Only the auditor issues the SOC 2 report, and it states its own opinion, not yours or your partner's.
Confirm with the firm, before signing, which preparation services it will provide and which it will decline.
Questions buyers ask about CC6.7 cost, effort and outside help
Is CC6.7 more effort than other CC6 criteria?
CC6.7 usually takes more coordination than most CC6 criteria, though the amount of work depends on how many data paths, endpoints and removable media channels are in scope. Criteria such as CC6.2 center on joiner and leaver records, which are repeatable. CC6.7 spans several technical layers at once: TLS settings on public and internal endpoints, encryption on backups and replicas, file sharing and sync tools, email and API transfers, and how laptops and removable media are handled. Each path needs an owner, a configuration baseline and evidence that the control operated across the whole period. The main drivers are the number of transfer paths, how many teams create them, and how much of the stack is managed by a cloud or SaaS provider whose controls you must describe instead of run.
Can our team meet CC6.7 without outside help?
Many engineering and security teams can meet CC6.7 without an implementation partner, provided someone owns each transfer path and can show evidence on a schedule. The skills involved are ordinary: certificate management, endpoint policy, data classification and change records. Outside help is most useful when the team lacks time to build evidence collection, when internal systems such as file sync or certificate authorities need to be installed or hardened, or when the team wants an independent view of gaps before the auditor arrives. Whatever the choice, the organization keeps ownership of its controls. A partner prepares systems and evidence, and only an independent CPA firm issues the report.
What does CC6.7 require from our staff and systems?
CC6.7 needs named owners and concrete inputs from several parts of the business. Expect to provide:
- System owners who can state which data flows leave the boundary, move between services or leave on removable media.
- A data classification scheme that tells engineers which transfers must be encrypted and which channels are approved.
- Administrative access, or a documented change path, to cloud load balancers, object storage, mail gateways, VPN and endpoint management settings.
- Time from security and operations staff for periodic exception reviews, where access to transfer channels or removable media is granted and then questioned.
Without these inputs, the control description will be generic and the auditor will ask for samples that the team cannot quickly produce.
How long must CC6.7 evidence be captured in a Type 2 period?
A SOC 2 Type 2 report covers an observation period, and the auditor tests whether CC6.7 controls operated throughout that period, not only on one date. The exact length is agreed with the CPA firm and is set out in the engagement, so confirm it in writing before work starts. Recurring capture matters because the auditor samples transfers, configurations and exceptions from across the period. If TLS settings were changed in month three and no record exists, the auditor cannot test the earlier or later state. Schedule evidence collection, such as configuration exports, certificate inventories and removable media logs, at a cadence that matches your change process, and keep the records in a place the auditor can reach.
Do self-hosted open-source tools remove the need for paid encryption or device management products?
Self-hosted open-source tools can replace some paid products, but they do not remove the need to design, operate and evidence the controls. Nextcloud server-side encryption protects file contents at rest on local and external storage, with keys kept on the server, but it does not encrypt file names or folder structure, so that gap needs a documented compensating decision. OpenBao's transit secrets engine can encrypt and decrypt application data without storing it and supports key rotation, which helps with application-level protection. cert-manager creates and renews TLS certificates for Kubernetes or OpenShift workloads from configured issuers, and FreeIPA includes a Dogtag-based certificate authority. Check that each tool covers your endpoint device and removable media needs, which are not the same as file transfer encryption, and confirm the vendor's current capabilities with the project documentation.
What should we request in a written scope before hiring a partner?
Ask for a written scope that names each transfer path in the CC6.7 boundary, the systems and owners for each path, and the controls the partner will configure or document. The scope should list deliverables such as a data flow inventory, encryption configuration baselines, certificate and key procedures, removable media and endpoint policies, and an evidence calendar that matches the observation period. It should state what the client must provide, which systems are excluded, which parts depend on a subservice organization's report, and how exceptions will be handled. Request the timeline, the acceptance steps and the change process for scope additions. DeployOpen works from fixed, written scopes agreed before work starts, and prices follow from that scope rather than a published list.
How DeployOpen scopes and prepares CC6.7 evidence
DeployOpen works on CC6.7 by agreeing in writing which transmission paths, data movements and removable or endpoint channels are in scope, then preparing the controls and records that show them working. We build and document the controls inside your own infrastructure, while your organization owns the controls and an independent CPA firm issues the SOC 2 report.
How we scope
- We list every path where information leaves or moves between systems, users and external parties.
- We agree with you which encryption, secure channel and removable media controls apply to each path.
- The written scope names the systems, deliverables, owners and timeline before any work starts.
How we build
- We configure encryption in transit and at rest on self-hosted platforms, such as server-side encryption with keys held on the server.
- We set up certificate issuance and renewal, and key rotation, so transmission controls keep working without manual steps.
- We restrict endpoint and removable media handling through configuration you can show an assessor.
How we hand over
- We document each CC6.7 control, its owner, configuration and the evidence it produces for the auditor.
- We test the controls in your environment and record the results, including any gaps we find.
- We walk your team through the evidence set so your staff can reproduce it for the examination period.
How we support
- Ongoing support, upgrades and continuous testing are available as a retainer after handover.
- Retained support covers configuration changes to transmission paths and certificate or key maintenance.
- Your organization still engages and pays the independent CPA firm, which decides what the report states.

