Nothing runs until Meta says it can
You cannot advertise a UK prize competition on Meta by simply writing a good advert. Prize competition advertisers are reviewed under Meta’s real money gaming requirements, and until that review passes, the account is not a marketing problem. It is a permissions problem.
We have been through this process repeatedly and it behaves the same way every time. It is a manual review. It takes weeks rather than days. The process tightened during 2026. Meta routinely asks for additional documents partway through, which operators read as a rejection when it is standard. And chasing it or resubmitting makes it slower, not faster. Meta’s own signalling does not help: we have seen an approval email arrive while the status page still said in review.
The part operators consistently underestimate is that Meta is not really assessing the advert. It is assessing the business behind it. That means the platform, the entry mechanics and the paperwork all have to agree with each other before you apply. If they do not, no amount of creative testing rescues the account. We wrote up why Meta bans competition ad accounts because it is the single most expensive mistake in this category, and it is almost always avoidable.
The legal opinion letter is the document that unlocks everything
The evidence Meta wants, and the evidence a specialist payment processor wants, largely overlaps. At the centre of both sits an independent legal opinion letter.
A legal opinion letter is a solicitor’s or licensing consultant’s written assessment of whether your competition is lawful as structured, referenced against the Gambling Act 2005. It is not marketing collateral and it is not something you can write yourself. Its whole value is that it comes from someone independent and qualified.
What it does is answer the only question that matters: is this a prize competition or is it a lottery? A lottery needs a Gambling Commission operating licence. A prize competition with a genuine skill element and a real free entry route does not. That distinction is the legal foundation the entire UK competition sector stands on, and the letter is what turns your version of it from an assertion into an assessed position. We have set out how a legal opinion letter works and what it covers in full, and separately whether you need a gambling licence at all.
Practical notes from arranging these repeatedly. Lead time is roughly one to two weeks, so it belongs at the start of a project rather than at the point you need it. It is prepared independently, and neither we nor the solicitor make Meta’s decision or the processor’s decision. And it is worth commissioning even when nobody has asked for it yet, because the two parties who will ask are the two you cannot launch without.
PrizeRunner’s own terms reference licensing and set out a free postal entry route across ten separate provisions, and prize claims on the platform are gated behind a skill question before a prize is released. That structure is what makes the platform advertisable. Our compliance service exists to get operators to that position, and the underlying mechanics are covered in our guides to skill questions and skill-based competitions versus prize draws.
We check payments before we spend anyone’s budget
This is not a payments project. We did not build PrizeRunner’s checkout. But we will not run acquisition into a payment setup we have not looked at, and the reason is straightforward economics.
Paid traffic is the most expensive traffic an operator buys. Point it at a checkout that fails under load and you have paid for the click and lost the sale. Point it at a processor that freezes the account mid-campaign and you have paid to acquire customers who cannot pay you. Both happen in this sector, and neither is a hypothetical.
The underlying problem is that competition operators are treated as high risk by default. Stripe, PayPal and Shopify Payments all refuse competition and prize draw businesses. Operators usually discover this after they have paid for a website, which is the worst possible order. We wrote why payment providers reject competition websites and a working comparison of UK banks and high risk merchant providers for competition sites because it is the first wall every operator hits.
On our own builds we solve it with Cashflows, a specialist acquirer that underwrites competition operators in the UK and Ireland. That is not a plugin. It is a merchant application, underwritten on the strength of the business case put in front of it, and it needs an accurate account of how entries are sold, what the skill question does, how the free route works, what appears on the customer’s bank statement as a descriptor, and what happens operationally when a draw closes. We prepare that evidence alongside the build so the journey the acquirer reviews matches the one described in the application. When those two disagree, underwriting stalls.
There is a load dimension too, and PrizeRunner has it in an extreme form. Instant win entries at 12 pence generate more orders per minute than any other competition mechanic. In our experience the failure point under that kind of concentrated demand is almost never the platform or the page loads. It is the place-order step, and the usual culprit is marketing and email tooling firing tracking calls synchronously before the order completes, so every customer waits on those tools before seeing a confirmation. That is a solvable engineering problem, and it is why we treat instant win mechanics and performance hosting as acquisition infrastructure rather than technical housekeeping.
One more thing worth stating because operators are rarely told it up front: VAT on paid prize draw entries is a live issue, and a February 2026 parliamentary answer put paid-entry prize draws on record as standard rated at 20 percent. That changes the arithmetic on what a cost per purchase can afford to be. We are a competition website agency, not tax advisors, and operators should take their own advice on it.
The problem was breadth, not budget
With the gate cleared, the actual advertising problem could start.
PrizeRunner gives a player several different reasons to be there. A £1 cash draw with only 100 tickets. Instant win games at 12 pence an entry, with names like Coin Fallout, Roulette Riches and Pharaoh’s Fortune. Ten collections running as a series, where completing all one hundred cards plays for £100,000. A membership tier on top.
That breadth is commercially strong and it is an advertising problem. One advert cannot explain a platform like that. Try, and you get an advert that describes a company rather than an offer, which is the most common way competition operators waste money on Meta.
Someone who has just seen a Coin Fallout clip needs a route straight into Coin Fallout. Someone meeting the brand for the first time needs an introduction, not a 12 pence instant win. Someone who has already bought needs a reason to come back on a day when nothing is closing. Three jobs. They cannot share a campaign.
