Should we run a sweepstakes or giveaway on Meta (free entry by email or SMS, extra entries for purchases), and how should we time it, structure entries and measure its lift?
Our answer
A sweepstakes can pay as a short full-price push in a slow stretch or before your biggest sale. Use free email or SMS entry, purchase entries only if free entry gets equal odds, official rules and a regional holdout.
- Confidence
- Low
- Strongest proof
- 5 Meta rules
- 1 dataset, 11 practitioners, 2 opinions or claims
- Sources
- 12 sources
- 19 statements
- CMRHDN
M
New here? Take the 1-minute tour
Why this is the answer
3 reasons and our judgmentSweepstakes can fill a slow stretch. Ridge runs a six-week sweepstakes, mostly at full price, in its late July and August lull, with free email entry and extra purchase entries. Partnership ad click-through rose up to 5 times, and Ridge says entrants buy for months.
Sweepstakes can also build a list. HexClad starts in the middle of September, before Black Friday, giving 25 entries each for email and SMS against 1,400 for a $700 knife set. Its 2024 entrants took 86 days on average to first purchase.
The law requires a free way to enter. Florida bars requiring a purchase, and free entry must give the same odds as buying (Thompson Coburn, Sweeppea). Florida and New York require a filing and a bond or trust account for prizes over $5,000.
Our judgment
- Time it to a slow stretch or the weeks before your biggest sale. Ridge fills late July and August, while HexClad starts in the middle of September and Pilothouse ends one before Black Friday. We judge both timings on what entrants buy later.
- Weight entries toward purchases, but check the free route first. HexClad does this to keep the list qualified. Have a lawyer confirm equal odds before you copy it, since neither brand describes how its free route works.
- Read lift from several rough comparisons, not blended revenue alone. HexClad's blended growth rate barely moved in its first sweepstakes. Ridge reads lift from growth going into the campaign, a shifted week, held-out markets or regions and the weeks after.
- Pick a prize people will talk about and filter list quality through entry weighting. Ridge picks prizes for attention. Jon Loomer's warning about unrelated prizes comes from lead generation and offers no data.
Our judgment, built from Ridge, HexClad, DTC Newsletter / DTC Podcast (Pilothouse), Thompson Coburn LLP, Sweeppea and Jon Loomer: our reading of these sources, not a finding from any one of them.
What the sources say
19 statements from 12 sourcesMeta says a business running a contest or sweepstakes on Facebook is responsible for its legality and official rules, and may not require entrants to share or tag friends to enter.
“We will not assist you in the administration of your promotion”
Florida law bars requiring a purchase to enter a game promotion. With prizes over $5,000, the operator must file the rules 7 days ahead and post a bond or trust account for the prize value.
“To require an entry fee, payment, or proof of purchase as a condition of entering a game promotion”
New York law says a sweepstakes with prizes over $5,000 must be filed with the secretary of state at least 30 days ahead, with a $100 fee and a bond or trust account covering the prizes.
“Failure to file such statement shall be a class B misdemeanor.”
Ridge used markets without a sweepstakes as a rough holdout: sweeps markets usually grow 30 to 40% year over year, and Canada and the UK kept growing faster for weeks after. It calls this unscientific.
“these are all again, like not particularly scientific ways of looking at it”
Ridge's Connor MacDonald suggests launching a sweepstakes a week earlier than last year so the prior year's same week acts as a holdout; going from 10% to 50% growth would show the sweepstakes adds revenue.
“if we're able to go from 10% growth to 50% growth, I think it's pretty clear that the existence of sweepstakes is driving incremental revenue.”
Ridge's Connor MacDonald says sweepstakes-specific ads grow to 40 to 60% of total ad spend during a campaign and often perform better than baseline ads.
“we're able to drive incremental ad dollars that perform often better than our baseline stuff.”
HexClad says its 2024 sweepstakes opt-ins took 86 days on average to first purchase, and it ties extra entries to purchase size to deter unserious entrants. A Ridge guest calls such leads very unqualified.
“The value of those emails is obviously questionable. It's a very unqualified lead that you're gathering at that point.”
HexClad says sweepstakes send email opt-in rates through the roof, but blended Shopify growth barely moved after its first one, so you have to dig into channel results to see the bump.
“we see our opt in rates go through the roof during sweepstakes”
Ridge's CMO says its six-week sweepstakes, run in a slow season at full price, drove about 50% revenue growth year over year at a better MER. He suggests holding out ads by region to read the lift.
“Comparing this year's campaign with last year's tells us whether it grew. It doesn't tell us how much revenue we would have generated without it.”
Ridge says sweepstakes leads are less qualified, but list growth and visibility pay off for months; it tags entrants and tracks what they buy. Lift is strongest for lower-priced items. No purchase rates are given.
“Tag everyone who enters and track what they buy over the next few months. Many free entries convert later, often during your next big sale.”
Ridge's CMO says after sweepstakes in Canada and the UK, those markets kept growing faster than others. He calls these rough reads, not clean holdouts, and suggests geo holdouts for brands selling in one market.
“None of these are clean randomized holdouts.”
Ridge's CMO says its sweepstakes run mostly at full price, with the prize doing a discount's work. Partnership ads raised click-through up to 5x in some cases, but the leads were less qualified.
“The prize does the work a discount usually does”
A Pilothouse operator on The DTC Podcast says giveaways are a good way to gather first-party data before Black Friday.
“Giveaways are a really good one and all it really requires is figuring out, what are we giving away?”
Jacob of Pilothouse, an agency, suggests a giveaway ending before Black Friday, with a squeeze page and lead gen ads to collect emails, saying non-winners often buy anyway. No results or costs given.
“Everyone that signed up that didn't win. There's a bit of FOMO there and a lot of the time they're going to come purchase anyways.”
Law firm Thompson Coburn says a sweepstakes' free entry method must be free, give equal odds and be as easy as paid entry, or the promotion may be treated as illegal gambling.
“A valid AMOE must be free, offer equal odds of winning, and be easily accessible and trackable.”
Woobox summarizes Instagram's promotion rules: official rules, a full release of Instagram by each entrant, a statement that Instagram doesn't sponsor it, and no inaccurate tagging.
“in no way sponsored, endorsed or administered by, or associated with, Instagram”
Jon Loomer says the lead magnet shapes lead quality: a giveaway unrelated to your business, like an iPad, brings worthless leads, while a free first lesson or chapter tied to the upsell works better.
“the lead magnet you offer will almost always impact the quality of your leads.”
Sweeppea, which sells sweepstakes software, says no purchase may be required to enter, any purchase entry needs a free route with equal odds, and New York and Florida often require registration over $5,000.
“No purchase or fee may be required to enter the sweepstakes.”
SweepWidget, a giveaway platform, reports that across 70,000+ contests, 8 to 14 day giveaways drew 956 entrants on average versus 525 for 4 to 7 days. Contests with 12 entry methods drew 1,926. No sales data.
“Our data from 70,000+ giveaways shows a clear correlation: more entry methods means more entrants.”
In the full answer
For readers who sign in- What to do this week3 steps
- The numbers behind it3 figures from cited sources
- Where each source standsA chart placing each source on the question
- The strongest disagreementThe best case against the answer, from a cited source
- How sure we areWhat its confidence rests on, and what would change the answer