How we verify giveaways before they go live
Short version: software finds the giveaways once a day, strips out the duplicates, and drops each one into a review queue as an unpublished draft. A person then reads it and decides whether it goes live. Nothing publishes itself.
That is broadly how giveaway aggregators work everywhere. The finding half is much the same wherever you go. The deciding half varies enormously, and it’s the half that determines whether the listing you clicked is real. “Verified giveaways” is a phrase almost every aggregator uses and almost none of them define, so here is ours defined: what our checks catch, and then, just as plainly, what they don’t.
How giveaway aggregators work
Any giveaway finder is two things bolted together: a crawler, and an editorial process. The crawler is the boring half. It visits public giveaway-listing sites and sponsor pages on a schedule, pulls out whatever structured information is sitting on them — title, prize, closing date, entry link, the platform running the widget — and hands a pile of results to whatever comes next.
Everything that distinguishes one aggregator from another happens in that second half. A site can publish the whole pile the moment the crawl finishes, which is cheap and produces what you’d expect: the same sweepstakes listed four times under four titles, links that died a fortnight ago, closing dates copied wrong from a source that copied them wrong, prize photos belonging to somebody else’s giveaway. Or it can put a person in front of the pile first. That is slower, carries fewer listings, and costs an actual salary. We chose slower.
Knowing how giveaway aggregators work is useful mainly for this: you can tell which kind you’re on in about a minute. A site doing the work names the sponsor and the platform, shows the closing date before you click rather than after, and doesn’t make you register to see what the sponsor published for free. A site skipping it has a front page full of campaigns that ended last month, because nothing on it ever re-checks.
How we verify giveaways, step by step
- Daily crawl. Automated scrapers run once a day against several public giveaway-listing sources, plus sponsor pages directly.
- Deduplication. Every listing is reduced to a canonical campaign identifier derived from its entry URL, before anything is imported.
- Enrichment. Prize details, winner counts and entry requirements are read from the sponsor’s own official rules text, where the page carries it.
- Review queue. New listings arrive as drafts. A human reads them and publishes, holds, or rejects.
- Rejection is permanent. Anything we turn down has its URLs suppressed, so a later crawl can’t quietly bring it back.
- Daily expiry sweep. A separate job re-checks end dates and marks finished campaigns as expired.
Why deduplication happens first
The same sweepstakes turns up on three different listing sites under three different titles, three different summaries and three slightly different URLs. Published raw, that’s three pages competing with each other, and a reader who enters all three has wasted two clicks on a campaign they were already in.
So before import, we derive a canonical campaign identifier from the entry URL. Two listings that resolve to the same campaign collapse into one page, no matter which source found them or what each source called it. The identifier comes from the campaign itself rather than the link text, which matters because the same Gleam-hosted campaign can be written several ways and still be one draw with one winner.
Why a person still has to look
A person is in this loop for one reason: automation is good at finding things and bad at judgement. It can’t tell you that a prize description is nonsense, that a listing site has garbled the end date, or that a “giveaway” is really a coupon offer with a spin wheel attached. Every new listing lands as a draft for that reason, and someone reads it before it becomes a page. Listings that fail get rejected rather than deleted, and rejection writes the URL to a suppression list, which is what stops tomorrow’s crawl re-importing the same rubbish.
We also drop non-English giveaways. A campaign whose rules and entry form are in another language is a dead end for almost everyone here.
Where the prize details come from
Prize values, winner counts, entry requirements and closing dates are pulled from the sponsor’s official rules text when that text is on the page. That’s the most authoritative source available to us, and it’s more reliable than a listing site’s summary, which is usually written for clicks.
Not every campaign publishes rules where a crawler can read them. When the detail isn’t there, the field stays empty rather than getting filled with a guess. An empty prize value is honest. An invented one sends somebody chasing a prize that doesn’t exist at that size.
Why some listings have no photo
This one is a deliberate trade-off, and it costs us. We only use an image the source page explicitly attributes to that campaign: the share image the page declares for itself, or an image sitting inside the campaign’s own container. If a page declares nothing, the listing gets a plain branded placeholder.
The obvious alternative is to grab the best-looking image on the page. We tried versions of that and stopped. Giveaway listing pages are wall-to-wall thumbnails of other people’s giveaways, and several platforms shuffle those blocks on every page load, so “the first image that looks right” is frequently a stranger’s prize. A wrong prize photo is worse than no photo. A placeholder tells you we don’t have the picture; a confident photo of the wrong thing tells you something false about what you’re entering to win.
Expiry, and why old giveaways stay up
A second daily job walks the catalogue and re-checks end dates, marking anything past its close as expired.
They don’t get deleted, and this is the number most aggregators would rather not print: roughly three quarters of everything listed here has already closed at any given moment. That sounds like neglect. It’s arithmetic. A campaign runs for a few weeks and then ends permanently, while the archive of campaigns that have ended only ever grows, so the expired share climbs towards a high plateau and stays there. Any listing site with a few months of history behind it sits in the same place, whatever its front page implies; the ones that appear not to are quietly deleting their past.
Deleting ours would break every old link, bookmark, forum post and search result pointing at a page that genuinely existed, and it would take the prize details with it. Those get looked up: people want to know what a sponsor gave away last time before deciding whether this year’s is worth entering. A page saying “this closed on the 14th” is more use than a 404.
The ratio never leaks into what you browse. Expired listings are dropped from the sitemap and kept out of every feed, archive and email that promotes live campaigns, so the front page and the category archives show active campaigns only.
What this means when you’re browsing
Every listing links out to the sponsor’s own campaign page; we don’t host entry forms or stand between you and the sponsor. The end date, prize and entry frequency came from the campaign itself, so you can weigh up whether a daily-entry sweepstakes is worth the habit before clicking through, and browsing by prize category skips everything you don’t care about in one go.
What we can’t promise
Here’s the honest limit of it. We link to campaigns other companies run. We don’t run them, don’t draw the winners, don’t hold the prizes, and have no way to confirm that a sponsor will actually ship one once it’s been won. Our checks catch duplicates, expired campaigns, junk listings and mismatched images. They cannot catch a sponsor who quietly decides not to award anything, and they sit well upstream of the scams that do happen in this category, most of which arrive by email or DM after the draw rather than anywhere near the entry form.
We can also just be wrong. A sponsor extends a deadline and doesn’t update the page. A prize value gets misread. A campaign changes its terms after we listed it. Read the official rules on the sponsor’s page before entering, and treat what we publish as a signpost rather than a contract. If you spot something broken, tell us and we’ll fix it.
What we miss, and how to tell us
The biggest gap is structural. A crawler can only find campaigns that are listed somewhere public. A brand running a quiet sweepstakes on its own site, promoted to its own mailing list and carried by no listing site anywhere, is invisible to us. Those are frequently the ones worth entering, precisely because far fewer people have seen them.
The only thing that has ever fixed this is a person who happened to stumble across one. If that’s you, send us the giveaway — it goes into the same queue as everything the crawlers found, duplicate check included, so you’re not making work for anyone by sending something we already have.
The other gap is timing. A campaign with a short entry window can open and close between two of your visits, and the listing was live the whole time you weren’t looking. The giveaway newsletter exists mainly to plug that: same catalogue, same review, arriving on whichever schedule you pick instead of waiting for you to remember.
