Campaign Pre-Flight: The Checks Worth Running Before Anything Goes Live
Most campaign errors that reach an audience are caught by nobody because checking was everyone's job. A pre-flight list built around what actually breaks — tracking, deliverability, and the offer itself.
By Robin Deane — Founder & Marketing Strategist, RD
The errors that reach a live audience are almost never the ones nobody thought of — they are the ones everybody assumed somebody else had checked. A broken tracking parameter, a link to a staging URL, an offer that contradicts what the landing page says, a send to a segment that includes people who unsubscribed. Each is trivially catchable and each recurs because checking is distributed across a team rather than owned by one person against a written list. A useful pre-flight is short, specific to what has actually broken before, run by a named person who did not build the campaign, and completed at a point where a failure is still fixable. Anything longer gets skipped under deadline pressure, which is precisely when the errors happen.
There is a particular kind of marketing error that is almost comic in retrospect. The email that went out addressed to "FIRST_NAME". The paid campaign that ran for six days pointing at a staging URL. The launch where every link worked and none of them were tracked, so nobody could say what the campaign did. The discount code in the subject line that the checkout didn't recognise.
What these have in common is not carelessness. They happen on campaigns run by competent people who were paying attention. They happen because in the final hours before a launch, the question "has anyone checked the links?" is asked in a group setting, receives no clear answer, and is understood by everyone present as having been resolved.
Why Do Obvious Errors Get Through Competent Teams?
Because responsibility for checking is distributed, and distributed responsibility reliably becomes nobody's.
This is not a marketing-specific phenomenon — it is the reason aviation, surgery, and construction all converged on written checklists run by a named individual. The insight those fields arrived at is that expertise does not prevent omission. A skilled person under time pressure, doing something familiar, will skip a step and not notice they skipped it, because the step is so routine that its absence produces no feeling of absence.
Marketing launches have every condition that produces this: time pressure, familiarity, multiple people each assuming coverage, and a final phase where attention is on the exciting parts (creative, messaging) rather than the mechanical ones (parameters, segments, links).
The corollary is important and slightly counter-intuitive: the more experienced the team, the more necessary the written list, because experienced teams are the ones confident enough to run on assumption.
What Should Actually Be on the List?
Not everything. A pre-flight list that tries to be comprehensive becomes a document nobody completes, and a half-completed checklist is worse than none because it creates a false record of verification.
Build it from what has actually gone wrong — in your own campaigns first, then the common categories below. Five to fifteen items is the working range.
| Category | What to verify | Why it's on the list |
|---|---|---|
| Destinations | Every link resolves, on production not staging, to the intended page — clicked, not read | Staging URLs and typos survive every visual review because a link looks correct until it is followed |
| Tracking | Parameters present and correctly formed on every link; a test conversion appears in the destination report | Untracked campaigns cannot be evaluated afterwards, and this is only fixable before launch |
| Audience | Segment size matches expectation; suppression and unsubscribe lists applied; no test addresses left in | A segment that is 10x expected size is almost always a filter that silently failed |
| Offer consistency | The claim in the ad, the email, the landing page, and the actual checkout or form all agree | These are written by different people at different times and drift without anyone deciding to change anything |
| Rendering | Email in at least one dark-mode client and one narrow viewport; landing page at 390px | Approval usually happens on one large desktop screen, which is not where most of the audience will see it |
| Deliverability | Authentication passing, sending domain warm, send volume in line with recent history | A campaign that lands in spam produces the same reporting shape as a campaign nobody wanted |
| Reversibility | Who can pause this, using what access, outside working hours | The moment you need this, you need it immediately — and it is never the moment to discover the answer |
The last row is the one most often absent and the one that turns a mistake into an incident. Knowing before launch who can stop a campaign at 9pm on a Friday, and confirming that person actually has the access, costs one message.
Who Should Run It?
Someone who did not build the campaign.
This is the single most important design choice in the whole practice, and the one most often compromised for convenience. The person who built something cannot reliably check it, because they will read what they intended to write rather than what is on the screen. This is not a discipline failure — it is a well-documented property of proofreading your own work, and no amount of care fully overcomes it.
The reviewer does not need to be senior, or a specialist, or deeply familiar with the campaign. In several respects it is better if they are not: an unfamiliar reviewer follows the list literally rather than skipping items that "obviously" apply. A junior team member with a clear checklist catches more than a senior colleague working from memory and goodwill.
When Should It Run?
A check that finds a problem an hour before send produces a decision between launching broken and launching late. Schedule pre-flight far enough ahead that a failure is fixable — typically 24 to 48 hours before, depending on what the fixes involve.
Checking something that is still being edited verifies nothing. Agree that the campaign is locked at pre-flight, and that any change after it re-triggers the affected checks rather than being waved through as minor.
Ticked by the person who checked, with their name against it. Not because anyone will audit it, but because a list you must mark completed is one you actually perform, while a list you read is one you skim.
Click the link, submit the form, complete the purchase, confirm the record arrives in the CRM and the conversion appears in reporting. Reading a tracking parameter and following it are different activities with very different catch rates.
Whenever something reaches an audience broken, add the check that would have caught it — and delete an item that has never once caught anything. A pre-flight list that only grows will eventually be abandoned under deadline pressure.
