The pre-launch compliance checklist that catches problems before the campaign runs

A UK television spot claiming a cheaper basket than a named rival cannot go to air until Clearcast has cleared it on behalf of the broadcasters. Clearcast reads the script first and the finished film second, and it wants the evidence for the price claim attached to the script, before anyone books a studio. That ordering rule is the whole logic of pre-launch compliance in one line: the proof has to exist when the claim is written, not when it airs.

Most retail campaigns never touch broadcast, so nobody imposes that discipline from outside. The sequence below imposes it internally: who signs, what file exists, and where in the production calendar a comparative price claim gets checked, then checked again.

Why retail campaigns need a formal gate

Rules bind at the moment of publication. Creative locks weeks earlier. Between those two points sit a shoot, a media booking, a print run and a store pack, and each one turns a cheap correction into an expensive one. Killing a claim at brief stage costs an email thread. Killing it after the shoot costs the shoot. Killing it once point-of-sale packs are picked and on vans costs the reprint, the store labour to swap them, and you still have media in market.

The stake behind the gate has moved recently. Under the Digital Markets, Competition and Consumers Act 2024, the CMA can decide consumer protection breaches itself and fine up to 10% of global turnover, without going to court first. A gate exists so that the decision to run a risky claim is made on paper, by someone senior enough to own that number, before the money is spent.

The sign-off sequence, step by step

Names vary between retailers. The order does not.

Step 1: The substantiation file

What counts as a claim and what counts as adequate evidence is settled in the foundations lesson. The operational question is where that evidence physically lives. One file per campaign, indexed by claim rather than by asset, because the same "30% off" appears in nine assets and nine copies drift apart. Each entry holds the claim exactly as worded, the source data, the date it was pulled, the person who pulled it, and an expiry date after which the claim cannot be reused.

Comparative prices need the date most of all. A basket priced on Monday and broadcast on Friday can be wrong by Friday. Aldi's price comparison advertising carries the comparison date and the data source in the supers for exactly this reason: that small print is doing compliance work, and it only holds up if the underlying basket data is filed alongside it.

Two things go wrong here. First, the reference price. In the EU and Northern Ireland, any announced reduction must show the lowest price applied in the previous 30 days, under the Price Indication Directive as amended by the Omnibus Directive, in application since May 2022. Great Britain has no fixed statutory window, so the test falls back on whether the higher price was ever a genuine selling price, which puts the weight on your own price history rather than a calendar rule. Second, multibuys: a "3 for 2" launched a fortnight after the unit price moved up needs that unit price history in the file, or the stated saving is fiction.

Step 2: Claims review, internal then external

Compliance reads the final creative against the file, never the brief. The gap between the two is where most problems are born: "up to 70% off" is defensible in a brief and indefensible on a hero banner if three SKUs out of 400 reachreachThe number of unique people exposed to your message in a given period. Unlike impressions, reach counts each person once, no matter how often they see it.View full definition → 70% and the banner shows a coat that is down 20%.

For broadcast and most VOD, the external review is Clearcast, and it helps to know what clearance is and is not. Clearcast can approve a script conditionally, requiring an on-screen qualification of a given size and duration, which changes the edit and sometimes the shot list. Clearance is also not immunity: a cleared ad can still be investigated and banned by the ASA afterwards, including on a competitor's complaint. In grocery, competitors complain quickly. Aldi and Lidl read each other's price claims far more carefully than any shopper will.

Red flags a reviewer should stop on:

• absolute superlatives with no dated data behind them
• environmental claims with no third party verification (the FTC's Green Guides in the US, CAP Code rules on qualification in the UK)
• comparisons drawn against products that are not genuinely like for like
• countdown timers and low-stock counters not wired to real inventory

Step 3: Fair treatment and data check

The fairness constraints on promotions and loyalty comms are the subject of the sibling lesson on fair treatment. The checkpoint here is narrower: does every recipient on the send list have a consent state that supports this message on this channel, and can you show when that state was captured?