One campaign, one job, one number
We split the account into four layers, each optimised for the one thing it was capable of delivering.
Awareness introduced the brand at scale. The Gem Drops campaign reached 236,562 people and served 734,404 impressions. Its job was recognition, so it was not asked to produce purchases and was not judged on them. Operators routinely kill awareness campaigns for failing at a job they were never given.
Traffic bought cheap product visits, 2,527 of them across the period. On a platform with this many entry points, cheap qualified visits are how you find out which products people actually want before you spend prospecting money finding out the expensive way.
Prospecting sales carried the commercial weight and produced 101 of the 102 purchases.
Retargeting ran as its own campaign so warm traffic had a dedicated route back rather than competing for delivery inside a prospecting campaign.
The full mechanics of how we structure and measure this are in our guide to Meta ads for UK competition websites.
The optimisation goal matters more than the budget
One decision worth spelling out, because it is where most competition accounts stall.
On a thin audience, optimising for Conversions does not work. Meta needs a volume of conversion events to leave the learning phase, and a young retargeting pool on a competition site cannot supply them. The campaign never stabilises, delivery stays erratic, and the operator concludes retargeting does not work when what actually happened is the ad set never left learning.
Where the audience is too small to feed a conversion objective, we optimise for landing page views instead and let purchase volume build underneath. It is a slower read and a more honest one.
Creative showed the game, not the company
PrizeRunner’s strongest asset is what the product looks like in motion. Treasure Hunt and Coin Fallout each have a recognisable world, a character and an action. Five seconds of either communicates more than a paragraph about competitions.
So the creative shows the game before the click, and the landing page continues the same visual idea after it. When the advert and the page look like the same product, the player does not have to re-orient. That is worth more than most copy changes.
We ran six separate angles rather than one general advert: PrizeRunner Royale, Door Knock, AirPods, Coin Fallout, Jump In and Play, and Collect Them All. Each is a distinct entry point, not a variation on the same pitch.
Organic is not an alternative to any of this. Facebook suppresses organic competition posts, and engagement mechanics of the tag, share, follow and like variety get them removed outright. Paid is the realistic route to reach in this category, which is exactly why the structure has to be right. Where organic does work is live draws and TikTok, which are different jobs again.
Paid, email and SMS as one system
A first purchase on a competition platform is worth very little on its own. The value is in the fifth.
That is especially true here, because the collections mechanic is built on incomplete progress. A player sitting on 6 of 10 cards has a reason to return that no advert needs to manufacture. The platform generates the message. The channels only have to deliver it.
Meta reaches new and returning audiences. Email carries detail: launches, competition mechanics, collection progress, what membership is for. SMS handles timing, which is closing draws and live draw reminders. Used the other way round, both underperform. Email is too slow for a draw closing in an hour, and SMS has no room to explain a collection.
Email only works if it arrives, which is its own discipline on a competition domain. We cover both email marketing for competition websites and email deliverability separately, because a competition operator sending draw reminders to a spam folder has a technical problem, not a marketing one. One recurring false alarm worth knowing: an operator decides sales have gone quiet, then logs in and finds the sales happened and only the notification emails failed.
What we would do differently
Publishing what worked is easy. This is the part worth reading.
The retargeting campaign should not have been a separate campaign yet. It produced a single purchase across the whole period. The audience was too thin to carry its own budget, and we would now fold that spend back into prospecting until the warm pool is large enough to sustain delivery. We reached the same conclusion on our own account, where a third audience layer had to be collapsed into the one above it for exactly this reason. Structure that is correct in principle can still be premature in practice.
Thirty days is not a verdict. A fresh ad account produces volatile early results, and the first weeks are about feeding consistent conversion data rather than judging return. We tell every operator this before we start, because it is the most common reason a competition account gets switched off two weeks before it would have worked. One operator asked whether spending roughly £17 to make £6 on their first purchase was normal. On a cold account, it was. For contrast, on a mature account for Cars and Kettles we have run a blended £2.65 per purchase across 468 purchases. That is what this looks like with data behind it.
Ads Manager should not be the only source of truth. Reported status and reported spend both lag and mislead. Adverts can read active after being switched off, can still appear in the admin’s own feed after ending, and dashboard spend can diverge from what actually leaves the bank once VAT and billing thresholds are involved. We reconcile against statements, not dashboards.
What other competition operators can take from this
If you already have a platform, you do not need a new one to fix acquisition. You do need the foundations underneath it to hold.
What made the difference here was sequence. Compliance and permissions first, because Meta decides whether you get to advertise. Payment resilience next, because paid traffic is the worst traffic to point at a checkout you have not checked. Then structure, and only then creative. Reverse that order and you spend the budget learning what you could have known in week one.
The advertising lesson is narrower and simpler: refuse to let one campaign do four jobs. Awareness was allowed to be awareness. Traffic bought cheap visits to locate demand. Prospecting was the only layer judged on purchases. Retargeting got its own route back, and we learned it got one too early.
Nera Marketing is a signatory to the DCMS Voluntary Code of Good Practice, in effect since May 2026, and we will not run acquisition into a platform that could not survive that review. That is not a marketing position. It is the reason the accounts stay live.
