Frequently asked questions about Connect Customer screen recording capabilities
This topic provides frequently asked questions about using Connect Customer screen recording capabilities.
Screen recording FAQ
General specifications
-
What is the file format of screen recordings?
The screen recording files are saved in MP4 format.
-
Which Connect Customer channels are supported?
You can generate screen recordings for voice and chat contacts. Screen recording is not supported for task contacts.
-
Do you capture the entire screen?
Screen recording captures only the Connect Customer agent workspace tabs, not the entire desktop.
-
Does screen recording support concurrent user sessions on Windows using Virtual Desktop Infrastructure (VDI) environments?
Yes, screen recording supports concurrent user sessions on Windows when using Connect Customer Client Application version 2.0.0 or later.
-
Where are the screen recording files stored in my AWS account?
The screen recordings are delivered to your Amazon S3 bucket and encrypted using the KMS key you specify. This is similar to how call recordings are stored and encrypted.
-
How can I be notified when there is a latest version of the client application?
-
For Windows, to be notified when there is an update to the Connect Customer Client Application, we recommend subscribing to the RSS feed of this administrator guide. Choose the RSS link that appears under the title of this page (it's next to the PDF link).
-
For ChromeOS, Isolated Web App and Chrome Extension are hosted and managed by Amazon Connect. They are automatically updated as newer versions are published.
-
-
Can I opt only for screen recording and not for call recording?
Yes, you can enable screen recording without call recording for a voice call.
-
How do I find the Amazon S3 location of the screen recording?
You can find the screen recording location in the RecordingsInfo section of the contact record. See the Location field.
-
How do I enable screen recording for a percentage of my contacts?
You can use the Distribute by percentage block in the flow to enable a percentage of contacts for screen recording.
-
Is screen recording PCI compliant?
Connect Customer, including the screen recording capability, is compliant with the Payment Card Industry Data Security Standard (PCI DSS). However, you are responsible for determining whether your specific implementation meets your compliance requirements.
Important
During a video call or screen sharing session, agents are able to see the customer's video or screen share even when the customer is on hold. It is the customer's responsibility to handle PII accordingly. If you want to change this behavior, you can build a custom CCP and communication widget. For more information, see Integrate in-app, web, video calling, and screen sharing natively into your application.
-
Does screen recording work with custom CCP and agent desktops?
Screen recording is designed to work with custom CCP and agent workspace built with the Connect Customer Streams JS library
. We recommend testing your custom solution before deploying screen recording in production. -
Can I use screen recording anywhere in the world?
Screen recording is available in AWS GovCloud (US) and all AWS commercial Regions where Connect Customer is available. However, your use of screen recording may be subject to compliance with privacy and other laws. Please consult your compliance team before enabling this capability for your agents.
To use screen recording in AWS GovCloud (US-West) requires client version 2.0.3 or later.
-
Are agents alerted when screen recording is enabled for a contact?
By default Connect Customer doesn't provide a notification feature. However, you can use the Connect Customer Streams JS library
to create a notice or other visual indicator on an agent's desktop to signal that screen recording is in use. -
What happens if an agent closes the browser during a contact, or immediately after a contact ends?
If the browser is closed at the beginning of contact before any screen capture data can be uploaded to Connect Customer, the final screen recording may not be published. If the browser is closed immediately after a contact ends but before the final screen capture data can be uploaded, the screen recording is published when the agent next logs in to CCP.
-
Does screen recording STOP when an agent places a customer on hold?
No, the screen recording continues recording when an agent places a customer on hold.
-
Is screen recording supported when Agents are logged into multiple CCP instances?
No, screen recording is not supported when Agents are logged into multiple CCP instances simultaneously either in the same or different browsers. You might see inconsistent behavior with screen recordings in these cases.
Configuration
-
Can I opt only for screen recording and not for call recording?
Yes, you can enable screen recording without call recording for a voice call. To do so, disable voice recording in the Set recording and analytics behavior block while keeping the screen recording enabled.
-
How do I find the Amazon S3 location of the screen recording?
You can find the screen recording location in the RecordingsInfo section of the contact record. See the Location field.
-
How do I enable screen recording for a percentage of my contacts?
You can use the Distribute by percentage block in the flow to enable a percentage of contacts for screen recording.
-
What is the average size of a screen recording file per minute in S3?
The average size of screen recording is 1.5MB/minute. This size can vary depending on factors like video encoding etc.
