Help center Customer portals Connect your customer portal to a custom domain: DNS, SAML and launch checks Updated Wednesday, October 28, 2026 · Marcel Gravelle A branded address such as help.example.com gives business customers a recognizable place to find knowledge and manage support requests. Core can verify a subdomain you control and arrange the TLS certificate that protects its HTTPS connection. The support operations owner coordinates the customer experience, while the DNS and identity owners make the corresponding infrastructure changes. Changing the address does not change which customers may see cases or articles. At a glanceFor: authorized portal administrators, support operations owners and the people who manage their organization’s DNS and sign-in provider. Applies to: Supportbench Core.Start here: Settings → Self-Service → Customer portal → select the portal → Portal address → Custom domain Understand the address, DNS and certificate · Prepare the people, access and change window · Give IT a complete, usable setup request · Request the new branded address · Add the displayed DNS records in order · Understand the provisioning status · Verify the customer experience before announcing it · Replace an active address or correct a pending request · Understand Remove before using it · Worked example: give a strategic account a recognizable support entry point · Troubleshoot setup and sign-in · Common questions before you launch Understand the address, DNS and certificate TermWhat it means for this setupPortal subdomainThe customer-facing hostname, for example help.example.com. Enter the hostname itself, without a path, port or https:// prefix.DNS providerThe service where your organization manages the domain’s DNS records. It may differ from the company that registered the domain. Ask your IT/DNS owner if you do not have access.TXT recordThe ownership-verification record. Core displays its exact name and value after the request is prepared. Copy both instead of inventing a verification string.CNAME recordThe record that points the chosen hostname directly to the target displayed by Core. It is different from a website redirect.TLS / SSL / HTTPSThe certificate-backed secure connection to the portal. The managed workflow arranges the certificate; it does not require you to buy a certificate, create a CSR or upload a PFX file.SAML ACS and Entity IDSign-in values that can depend on the portal hostname. They are separate from the TLS certificate and must match the identity provider when SAML is used.For example, example.com is an apex domain and help.example.com is a subdomain. Core rejects apex domains, wildcards, IP addresses, ports and Supportbench-owned domains for this field. These examples explain the format; they are not DNS values to copy into a real account. Prepare the people, access and change window Confirm you can edit the intended customer portal and that the Custom domain card is available. If it says custom domains are not enabled for the environment, ask the account administrator or Supportbench to review availability.Choose the exact subdomain and confirm that the DNS owner controls it. Check whether it already serves another application before changing any records.Record the current portal address and its working sign-in method. Keep a known working route available while preparing the change.If the portal uses SAML, include the identity-provider administrator. Agree how existing users will sign in while the old and new addresses are being coordinated.List the customer-facing places that will need the new URL after verification: website support links, onboarding material, saved account instructions and relevant email content.Choose a controlled test customer and an anonymous test route. You need to verify both useful access and intended restrictions, not just that a home page opens.Existing Legacy certificate or domainDo not remove an existing certificate or change a live CNAME as the first step of a migration. The old PFX-upload instructions describe a different setup. Coordinate a move from an existing Legacy binding with Supportbench and your DNS/sign-in owners, retaining a working fallback until the managed replacement is ready. Give IT a complete, usable setup request A support lead does not need to become a DNS specialist to own a successful portal launch. Identify the people who can make each change and send a clear request. The portal administrator selects the correct Supportbench portal and starts the managed-domain request. The DNS owner adds the displayed records. If SAML is enabled, the identity owner updates its application settings. The support lead verifies the actual customer journey and updates customer-facing links.Include in the handoffWhy it mattersPortal name and intended hostnameAvoid changing the production portal for a different product or audience.The current working URL and sign-in methodGive the team a tested reference while the replacement is being prepared.Exact TXT name/value and CNAME name/target from CoreEach request has its own evidence and destination. Copy these only after Core displays them.SAML ACS and Entity ID, when enabledThese identify where the provider sends the sign-in response and the application it represents.Owners, agreed change window and test customerSomeone must validate the complete result and deal with a failed check.Public and restricted content to testA secure homepage alone does not prove the customer has the right knowledge and case visibility.Do not send an administrator password, DNS account password, private key or sign-in token in this handoff. Each owner makes changes through their normal authorized access. If you cannot identify the DNS owner, stop before editing records and ask your IT team who maintains the domain. Request the new branded address Open Settings → Self-Service → Customer portal and select the correct portal.Choose Portal address, then find