Menu

5 Steps to an Ad Campaign Optimization Audit

While our accounts here at Nanigans differ in terms of vertical, spend and frequency, a unifying factor between our gaming, eCommerce, travel and pure play clients is their drive to scale and their willingness to invest budget into high-performing audiences. In an ideal world, performance marketers would be able to run a postmortem on every campaign to analyze what didn’t meet expectations in terms of bids, and what audience targeting should be duplicated in the next push, but it’s easy to get in the weeds when you’re spending thousands a day. To avoid campaign sprawl, we advise an optimization audit. Taking time out to switch off unsuccessful ads and focus on the successful ones is an easy way to identify winning patterns in creative, bids and targeting.

Without further ado, here are five steps you can take to look beyond the ad level and reconnect with what works:

1. Determine that the timing is right

Before you take any action, make sure you have enough data to make an honest judgment (ads should have at least 100 clicks), that the ad campaigns you’re assessing are evergreen (if things are running well on a sale, you don’t want to mess with the momentum), and that you’re not pushing for volume.

2. Get a fresh set of eyes

With the ease of campaign cloning comes the challenge of making the necessary adjustments. We recommend that you review the way your strategy groups are set up. Look at the campaign with these two questions in mind: (a) How are we bidding? and (b) How are we allocating our budget? Make sure these goals align. For example: If your campaign is set up for maximum volume but you are bidding for cheap cost per action, you will want to change that one setting.

3. Pause down poor performing ads

Prime candidates for shutdown are ads with an egregious amount of spend (i.e. $50) without a conversion or really high cost per action (CPA). If your CPA is over 25% higher than your goal, that placement is most likely hindering your performance.

4. Focus on your top 10

The goal is to cut through the noise (running hundreds of ads with varying degrees of success for any given strategy group or brand) to running a small number of the best or most efficient placements. It’s important to keep in mind that the ratio of ads should be relative to the amount of target spend. In the case of one of our longstanding eCommerce clients, we brought their brands that had 100, 200, or 300 ads running down to 20 or 30 max. In the two weeks that followed, their CPAs were at, or right around, what they wanted. And after running an optimization audit for one of our newest eCommerce clients, their yield revenue went from 30% to 90% and has stabilized to 80% over several days.

5. Run a controlled experiment

  • Look at your audience segments (country, gender, age, targeting)
  • In each, just run the top one or two ads in terms of spend/performance (pause remaining)
  • For top-selling products, try new creative
  • For top-performing audiences, target new lookalikes
  • Unpause ads that saw no spend in the hopes that they’ll get traction
  • If you have a clear creative winner, push it in any instance where you have a new audience

Why is an optimization audit helpful?

Trimming down your campaigns to identify best performers allows our algorithm to hone in on what’s performing best and iterate. Pushing top creative to new audiences is also a best practice.

An optimization audit is most necessary when there is a marketing personnel change, but if you are taking on an important new campaign, ending an ad and redeploying it, or if you are cloning a campaign and putting it in a new budget pool, it’s also an excellent idea.

If you have thousands of ads live, focusing on the few ads that perform extremely well will help re-align your strategy for future campaigns.

Leave a reply

Your email address will not be published. Required fields are marked *

* Copy This Password *

* Type Or Paste Password Here *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>