METATESTBLOG

Meta Bid Strategy Test: Cost Per Result vs. Highest Volume

Every time you run an ad campaign on Facebook or Instagram, you’re entering an auction and competing on getting your ad on a user’s screen. How much you pay is simple supply & demand (along with other algorithm factors) based on how many advertisers are also trying to get in front of a 29 year old female based in Sydney who has an avid interest in fashion, puppies and basketball (for example).

One of the first steps in setting up your campaign, is to select a bid strategy.

Earlier this year, Meta decided to re-work and re-name their bid strategies for ad campaigns. Whilst there has always been the ability to manually select how FB/IG ads buys your inventory, this change simplified the categories they fall under – lowest cost, or controlled cost.

In May, we tested cost per result bidding versus highest volume of results bidding. The test was conducted across 8 clients in different industries. All campaigns were targeting a result falling under lead generation or ecommerce.

It goes without saying each industry and business has its unique challenges and campaigns within it will behave differently. Data has been given parity by applying interpolation and extrapolation to base volume metrics on an even $1,000 of spend. Whilst we have found some macro trends, please consider everything from your goal to your creative when considering what objective to use!

Some context…

Every time someone opens Facebook, Instagram or any other owned and targeted placements, the ad inventory is an open auction for the advertisers trying to get their ad in front of that individual (based on many possible targeting mechanisms).

Cost per result (or cost capping) gives advertisers control on how much they’re willing to spend to achieve a result, over time, that all begins with the auction price for the impression. Whilst it can be quite volatile especially in a campaigns infancy, it should, theoretically balance out over time.

In Meta’s words:

“Cost cap is one of Facebook’s bid strategy options, meaning that it tells us how to bid in the ad auction. The cost amount you provide is an average amount that we try to deliver against over the campaign lifetime, while dynamically bidding as high as needed to maximise results. This means that the average cost per result may exceed the set amount.”

Highest volume bidding aims to buy as much inventory as possible and spend the full budget. It doesn’t worry so much about if your results are efficient.

In Meta’s words:

“Lowest cost is one of Facebook’s bid strategy options, meaning it tells us how to bid in the ad auction. When you use the lowest cost bid strategy, we’ll aim to get the most results possible from your budget.

For example, an event planner could use the lowest cost bid strategy to get as many people as possible to attend an upcoming music festival, where cost per attendance doesn’t matter.”

WHAT WE FOUND

  • Variance to CPR goal
    • Half of CPR campaigns ended up with a CPR that was higher than the target. Of the campaigns that exceeded their CPR target, they overshot it by an average of 120%.The campaigns that beat their target, they outperformed it by an average of 43%. Interestingly, the only campaigns that beat their target were eCommerce brands with a goal of purchases. Lead generation saw a majority of CPR actuals come in more expensive than target. 
  • CPR struggles to spend early. This resulted in proactive optimisation to continually increase the CPR until just before the efficiency became unprofitable.
  • Highest volume will get your ads in front of more people, more times. It could be argued that some of this inventory is low quality given the framework of highest volume. If your creative is quite informative and potentially front loads information for the user, consider this approach.
    • There were marginal differences in efficiency metrics of CPLC and ULCTR. This could indicate several things, but one that is likely prevalent is asking if that of the additional reach and impressions that highest volume can achieve, are these relevant people even remotely at the top of your funnel?
  • CPR should be used when you have time. If you’re only running a campaign for a few weeks, the time and budget it needs to learn could result in a serious underspend and not enough time for the bid strategy to truly learn the most efficient way to spend budget to achieve your desired result. If you can’t commit to at least $5k to test, you’re likely going to cut it at the knees before it has a chance to prove itself.
  • Run your own test! There are alot of factors at play here and even though parity was applied where it could be, this test still involves individual industries and campaign information. Whilst we’ve shared individual insights and learnings on a 1:1 basis with the clients involved, we’ll keep that private!
  • Want us to run an individual test with your campaigns? Get in touch!

Hello. Get in Touch.