-
What is the frame rate for screen recording and is this configurable?
The screen is recorded at 5 frames per second and this is not configurable.
-
What codec is used for screen recording?
Screen recording uses OpenH264 codec.
-
Is there a way to choose which audio (redacted or unredacted) gets used for screen recording?
By default, the screen recording uses the unredacted audio. If you enable rule-based redaction for the contact, the redacted screen recording is stitched with the redacted call recording when Contact Lens call recording redaction is also enabled for the contact, and has no audio otherwise.
-
Is there a service limit for screen recording?
No, there is no service limit or quota for screen recording service.
-
Is there a maximum duration for screen recording?
No, the screen recording solution imposes no maximum duration for a recording.
-
How many agent monitors can be recorded?
Screen recording can record up to 3 screens/monitors.
-
Can I configure my call/screen recording storage S3 bucket to enable bucket level encryption with a KMS key that is different from the KMS key used as part of instance data storage configuration?
No, the same key should be used at bucket level and also as part of instance data storage configuration.
-
Where are screen recordings stored?
Screen recordings are delivered to your Amazon S3 bucket under the following prefix:
s3://your-bucket/connect/your-instance-alias/ScreenRecordings/year/month/day/contact-id_UTC-timestamp.mp4By default, the path prefix is
ScreenRecordings, but this prefix is configurable.
Performance
-
What are the bandwidth requirements for screen recording?
We recommend 500 Kbps bandwidth for screen recording, regardless of the number of concurrent contacts. The system requirements specify 600 Kbps total network bandwidth to account for additional agent workstation traffic.
-
Why do I see higher CPU usage after installing screen recording client application on my windows machine?
Screen recording in general is a CPU intensive application and hence CPU use increase is expected. We recommend ensuring you provide sufficient resources as documented in System requirements to avoid any resource contention issues.
Rule-based redaction FAQ
General specifications
-
Do I need to replace my existing screen recording setup to use rule-based redaction?
No. Rule-based redaction for screen recordings is an extension of Connect Customer agent screen recording. You enable it on a per-contact basis through a contact flow.
-
Is there an additional charge for rule-based redaction?
Rule-based redaction is included with Connect Customer agent screen recording at no additional charge. Standard recording and storage rates apply.
-
Does rule-based redaction affect storage costs?
Connect Customer stores both the unredacted and redacted recordings in your Amazon S3 bucket. You pay standard Amazon S3 storage rates for both files. If you do not need to retain the unredacted originals, you can use an Amazon S3 lifecycle policy to expire them on a shorter schedule than the redacted versions.
-
Does rule-based redaction apply to call recordings?
Rule-based redaction applies only to agent screen recordings. It does not redact audio. When rule-based redaction is enabled for a contact, Connect Customer stitches the redacted video with the redacted call recording if Contact Lens call recording redaction is also enabled for the contact, and with no audio otherwise. To redact audio from call recordings, see Use sensitive data redaction with Contact Lens.
-
Does rule-based redaction use artificial intelligence or machine learning?
No. Rule-based redaction is purely pattern matching. It evaluates each browser page URL and application window title against the rules you configure, and does not use artificial intelligence, machine learning, or generative AI.
-
Can I redact only part of a page, such as a single form field?
No. Rule-based redaction masks entire browser or application windows that match a rule. Field-level redaction is not supported.
-
Does rule-based redaction work in Citrix or other virtual desktop environments?
Yes, provided that the Connect Customer Client Application and the Connect Customer browser extension are installed in the virtual desktop session. Performance depends on the virtual desktop configuration.
-
What happens if an agent opens a matching URL in a private browsing window or an unsupported browser?
Private browsing windows are reported by the Connect Customer browser extension the same as regular windows and are redacted according to your URL rules. Browsers other than Google Chrome, Microsoft Edge, and Mozilla Firefox do not report URLs to the Connect Customer Client Application, so URL rules cannot match pages in those browsers. To cover those cases, add window title rules that match on the browser's window title, or use group policy to restrict agents to Chrome, Edge, or Firefox during recorded contacts.
-
How does redaction behave with multiple browser windows or tabs, or when a matching page is left open in the background?
Redaction follows what is visible on screen, not which window the agent is actively using:
-
Each browser window is evaluated independently. A matching window is masked; non-matching windows remain visible.
-
For multiple tabs in the same window, redaction follows the visible tab. If the visible tab matches a rule, the whole window is masked. If a matching page is open but is not the visible tab, that window is not masked.
