Giveaways are easy to call a success. You run a collab with three other brands, the entry form collects 12,000 emails, and the recap slide shows a huge number in green. Two months later, most of those subscribers have never bought, a chunk have unsubscribed, and your email engagement rate has dropped enough that inbox placement suffers.
I am not against giveaways. Some of them bring in customers who stick. But you only find out which ones by measuring something other than entries. Here is how I do it.
Entries are not the result
The goal of a giveaway is customers, not contacts. Entries tell you the prize was attractive. They do not tell you the people who entered want your product. A $500 gift card giveaway will usually get more entries than a giveaway of your own products, and it will almost always bring in worse subscribers, because people entered for the gift card.
Set the success metric before launch. I use first purchases from giveaway entrants within 60 days, the cost per first purchase including the prize and any paid promotion, and the share of entrants still subscribed after 30 days.
Tag every entry at the source
Give each giveaway its own source tag in your email platform, and if you run it with partner brands, tag the entry by which partner sent the person. Collab giveaways are usually lopsided. One partner’s audience converts and another’s does not, and you only see that if you tagged it.
Track the traffic too. Use a unique landing page and UTM parameters for each partner and placement so your analytics and attribution tools can see where entrants came from. If you count sign-ups as leads, report them by channel the same way you report orders. ThoughtMetric, for example, reports leads alongside orders by channel and campaign, so giveaway sign-ups can be compared with the purchases they lead to. (Disclosure: ThoughtMetric sponsors this site.) Any tool that counts a lead event can do a version of this.
Follow the cohort, not the campaign
Most of the value, if there is any, shows up after the giveaway ends. Build a segment of entrants and watch what they do for 60 to 90 days. Look at first purchase rate, first order value, the discount used, and whether they buy a second time.
Compare that cohort with your normal new subscribers from the same period. If giveaway entrants buy at a quarter of the rate of organic sign-ups, you know roughly what each giveaway subscriber is worth relative to your usual list growth. That ratio is the number to bring to the next planning meeting.
Count the costs honestly
The prize is the obvious cost. The less obvious ones are paid promotion, partner fees, the discount code you sent everyone who did not win, and the deliverability hit from a wave of disengaged subscribers. Add them up before you calculate cost per customer.
The discount code deserves its own line. Giveaways usually end with a consolation offer, and that is often where most of the purchases come from. If the code is strong enough, you are running a discount campaign with a giveaway attached, and it should be judged like one. I covered how promotions distort channel numbers in reading ROAS during promotions.
Clean up after
Plan the sunset before you send the first email. Put giveaway entrants in a short welcome series built for them, and suppress anyone who has not opened or clicked within 30 to 45 days. It is better to lose most of the list on purpose than to let it drag down engagement on every campaign you send for the rest of the quarter.
Watch your unsubscribe and spam complaint rates in the first two weeks. A spike there is often the clearest sign that the giveaway audience was not a fit.
When giveaways work
The giveaways that work tend to share a few traits. The prize is your own product or something your customers would buy anyway. The partners sell to the same customer at a similar price point. And the follow-up offers a clear reason to make a first purchase that does not depend on a deep discount.
When those line up, the entrant cohort can look close to normal new subscribers, and the giveaway becomes a reasonable way to buy list growth. When they do not, it is usually cheaper to spend the same budget on prospecting ads and measure the customers directly.
What to report
- First purchases from entrants within 60 days, not total entries.
- Cost per first purchase, including the prize, promotion, partner fees and the consolation discount.
- Results split by partner and placement, using unique landing pages and UTMs.
- Entrant purchase rate compared with your normal new subscribers in the same period.
- Share of entrants still subscribed and engaged after 30 days.
- Unsubscribe and complaint rates in the first two weeks, as an early fit check.
Leave a Reply