Media buyers in mainstream categories can treat the landing page as a variable to test. Swap the hero, move the form, measure the lift. The page affects performance, but it does not usually affect whether the campaign is allowed to run.
In restricted categories that assumption breaks. The landing page is part of the ad review. It determines whether the account stays open, whether the payment processor keeps the merchant, and whether the spend you have carefully optimised has anywhere to land. Design stops being a conversion lever and becomes a precondition.
UK prize competitions are a good vertical to reason through, because every constraint is explicit and documented. The principles transfer to any category where a platform reviews where the click goes.
The Reviewer Opens the Page
This is the fact that changes the brief. When Meta reviews a restricted-category advertiser, a human looks at the destination. So does the payment provider during merchant onboarding. Two separate reviewers, overlapping checklists, both assessing the site rather than the ad.
For a competition operator they are looking for a skill question in the entry flow, a free entry route displayed at equal prominence to the paid one, terms specific to that competition rather than a boilerplate page, published closing and draw dates, an 18+ restriction, and a privacy policy that explains what happens to entrant data.
If any of that is missing or buried, the outcome is not a lower conversion rate. It is a declined application or a restricted ad account, and no amount of budget fixes either.
Which means the compliance elements are not things the client adds later. They are structural components of the template, and they have to be designed properly rather than bolted on, because a badly integrated compliance element reads as evasive to a reviewer and as clutter to a user.
Equal Prominence is a Design Problem, not a Legal Footnote
The single most interesting constraint in this vertical: under Section 339 of the Gambling Act 2005, the free entry route has to carry the same visual weight as the paid one, with identical odds of winning.
You are being asked to present a free alternative to the thing you are spending money to sell, on the same screen, at the same hierarchy level. No smaller type, no muted colour, no link in the footer.
Most attempts at this fail in the same way. The designer treats it as an obligation to satisfy minimally, produces something technically present and visually demoted, and it reads as exactly what it is. Reviewers notice. Users notice too, and in a category built entirely on trust, looking evasive costs conversions.
The layouts that work treat the two routes as a genuine choice presented in parallel. Counter-intuitively, honest presentation tends to perform better, because the free route is used by a small minority and its visible presence reassures the majority who were going to pay anyway.
Urgency has to be Real
Countdown timers are standard in this category and they are also the component most likely to attract regulatory attention.
The CAP Code polices how urgency and scarcity are communicated. A timer next to a genuine deadline is useful information. A timer that resets, a ticket counter that does not update in real time, or scarcity messaging on a draw closing in eleven days are all things the ASA has acted on.
For a media buyer this matters because ASA rulings are public and platform policy teams read them. A dark pattern that lifts conversion by four percent is not worth an enforcement action that closes the acquisition channel.
The workable version is honest urgency, well designed. Real deadline, accurate counter, clear presentation. It converts nearly as well and it does not carry tail risk.
Speed is a Campaign Constraint, not a Technical one
Competition traffic has an unusual shape. Volume concentrates into the final hour before a draw closes, sometimes at ten times normal load, and it lands on basket, checkout and ticket-count pages that cannot be cached.
That hour is also when the highest-intent traffic arrives and when retargeting budget is doing its most valuable work. If the site slows or falls over then, the loss is not spread evenly across the campaign. It is concentrated at the exact point the campaign was designed to produce.
Roughly seventy percent of entries on a typical UK competition site come from a phone, so the relevant test is mobile checkout under load rather than desktop homepage speed. Shared hosting does not survive this. VPS hosting or cloud with server-side caching and a CDN, around £99 a month, is the baseline.
Load-test the checkout path before the first campaign, not after the first draw.
Checkout Mechanics Decide the ROAS
Everything upstream is persuasion. Checkout is where the money either arrives or does not, and on mobile the failure modes are mechanical rather than psychological.
Apple Pay and Google Pay as the primary path, presented before any card form. One-tap removes the largest drop-off point on a phone. Card entry stays available, but it should not be the default visual treatment.
Beyond that: no forced account creation before payment, visible basket state, age confirmation handled inside the flow rather than as an interstitial that reads like a new obstacle, and correct keyboard types on every field. Small mechanical failures compound on mobile in a way they do not on desktop, and in this vertical they compound during the highest-value hour of the campaign.
Tracking, Before You Spend Anything
Restricted categories complicate measurement. Some events are harder to pass through platform APIs, attribution windows behave differently, and you cannot always rely on the standard ecommerce event set.
Get GA4 and the platform pixels configured and verified before launch, with server-side events where the client’s setup supports it. Confirm that a purchase event actually fires on a real transaction rather than assuming the plugin handles it. Diagnosing broken attribution three weeks into a campaign, with spend already committed, is a bad way to spend a Tuesday.
What this Means for Scoping
If you are running paid media in a restricted vertical and someone else built the site, audit it before you turn campaigns on. Check the compliance elements, the mobile checkout, the hosting and the tracking. Any one of them can invalidate the media plan.
If the client is still choosing a build partner, the specialist route usually costs less overall. Studios working exclusively in this category exist because the problems repeat: raffle web design from a specialist such as Nera Marketing ships the skill question, free entry route, draw mechanics and mobile checkout as standard components rather than as bespoke problem-solving on every project, with builds quoted from £2,995. A generalist build in this vertical usually gets rebuilt, and the rebuild costs more than the original saving.
The wider point is one this agency has made before in different words. Media buying does not sit above the site, it sits on top of it. In restricted categories, that dependency is not a matter of degree. Get the build wrong and there is no campaign to optimise.
