Define the brand and the campaign that will use it
Robly Branding combines a custom domain with branded campaign links and domain-authentication setup. Its documentation requires a domain the customer owns and controls; a personal address at a public mailbox provider is not a domain the customer can add to Branding. Start with the organization’s approved From domain, reply mailbox, and intended link hostname. Robly Branding guide
Create an inventory of brands and the campaigns that should use each one. Record the account owner and the person authorized to publish DNS. For an agency or multi-brand organization, identify who may send under each client’s domain. Keep these decisions separate from the visual appearance of the campaign: matching a logo and display name does not demonstrate permission to use a domain.
Check whether the proposed link hostname already serves a website, another mailing platform, or old campaign links. Agree how it will be used before generating and publishing records. A convenient label is not worth breaking an existing service. Plan sender replies as well, including who monitors the address after a campaign goes out.
Generate the brand’s records and preserve their identity
The documented path is My Account > Settings > Branding > Add New Brand. Enter the brand name, From domain, and link subdomain. Robly’s guide describes a generated set of four DNS records, consisting of three CNAME records and one TXT record. Use the actual values presented for the brand, not a recreated set based on the guide’s screenshots. Robly Branding guide
Copy each public owner, type, and target into a change ticket. Keep the brand identifier and the generation date beside the records. If the account currently produces a different set, pause and reconcile it with current provider guidance rather than forcing the old example onto a new configuration. Do not infer a universal selector or transfer another brand’s records.
At the authoritative DNS host, export the previous values and check for conflicts. Confirm how the editor handles complete names and zone suffixes. A CNAME owner must not be silently repurposed from another active service. Publish the approved set, then query the records independently and compare every owner and value with the generated instructions. Record TTLs and collection times so a reviewer can distinguish cached data from an incorrectly entered record.
Robly documents Verify and per-record Valid status, with Debug Validation available when a record fails. Treat those as provider checkpoints and preserve the error for the specific record. A failure on a link record and a failure on a signing record can need different corrections; repeatedly replacing the whole set makes that distinction harder to see. Robly Branding guide
Review SPF without copying the whole example
Robly’s separate SPF instructions identify include:_spf.robly.com and tell customers to merge with an existing policy. Preserve existing legitimate senders and confirm the appropriate DNS owner from the message route. Do not add sample a or mx permissions merely because a generic example contains them. Robly SPF instructions
Use RFC 7208 for the technical limits. Its DNS-lookup limit covers evaluated lookup-causing terms and nested evaluation, not just ten visible include words. Also distinguish the 255-octet limit on a single DNS TXT character-string from the length of an SPF policy carried in a TXT record containing multiple strings. Do not split one policy into multiple independent SPF records to address a length problem. SPF RFC 7208
The envelope domain still matters. A passing SPF result for a provider domain does not become aligned merely because the customer’s root domain contains an additional include. Preserve the evaluated identity in the evidence and determine whether DKIM supplies the aligned path. This prevents a campaign problem from turning into unnecessary expansion of the organization’s SPF permissions.
Select the brand on the actual campaign
Once the records validate, make the campaign selection part of the release check. Robly documents choosing a Brand in the campaign’s From settings; selecting No branding returns to the original settings. A valid brand sitting in the account is therefore not sufficient evidence that the campaign uses it. Robly Branding guide
Review the selected brand, visible From address, and reply behavior before the controlled send. If the campaign was copied from an older one, inspect its settings rather than assuming it inherited the intended new configuration. For multiple brands, test a representative campaign for each released sender domain and retain the campaign-to-brand mapping.
Robly documents alias handling for certain public-mailbox sender addresses using send-robly.com. If that route is relevant, inspect the actual received identity instead of treating the address typed into the editor as the final From domain. Do not claim that the organization has authenticated Gmail, Yahoo, or another mailbox provider’s domain. Robly DMARC and freemail guidance
Do not follow broad suggestions to relax a domain’s DMARC policy simply because one campaign bounces. First identify the authenticated domains and the failing route. The organization may have other critical senders whose policy should not change as a side effect of marketing setup.
Verify delivery identities and link behavior separately
Send harmless campaign content to authorized test mailboxes. Collect the visible From, envelope domain, every DKIM d=/s= pair, and the trusted receiver authentication results. Compare the signature with the brand’s public records. DKIM signature fields identify the actual signing domain and selector; a record in DNS alone does not show that this message used it successfully. DKIM RFC 6376
DMARC requires a passing aligned authentication path. Compare the passing SPF or DKIM identity with the visible From under the applicable alignment mode. Do not use link-host similarity, a brand name, or disappearance of a display label as a substitute for that comparison. DMARC RFC 9989
Then test the campaign’s links, including the expected tracking hostname and important recipient actions. Record these functional results separately from authentication. A correctly signed message can contain a broken link, while a working branded link can appear in mail whose sender is not aligned. Neither test establishes consent, recipient engagement, or guaranteed inbox placement.
Fictional diagnosis: a valid brand is not selected
A fictional team validates the DNS for news@studio.example and sees all brand records marked Valid. It duplicates an old campaign and assumes the new brand applies automatically. In this invented case, the campaign still uses its previous sender configuration, and the received message lacks the expected aligned signature. This is a teaching scenario, not a Robly tenant observation.
The investigator checks the campaign’s brand selection before rotating keys or adding another SPF record. After choosing the intended verified brand and confirming the From address, the team repeats the same controlled send. Closure requires the expected passing aligned identity and functional links. The original verification screenshot remains useful evidence of DNS setup, but not proof of the first campaign’s behavior.
Campaign authentication worksheet
- Brand: account owner, approved domain, link hostname, reply mailbox, and authorized users.
- Generated configuration: brand identifier, public record owners and values, issue date, and current provider instructions.
- DNS: previous values, published values, independent lookups, TTLs, and each verification result.
- Campaign: selected brand, visible sender, test time, recipient, and copied-template checks.
- Authentication: envelope domain, DKIM d=/s= values, trusted results, and alignment conclusion.
- Function and recovery: link tests, previous supported settings, rollback owner, and old-link dependencies.
Before retiring a brand, check existing campaigns and previously sent links. Remove unnecessary sending trust only after dependent traffic is accounted for. Tenant, DNS, received-message, visual, human SME, and publication gates remain pending for this documentary candidate.
Match reply and From addresses to the chosen brand
Robly documents that choosing a brand can prepopulate verified addresses or leave those fields blank. Both the From and Reply-to addresses must match the chosen brand domain. Review both after changing the brand rather than preserving an incompatible reply address from a copied campaign. This form check precedes a controlled received-message and reply test; it is not proof those tests passed.
Separate message acceptance from policy rollout
A passing aligned SPF result can satisfy a message evaluation, but does not alone justify moving an entire domain to reject. RFC 9989 Section 7.4 requires valid DKIM for domains publishing p=reject and warns about indirect-mail interoperability. Inventory all legitimate senders and indirect routes before the policy owner approves enforcement. No DNS change, tenant operation or received-message test was executed for this article.