THISTLE

Our goal was simple – maintain efficiency, but double scale meaning ad spend and growth. The first ad-buying strategy we tried was a day trading strategy. We essentially launched 20 – 30 ad sets each morning with a few really aggressive rules. If we didn’t receive the desired purchase rate or efficiency rate by a certain time of the day we’d cut the ad set. We turned on and off ad sets throughout the day until we were left with 6 – 8 ultra high performing, efficient ad sets. The next day we restarted the process again. This worked for a little bit and helped us scale 1.2x or so, but it was extremely volatile.

Next, we switched to a low bids, accelerated delivery strategy. We actively and regularly bid on audiences throughout the day in an attempt to buy the audiences at the price we wanted. The issue with this strategy was again volatility – on some days we were able to spend our budget, on others we didn’t get close.

We ended up with a cost cap strategy that worked really well. This strategy with new creatives, including videos, carousels, and still images that focused specifically on the menu options instead of the company as a whole worked extremely well. Because we had a ton of data to work, with we also granularly segmented the look-a-like audiences. Non-conventional audiences included highest revenue, highest retention, meal based (# of meals per week, breakfast, lunch, dinner), users that shared their referral code, local delivery vs shipping, and geographic audience segmentation are a few examples.

The result was maintained efficiency at almost 3x the spend. Growth is steady and consistent and the team couldn’t be happier.