the Custom domain card. The nearby Supportbench subdomain/alias setting is a separate address; do not confuse its validation action with a custom-domain request.Enter the full hostname in Portal subdomain, for example help.example.com. Use the exact intended spelling.If SAML is enabled, read the proposed ACS and Entity ID displayed for this hostname. Have the identity administrator update the provider configuration, then select I updated the SAML provider metadata for this portal. Changing the hostname clears this acknowledgment, so check the displayed values again.Choose Request custom domain. This sends and saves the domain request immediately; it is separate from the portal’s Save Changes action.Read the returned status and DNS instructions. Wait while validation information is prepared, then copy the records shown for this specific request.Supportbench does not update your identity provider for you. The acknowledgment records that this prerequisite has been handled; it is not a button that performs the provider change. Use the SAML guide for the separate sign-in setup and provider-specific checks. Add the displayed DNS records in order Give the DNS owner the exact TXT name and value shown in the card. Add that TXT record in the DNS zone for your domain.Check how your DNS provider represents record names. Some providers append the domain automatically. Verify the resulting full name so the zone is not accidentally repeated.Return to the Custom domain card and follow its status. The editor refreshes provisioning status while the request is in progress.When the card asks you to point the direct CNAME, copy the exact CNAME name and target displayed there. Do not substitute your current Supportbench portal URL or another customer’s target.Resolve conflicting A, AAAA or CNAME records for that exact hostname with the DNS owner. Avoid changing records used by other services.Keep the CNAME direct. A proxy, record flattening or an intermediate redirect is not the direct CNAME that the managed workflow requires.Wait for DNS validation and certificate provisioning. Continue reading the status instead of repeatedly removing and recreating the request.Keep the exact direct CNAME in place after activation because it serves the portal and supports managed certificate renewal. As an operational recommendation, retain the TXT record in your DNS change record and coordinate later removal or replacement with the domain owner and Supportbench. A future ownership-validation request can require a new value; an old TXT value is not a permanent guarantee of validation. DNS propagation and certificate issuance do not have a fixed completion time promised by this guide. Understand the provisioning status Status shown in CoreMeaning and next actionPreparing DNS validationThe request is preparing its ownership-verification details. Wait for the specific records.Add the TXT record below to verify ownershipAdd or correct the displayed TXT record, then allow time for it to become visible.Point the direct CNAME below to SupportbenchOwnership validation has progressed far enough to provide the required traffic target. Add that exact direct CNAME.Azure Front Door is issuing the TLS certificateManaged HTTPS is being provisioned. Do not announce the address as ready merely because DNS was accepted.Active and protected by managed TLSThe managed domain is active. Perform the customer workflow checks before distributing the address.Action is required before this domain can continue / The domain request needs attentionCheck the exact hostname, DNS records and displayed status. Correct a confirmed cause and use Retry when it is offered. Ask Supportbench to investigate if the status persists; detailed failure information is not always displayed.Removing the managed domain / RemovedRemoval is in progress or completed. The domain is no longer a normal active setup; verify the intended replacement route.A successful request is not the same as an active HTTPS portal. Equally, a certificate-ready hostname does not prove that customer authentication, navigation or article visibility is correct. Verify those separately. Verify the customer experience before announcing it Wait for Active and protected by managed TLS, then open the new HTTPS hostname in a fresh browser session. Confirm the intended portal branding and a normal secure connection.As an anonymous visitor, test the knowledge and request-entry experience your portal is meant to provide. Confirm restricted content is not unexpectedly visible.Sign in with an authorized test customer. If using SAML, complete the full provider-to-portal round trip and confirm the correct customer identity.Open an allowed article and a representative permitted request. Check that the customer can see the intended Company/Contact context and cannot access unrelated customer work.Check the portal’s navigation, files/images, request links and sign-out behavior on desktop and a narrow screen. A working home page alone is insufficient.If testing request submission, use an approved test record and verify its routing, ownership and customer acknowledgment. Record the test rather than leaving unexplained production work.Update the intended customer-facing links only after verification. Keep a record of the hostname, portal, DNS owner, sign-in owner and date tested. Replace an active address or correct a pending request When a managed domain is Active, the action changes to Replace custom domain. Enter the replacement hostname, review any new SAML values, and follow the new TXT/CNAME process. Core keeps the current active domain serving while the replacement is provisioned, then switches the portal after the new domain and managed TLS are ready and schedules cleanup of the old binding.That overlap is not a promise that the old managed address will remain an indefinite redirect. Plan link updates and customer communication, and do not remove old DNS or