Email deliverability for digital agencies
Where the addresses come from
- Client-supplied exports, usually with no documented consent path and no answer about where the data came from.
- Lists inherited from a previous agency, carrying years of accumulated decay and nobody left who remembers the sources.
- Lead generation campaigns the agency ran itself, which are the only rows with a provenance story attached.
- Event and conference lists a client acquired through a partnership, which is the weakest consent story in common circulation.
The reputation is yours and the data is theirs
When an agency sends from its own platform account, mailbox providers judge the agency. One client's purchased list damages placement for every other client on the same systems, and the client who caused it experiences no result at all.
Campaign Monitor and others separate clients for reporting purposes, which helps the invoicing and does nothing for the sending reputation underneath.
The practical defense is a written standard for lists you will accept: age, source, consent evidence and a verification pass before the first send. It converts an argument during an incident into a talk during onboarding.
What the results tend to look like
Varies more than any other sector, because it is whatever the client's business is. A B2B software client and a retail client produce completely different verdict distributions from the same agency account, which is why per-client measurement matters here more than anywhere else.
What to do about it
Write a list acceptance standard before you need one
Age, source, consent evidence and a verification pass as conditions of onboarding. It is far easier to agree during a sales talk than during an incident.
Test-send a few thousand addresses first
On an inherited list this is the cheapest insurance available. You get the bounce rate before the whole list touches your systems.
Bill checking through as a line item
At $89 for 50,000 addresses it is small next to a campaign fee and it converts an unknown risk into a disclosed cost.
Use separate sending subdomains per client where volumes justify it
It does not fully separate reputation, and it does give you a way to isolate a problem rather than watching it spread.
Questions people ask
Should an agency check every client list?
Every inherited one, certainly. You did not collect it, you cannot vouch for it, and the sending reputation risk lands on your account rather than the client's.
What if a client refuses?
Then you are deciding whether to put your systems behind data you cannot inspect. A small test send is the minimum compromise worth accepting before a full campaign.
Does checking prove a list is compliant?
No. It measures delivery. Consent is a separate question and it is the one that carries legal risk, so a clean result is not an answer to it.
Related sectors
See the breakdown on your own list
100 free checks a month, no card. Addresses we could not get an answer on come back labeled rather than guessed at, and we do not bill them.