Email Verification for Agencies Managing Client Lists

Agencies have a structural problem with email data: you did not build the list, you cannot verify how it was collected, and you will be held responsible for what happens when you mail it.

A client hands you 60,000 contacts and says the list is good. You send the first campaign. Bounce rate is 14%, complaints are above the threshold, and placement collapses. The client's read on that is that the agency ran a bad campaign. Your read is that you were given a bad list. Both of you are looking at the same numbers and neither of you can prove anything, because nobody measured the list before the send.

The fix is procedural rather than technical, and it happens before the first campaign.

01The onboarding audit

Run this before you send anything for a new client. Not as a service you charge for, though you can. As protection.

1. Verify the whole list, or a representative sample of it.

You need a number, and you need it dated. A 500-address random sample is enough to characterise a list of any size, and the free plan covers exactly that per month if you are auditing one client at a time. For a full audit at scale, 75,000 credits a month on the Max plan handles most agency volumes.

2. Write the result into the record.

Not a mental note. A dated document the client has seen. Something like:

List audit, 4 October. 60,412 contacts. Deliverable 71%. Undeliverable 19%. Risky 6%. Unknown 4%. Recommendation: remove undeliverable before first send, exclude risky and unknown from the initial campaign.

That paragraph is the single most valuable thing in your onboarding pack, and it costs you one verification pass. It converts a future argument into a documented starting position.

3. Ask where the list came from, in writing.

Three questions: where were these addresses collected, when, and was it single or confirmed opt-in. If the client cannot answer, that is important information about what you are about to send, and it belongs in the same document.

If the answer indicates purchased data of unknown provenance, stop and have a different conversation. Our post on inherited purchased lists covers why cleaning does not fix that situation.

4. Set the first-campaign expectation from the audit, not from a benchmark.

If the audit says 19% undeliverable, do not promise industry-average open rates on the first send. Say what the data supports. Clients handle a lowered expectation set in advance far better than a missed one explained afterwards.

02Keeping clients from contaminating each other

This is the risk unique to agencies and the one most likely to cause real damage.

Sender reputation attaches to sending domains and IP addresses. If you send multiple clients through shared infrastructure, one client's terrible list degrades placement for every other client on the same reputation. You can do excellent work for nine clients and have it undermined by the tenth.

Three levels of separation, in ascending order of effort:

Separate sending domains or subdomains per client. The baseline. Every client should send from their own domain, authenticated to their own domain. This is standard practice and also the right answer commercially, because the reputation being built belongs to the client rather than to you.

Separate ESP sub-accounts. Most platforms with agency tiers support this. It gives you separated reputation, separated reporting, and clean handover if the relationship ends.

Dedicated IPs for high-volume clients. Worth it above a few hundred thousand sends a month. Below that, a dedicated IP with insufficient volume warms poorly and can perform worse than shared.

The audit connects to this directly. A client whose list audits at 19% undeliverable should not go anywhere near shared infrastructure until the list has been cleaned and a small test send has come back with acceptable numbers.

03Verification as a service line

Most agencies absorb this cost. Some sell it, and selling it works better than you might expect, for a reason that is not obvious.

The reason is that the deliverable is a finding, not a task. Clients understand and pay for findings. "We audited your list and 19% of it is dead, here is what that has been costing you" is a piece of consulting. "We ran your list through a tool" is a line item they will try to remove.

If you want to price it, the framing that works is the cost model rather than the tool cost. Our cost arithmetic post works through a 50,000-address list where the reputation damage runs to roughly $4,000 a month while the ESP saving is about $80. Present the audit as recovering the first number and the price of the engagement stops being the question.

Two cautions. Do not present verification as a fix for engagement, because it is not and the client will notice. And do not promise spam-trap detection, because nobody can deliver it.

04Reporting that actually helps

Agency reporting on list health tends to be a single number, which is not enough to act on.

What clients can actually use:

The four-status split, tracked over time. Deliverable, risky, undeliverable, unknown, measured at each audit. The trend line is more useful than any single reading, because it shows whether capture is improving or the list is degrading.

Decay rate, once you have two data points. Verify the same sample ninety days apart and you have the client's actual monthly decay rate. That number justifies your ongoing cleaning cadence far better than a general recommendation. The decay measurement method is straightforward.

Undeliverable rate segmented by acquisition source. This is the finding that changes client behaviour. If contacts from the webinar form audit at 4% dead and contacts from the trade-show list audit at 31%, you have just told the client something about their marketing spend that has nothing to do with email. That is the kind of insight that extends a contract.

05The handover question

Agency relationships end. When they do, the client should be able to leave with a clean list and their own reputation intact.

Practically: the sending domain should be theirs, the ESP account should be transferable or exportable, and the suppression list should go with them. An undeliverable address that is suppressed only inside your agency's account will come straight back into their list on the next platform, and they will hit those bounces again.

Suppression lists are the thing most commonly lost in a handover. Export it and hand it over. It costs nothing and it is the difference between a client who speaks well of you and one who discovers on day one at the next agency that their bounce problem is back.

06The short version

Audit before the first send and document the result with a date. That single artefact converts the inevitable future argument about who caused a bad campaign into a settled question.

Keep every client on their own sending domain so one bad list cannot damage another client's placement. Sell the audit as a finding rather than a task. Report the four-status split over time and segment the undeliverable rate by acquisition source, because that is where the insight the client cannot get elsewhere lives.

And when the relationship ends, hand over the suppression list.

Run a client audit. Free plan is 500 credits a month for sampling, Max is $49 for 75,000 if you are auditing full lists across several clients.