The recurring failure is timing. A CRMCRMCustomer Relationship Management: software and strategy to manage and analyse customer interactions throughout their lifecycle.View full definition → segment is often built days before send, while consent, suppression and bereavement flags keep moving. Freeze the segment on Tuesday, send on Saturday, and you have mailed people who opted out on Wednesday. The fix is procedural: rebuild the audience against live consent inside a defined window before send, and log that rebuild as its own approval rather than folding it into creative sign-off.

Step 4: Channel and territory check

• Broadcast and VOD: Clearcast at script and again at finished film. Budget schedule for both. A conditional clearance at film stage means a re-edit and a resubmission.
• Email: CAN-SPAM in the US requires working opt-out and accurate sender information; GDPRGDPREU regulation governing how organizations collect, store and use personal data, with fines tied to global revenue for breaches.View full definition → governs the consent basis in Europe.
• SMS: the TCPA in the US requires prior express written consent for marketing texts.
• Influencer and affiliate: paid partnership disclosure under the FTC Endorsement Guides and the ASA's equivalent rules. Affiliates build creative from your feed, which makes them the slowest surface to correct.
• In-store: shelf-edge labels and POS packs are printed and shipped days ahead. If a claim changes at T-3, digital changes and cardboard does not.

Step 5: The sign-off matrix

The final gate is a grid listing every campaign asset against every required approver, with a version number against each row. No asset ships with an open cell, and an approval attaches to a version, not to a name.

Give each territory its own row. A price claim cleared for Great Britain is not cleared for Northern Ireland or the Republic. The matrix is usually owned jointly by legal and the campaign project manager, and it becomes the audit trail when someone later asks who approved this.

What happens when the checklist is skipped

Enforcement outcomes are read closely in the enforcement lesson. The process failures that produce them are worth naming here, because they repeat:

• verbal approval under deadline pressure, no matrix entry, nobody able to answer who signed
• version 3 approved while version 5 ships
• one territory cleared and the rest assumed
• the substantiation file assembled after the ad ran, from memory

The second-order cost is that pulling a campaign is not a single action. A homepage changes in minutes. Paid social has a review lag. Affiliates cache your feed and keep serving the old banner for days. Catalogues and POS are physical objects in vans. Email that has already sent cannot be recalled at all. So the real question at the gate is not only whether a claim is defensible, but what you could withdraw on day three and how fast. A claim that is hard to retract deserves a higher evidence bar than one living on a page you control.

🎬 [VIDEO: "How the FTC Regulates Advertising" - youtube.com/@FTCvideos - an accessible overview from the FTC itself explaining core deceptive advertising rules retailers must follow]

A practical mini-checklist for any campaign

Before a multichannel retail campaign goes live, six checks:

1. Does every price, discount and comparison claim have a dated substantiation entry naming its source and its owner?
2. Has compliance reviewed the final creative in each locale, not the brief and not one master?
3. For any comparative price, has it been re-verified against live prices within 24 hours of launch?
4. Was the audience rebuilt against live consent inside the send window?
5. Are disclosures satisfied on the slowest channel to correct, not the fastest?
6. Is the matrix fully green against the exact versions that will ship?

If any answer is no, the launch date moves, not the checklist.

Key takeaways

• The evidence for a claim has to exist when the claim is written. Clearcast enforces that ordering on broadcast by asking for substantiation with the script; internally, nothing enforces it except your own sequence.
• Comparative price claims get checked twice: once when the data is pulled and filed with its date, once against live prices immediately before launch.
• Reference prices are jurisdictional. The EU and Northern Ireland require the lowest price of the previous 30 days; Great Britain tests whether the higher price was genuinely charged, which makes your internal price history the evidence.
• Clearance is not immunity. An ad approved for broadcast can still be ruled against afterwards, often because a competitor rather than a consumer complained.
• Weigh evidence against retractability: email cannot be recalled and affiliate feeds keep serving old creative for days, so the assets you cannot pull quickly need the strongest file behind them.