You may see an option to add "Event Details" when creating your ad. Not to be confused with Facebook Events, this allows people to get notified of an event, like a time-sensitive sale. This may just be a repackaging of Get Reminders. I'm not sure if this is a test…
Should you separate countries by ad set? There is a potential advantage to doing this, but only in certain situations.
Are daily budgets going away? That's the claim from a recent marketing newsletter, though it didn't cite sources. I didn't name the newsletter here because this isn't about dragging them but about responding to the rumors that have come from it. My guess is that they misunderstand Daily Budget Flexibility,…
If you feel that Ads Manager is overcounting or undercounting conversions, or you want a more precise measurement of results, here's an example of what I do.
Don't discount the potential impact of randomness on your results, especially when the difference isn't statistically significant. If you find yourself micotesting every minor difference and making big changes on the smallest of differences, give this a thought.
I ran a test. I created a campaign with three ad sets, each optimized for conversions with a 1-day click attribution setting. The purpose of the test was to find how performance differs when using three different targeting approaches. Learn more about my test at https://jonloomer.com/split
I understand advertisers' frustration when creating multiple ads with different formats. The algorithm almost always chooses the static image version when optimizing for conversions. Here's a way to get the best of both worlds: A single ad that will show both a static image and video version, depending on the…
My bot's been updated! I've had an AI-powered chatbot that was trained by more than 500 pages of my website for the past year, but now it's received an upgrade. Thanks to a new integration with Messenger, you can access it at https://fb.com/jldchatbot from its Facebook page. You can otherwise…
The most overlooked mistake that advertisers make is a failure to properly interpret results. They take them at face value and refuse to dig deeper. Here's what I mean...
What do you miss the most about old school Facebook advertising? Power Editor, insanely low costs, and the new features that were steadily coming out. