Blog
Every recommendation in this category starts with data. Gleam analyzes real campaign performance across giveaways, user generated content, lead capture and rewards, then publishes what the numbers actually show.
Each study gives you a benchmark to measure your own campaigns against: entry rates, referral lift, prize performance, differences by industry and the mechanics that move conversion. Use these reports to plan on evidence instead of guesswork.
The typical paid Gleam giveaway with a refer-a-friend action drew 7.5% of its entrants through referrals. Benchmarks from 4,219 campaigns run by 1,028 businesses on entries per referral, widget position, promotion breadth and campaign timing.
Every giveaway starts with one decision: which prize to run and how much to spend on it. This research has been conducted by Gleam to provide giveaway organizers with a reference to plan prizes budget, the prize fit to the audience, the actions used and the plan tiers leading to better outcomers.
December holds 10.4% of giveaway launches. We compared 117,206 campaigns to see how start date, run length, prizes and actions differ in December.
Five findings from Gleam giveaway data, from how your next giveaway compares with your last to how many entry actions to ask for.
Every report here is built on first party campaign data from the Gleam platform, combined with external benchmark studies where wider context helps. We publish the sample size, the filters applied and the method behind each figure, so you can judge how closely a benchmark fits your own market.
Studies run on a quarterly cycle, and each one covers a single theme: competitions, user generated content, lead capture or rewards. Between full reports we release shorter statistics you can cite in your own decks, pitches and growth marketing plans.
A benchmark only pays off when you act on it. Once you know how your entry rate, referral rate or capture rate compares to the field, the next step is closing the gap. Pair these findings with our proven growth tactics and test one mechanic at a time.