certificate material prematurely during a coordinated Legacy migration.A different hostname cannot overwrite a still-pending request. If the pending hostname is wrong, first confirm the impact and remove that pending request through the managed workflow. Core can accept a different hostname after the old request enters Removing, while cleanup continues. As an operational safeguard, monitor removal and wait for Removed before starting the next request so your team can verify each change separately. Resubmitting the same pending hostname can update the SAML acknowledgment; it does not provide a second independent pending domain. Understand Remove before using it Remove starts a separate, immediate removal workflow after confirmation. The warning explains that the custom domain will stop serving the portal after the managed endpoint is removed. It is not an unsaved field change that Cancel or leaving the page will discard.Confirm which hostname is being removed and why. Have a verified alternative route for customers who use it.Review website links, bookmarks, sign-in provider settings and other references with their owners.Use Remove only when you intend to stop that hostname serving this portal, and read the confirmation.Wait for the removal status to complete, then verify customer entry through the intended alternative. Worked example: give a strategic account a recognizable support entry point A supplier supports several contacts at each business customer. Its support lead wants everyone to use help.example.com rather than different bookmarked addresses. The lead selects the right portal and documents the intended customer access; IT adds the exact TXT/CNAME records; the identity owner updates SAML values; and a test contact verifies knowledge and permitted requests after managed TLS becomes active.The team then updates onboarding and account-review links. The benefit is one verified entry point with clear ownership. The branded address does not expand customer access or replace good Company/Contact setup, and the team does not announce completion until both sign-in and the actual support workflow work. Troubleshoot setup and sign-in ProblemLikely cause and corrective stepThe hostname is rejectedCheck for a path, port, wildcard, IP address, apex domain or Supportbench domain. Use an organization-controlled subdomain.TXT validation remains pendingCompare the complete DNS name and value with the card. Check accidental domain duplication, the correct DNS zone and propagation.CNAME validation does not completeCheck the exact direct target, conflicting records, proxy/flattening behavior and propagation. A browser redirect does not satisfy the record requirement.The page has a certificate warningDo not instruct customers to bypass it. Confirm Active status and the exact hostname; investigate DNS/certificate state before distributing the URL.The portal opens but SAML failsCompare the identity provider’s ACS and Entity ID with the values for the new hostname. Supportbench does not update the provider automatically.Changing the hostname cleared the acknowledgmentExpected: the sign-in values may have changed. Review and update the provider before acknowledging again.A different pending hostname cannot be requestedRemove the current pending request through the managed workflow. A new hostname can be requested after the old request enters Removing; as an operational safeguard, wait for Removed and verify cleanup first.Retry is offeredReview the status and verify the DNS and hostname configuration. Correct a confirmed cause and then retry. Repeated retries do not correct DNS or identity-provider settings.Setup remains Failed or Action requiredGive Supportbench the portal name, requested hostname, displayed status, any visible error message and time observed. Do not send certificate private keys, passwords or sign-in tokens.The new address works but a customer cannot see an article or caseCheck the existing portal, article and customer access rules. A hostname change does not grant visibility. Common questions before you launch QuestionAnswerDo I need to purchase an SSL certificate?The Core managed-domain flow provisions managed TLS. It does not ask you to purchase a certificate or upload a PFX file. Follow the separate HTTPS guide if you are moving an existing Legacy certificate binding.Can I use example.com or *.example.com?No. This Supportbench control accepts a hostname subdomain you control, such as help.example.com. General Azure capabilities do not widen the formats this control accepts.Can I point the domain at my current Supportbench alias?Use the exact direct CNAME target displayed for the managed request. The current portal alias, an HTTP redirect and an intermediate CNAME are not substitutes for that instruction.Will customers automatically get more access?No. Portal, customer, article, Topic and Division restrictions remain separate from the address. Test authorized and excluded readers.Does Active mean SAML has been tested?No. Active confirms the managed domain state. Complete a real sign-in round trip with the intended test identity before announcing the new URL.Can I just press Save Changes?The Custom domain card has its own Request custom domain, Replace custom domain, Retry and Remove actions. A general portal save does not replace those lifecycle steps.How long will it take?There is no guaranteed completion time in this procedure. DNS visibility and managed provisioning can take time. Use the current status and investigate a persistent failure with the responsible owner instead of repeatedly deleting the request. Continue learning Understand managed HTTPS and troubleshoot portal certificate problemsConfigure the customer portal and its audienceConfigure and verify customer-portal SAMLCustomize portal header links and brand colorsBuild product-specific knowledge portals Was this article helpful? Yes No Thank you for your feedback.