-
A matching window stays masked whenever it is visible, even if the agent is working in another window beside it. Only the visible portion is masked — if the matching window is covered, minimized, or on an inactive tab, there is nothing on screen to mask.
-
On the first contact after an agent begins their session, if a matching page was opened before the extension observed it, that window may briefly not be masked at the start of the recording. Redaction applies as soon as the agent selects any tab in the window, and subsequent contacts are unaffected.
-
-
Does rule-based redaction pause or stop the recording when a matching window is detected?
No. The agent's screen is captured continuously throughout the contact. Matching windows are masked only in the final redacted recording that is produced after the contact ends; the capture itself is never paused.
-
Can I preview a redacted recording before the contact ends?
No. Redaction is applied when the recording is assembled after the contact ends.
-
Are agents notified when they visit a page that matches a redaction rule?
No. Agents are not notified when a page or window matches a redaction rule. Connect Customer does not provide a built-in notification for rule matches.
-
What browser information can the Connect Customer browser extension access?
The extension only reads the URL of each browser page and the title of the visible tab. It cannot access other browser data such as cookies, page content, or browsing history. These URLs and window titles are the only information used to match your redaction rules.
-
What APIs does rule-based redaction consume, and how are they authenticated?
Rule-based redaction introduces no new public APIs, and the browser extension consumes none. The extension communicates locally with the Connect Customer Client Application on the agent's workstation using the browser's native messaging channel; the extension is signed and scoped to the Connect Customer native messaging host. Redaction reuses the existing screen recording upload path, so it inherits that path's authentication and authorization — no additional credentials or permissions are required.
-
What happens if the Connect Customer browser extension fails or is not running?
Because redaction is applied after the contact ends, the extension does no real-time video processing during the contact — it only reports page URLs and window titles — so it is unlikely to affect the responsiveness of the agent's desktop. The possible failure cases are:
-
If the extension is not installed or not running, URLs are not reported, so URL rules cannot match and browser pages that should be redacted by URL may appear in the recording. Window title rules do not depend on the extension and continue to work.
-
The extension does not connect to Connect Customer directly; it communicates with the Connect Customer Client Application on the agent's workstation. If the Connect Customer Client Application is not running, screen recording does not function at all — not only redaction.
-
If the extension fails to report a matching URL during a contact, that page is not masked in the final output, and the unredacted recording still contains the full capture.
For troubleshooting, see Download log files for the screen recording app.
-
Configuration
-
Does rule-based redaction require Conversational Analytics (Contact Lens)?
No. The URL and window-title matching that drives rule-based redaction is independent of Contact Lens and works without it enabled. However, audio in the redacted recording does depend on it: when rule-based redaction is enabled for a contact, the redacted screen recording is stitched with the redacted call recording only if Contact Lens call recording redaction is also enabled for that contact — otherwise the redacted recording has no audio. The unredacted screen recording always uses the unredacted audio.
-
Can I change the redaction rule during an active contact?
No. The redaction configuration is fixed for the duration of a contact.
-
Where are redacted recordings stored?
Redacted recordings are stored in the same Amazon S3 bucket as unredacted recordings, under a separate prefix:
s3://your-bucket/connect/your-instance-alias/Analysis/ScreenRecordings/Redacted/year/month/day/contact-id_screen_recording_redacted_UTC-timestamp.mp4 -
How can I audit whether a specific contact was recorded with redaction enabled?
The flow that applied the redaction configuration is captured in the contact record. For information about accessing contact records, see Contact records data model.
-
Can I use rule-based redaction with the SuspendContactRecording and ResumeContactRecording APIs?
Yes. You can use the SuspendContactRecording and ResumeContactRecording APIs for screen recording in conjunction with rule-based redaction.
Performance
-
Does enabling rule-based redaction delay when a recording is available?
Your unredacted screen recording is unaffected — it is captured, delivered, and available on the same timeline as before. Redaction runs as a separate post-processing step, so the redacted recording becomes available once both the redacted screen recording and the redacted call recording have finished processing.
-
How much CPU and memory does the browser extension add?
The browser extension is lightweight — it only reports page URLs and window titles to the Connect Customer Client Application — and adds roughly 1% additional CPU and memory usage on top of the existing screen recording baseline. The most resource-intensive component remains the screen capture process in the Connect Customer Client Application. Follow the minimum system requirements in System requirements.