Step five is what keeps the practice alive. The list should reflect your team's actual failure history, not a generic template downloaded from somewhere. If an item has never caught anything in two years, it is costing attention that belongs to the items that do.
Doesn't This Slow Everything Down?
It adds an hour and removes a category of delay that costs considerably more.
The comparison worth making is not "pre-flight versus no pre-flight" but "an hour before launch versus the cost of the failure it prevents." A broken tracking parameter discovered after a campaign ends does not cost an hour — it costs the ability to evaluate the campaign at all, which means the next budget decision is made without evidence. A staging URL in live paid media costs the spend plus the recovery. An offer inconsistency between email and checkout costs conversions and, occasionally, a complaint you have to answer formally.
There is also a second-order effect that shows up over a few months: teams with a reliable pre-flight step become willing to launch closer to the deadline, because they trust the check. Teams without one accumulate an informal safety margin of nervousness — extra review rounds, additional sign-offs, launches held back a day "to be safe" — that costs far more in aggregate than the checklist ever would.
What Does This Look Like Working?
- Live campaign errors are rarely things nobody considered — they're things everybody assumed someone else had checked
- Distributed responsibility for checking reliably becomes nobody's; this is why aviation and surgery use named-owner written checklists
- The more experienced the team, the more necessary the written list, because experienced teams run on assumption
- Build the list from what has actually broken in your campaigns — five to fifteen items, not a comprehensive document nobody completes
- A half-completed checklist is worse than none, because it creates a false record of verification
- The reviewer must be someone who didn't build the campaign; you read what you intended to write, and care doesn't overcome that
- Run it 24–48 hours out against a frozen version, not on launch day when a finding forces a choice between broken and late
- Test end-to-end for real — click the link, submit the form, confirm the conversion lands; inspection catches far less than execution
- Add a check after every incident and delete items that have never caught anything, or the list grows until it's skipped
- Include who can pause the campaign out of hours, and confirm they actually have the access
The measurable outcome is a drop in post-launch fixes — the small corrections, resends, and "we've updated the link" follow-ups that teams stop noticing because they feel like normal operating cost. They are not normal. They are the visible residue of checks that didn't happen, and a team running a real pre-flight sees them fall close to zero within a couple of quarters.
The less measurable outcome matters more: launches stop being tense. The nervousness around a send button is almost entirely uncertainty about whether something was missed, and a completed checklist with names against it resolves that directly. That is worth more than the hour it costs.
Our campaign management service builds this into the delivery process rather than leaving it to individual discipline. For the causes of the deadline pressure that makes teams skip it, see why campaign launches slip; for catching the failures that recur across campaigns, the campaign post-mortem nobody runs. On the deliverability row specifically, why good campaigns land in spam covers what to verify before a send.
Frequently Asked Questions
What is a campaign pre-flight check?
It is a short written checklist, run by a named person against a frozen version of a campaign before it goes live, covering the specific things that have historically broken — link destinations, tracking parameters, audience segments, offer consistency, rendering, deliverability, and who can pause the campaign. It is deliberately narrow rather than comprehensive, so that it is actually completed under deadline pressure.
Why do obvious errors get through experienced marketing teams?
Because responsibility for checking is distributed across several people, and distributed responsibility becomes nobody's. Expertise does not prevent omission: a skilled person under time pressure doing something familiar will skip a routine step without noticing, since the step is routine enough that its absence produces no sense of absence. Experienced teams are more exposed to this, not less, because they are confident enough to work from memory.
Who should run the pre-flight check?
Someone who did not build the campaign. People read what they intended to write rather than what is actually on screen, and care does not overcome that. The reviewer does not need seniority or deep familiarity — an unfamiliar reviewer follows the list literally instead of skipping items that seem obviously fine, so a junior colleague with a clear checklist often catches more than a senior one working from memory.
When should a pre-flight check happen?
Typically 24 to 48 hours before launch, against a version that has been frozen. Running it on launch day means a finding forces a choice between launching broken and launching late. Any change made after the check should re-trigger the affected items rather than being waved through as too minor to matter.
What belongs on a campaign launch checklist?
Link destinations verified by clicking, tracking parameters confirmed with a real test conversion, audience segment size and suppression lists, consistency of the offer across ad, email, landing page and checkout, rendering in dark mode and at narrow widths, deliverability authentication, and who can pause the campaign outside working hours. Beyond those, add whatever has actually broken in your own campaigns before.
How long should a pre-flight checklist be?
Five to fifteen items. Longer lists get skipped under deadline pressure, and a partially completed checklist is worse than none because it creates a false record that verification happened. Add an item after any incident that reaches an audience, and delete any item that has not caught anything in a long period, so the list stays proportionate to the time available.
Does a pre-flight process slow campaigns down?
It adds about an hour and removes larger costs — untracked campaigns that cannot be evaluated, wasted spend on broken destinations, and post-launch corrections. Over time it usually makes teams faster, because a trusted check removes the informal safety margin of extra review rounds and cautious delays that teams accumulate when they are unsure whether anything was missed.
Keep Reading
Want this handled properly?
If this is the kind of problem you're wrestling with, a short conversation is usually enough to tell whether there's a real opportunity here.