What's a fast way to test rendering across email clients without a dedicated testing tool?

Send to a small internal seed list covering the handful of clients your subscriber base uses, rather than assuming one desktop preview represents every inbox. A seed list of five or six accounts across the most common clients and one mobile device catches the rendering breaks that matter most, even without a paid preview service.

Should QA happen before or after the campaign is scheduled?

After — scheduling itself can introduce errors, such as a timezone default that does not match what was intended or a list swapped during the scheduling step, and those errors will not show up in a check run only against the draft. Running the checklist against the fully scheduled, queued version catches a different class of mistake than checking the content before it was ever scheduled.

What happens when QA finds a problem an hour before launch?

The checklist should include a severity call: cosmetic issues can ship with a fix queued for the next campaign, but a broken tracking link, the wrong audience, or a missing compliance footer should hold the send even at the cost of missing the planned time. Missing the slot costs one window, and the next send recovers it; a dead link or a wrong audience cannot be recalled once the send has left, and the record it leaves behind stays wrong afterward.

Does a resend to non-openers need the same QA pass as the original send?

The content does not need re-approval if nothing changed, but the mechanical checks do need to run again — dates, any time-limited offer mentioned in the copy, and every link should be reverified, since a link that was live when the original went out can have expired or moved by the time the resend fires days later.

How to QA marketing campaigns before they ship

Guide · Market Growth · 4 min read · last verified 2026-07-29

Reviewed before publication Editorial board Independent commercial review

In shortA practical pre-flight checklist for marketing campaigns: what to check against the live assets before launch, who should sign off, and where mechanical QA ends and editorial or account-setup checks begin.

The campaign failures a checklist can catch are the mechanical ones — a broken link, the wrong audience segment, a stale date left over from an earlier draft, an email that renders fine on a desktop preview and breaks on a phone. QA is the pass that catches that class of problem, and it only works as a named step run against the live assets about to ship, not against the plan that was approved for them days earlier.

What a pre-flight pass checks

A working QA pass runs down a short, specific list against the real thing, not a description of it:

Links resolve to the intended live page, not a staging URL or a 404

Tracking parameters are present and correctly formatted on every link

The audience or segment matches the brief, not a leftover test list

Send date, send time, and timezone match what was intended

Personalization tokens render with real data, not the raw {{token}}

The unsubscribe or compliance footer is present and functional

None of these require judgment about whether the campaign is a good idea — each one has a checkable right answer, which is what makes the pass fast to run and fast to hand off to someone other than the person who built the campaign.

Test the object that will ship

Content gets approved once and then edited again afterward for a last-minute fix — a subject line swapped, a segment adjusted, a link updated — and the version that goes out is not always the version anyone reviewed. QA has to run against the actual built object: the rendered email, the live landing page, the ad creative as it sits inside the ad platform, checked after every change rather than only once at the start. Tracking parameters are exactly the kind of detail a late link swap can break silently, which is why they belong on the pre-flight list rather than assumed carried over from an earlier check — see what a UTM parameter is for what those tags need to look like once they are on the link. The same goes for the files a campaign pulls in: a logo or an image taken from an organized asset library can be checked against a known current version, where one saved to a laptop months ago cannot.

Who signs off, and why one name matters

A checklist with a name attached to it and a completion date is a different artifact than a thumbs-up in a message thread that scrolls out of view within a day. The name matters less as credit and more as a decision record: when something breaks anyway, the useful question is which check missed it and how to fix the checklist, not who to blame — and that question only has an answer if the sign-off was recorded somewhere durable in the first place.

QA is not the same job as checking whether the content is true

Reviewing AI-written content before publishing is an editorial pass — are the claims accurate, are they sourced, does the piece say what it means to say. QA is a mechanical pass — does the link go where it claims to go, does the segment match the brief, does the page render. Both gates need to exist, and neither substitutes for the other: a factually accurate email pointing at a broken link has failed exactly as completely as an inaccurate one that renders perfectly on every device.

Where this sits next to ad-account setup

Campaign readiness before connecting ad accounts is account-level plumbing — permissions, pixel installation, billing — done once when an account gets connected. The QA pass described here runs every single time a campaign ships, not once at setup, because a correctly connected account says nothing about whether today's specific campaign has the right link, the right segment, or the right date on it.

Let the checklist grow after every miss

When something breaks that the checklist did not catch, the fix is not only patching that one campaign — it is adding a line to the list so the next campaign gets checked against the same failure before it ships. A checklist frozen at its first version just waits for the same mistake to reappear under a different subject line. A borrowed checklist from another team is a reasonable starting point for the first six or seven items — links, tracking, audience, dates — but a poor permanent substitute for one built from a team's own record of what has gone wrong, since that record is the only version tuned to the failures these campaigns keep producing.

Why the gate exists at all

As AI-assisted tools shorten how long a campaign takes to produce, the number of campaigns a team can generate grows faster than the number of people available to check each one by hand. The QA gate is what keeps that increase in output from becoming the point in itself — a named, mechanical pass that scales with volume in a way that trusting each new campaign by default does not.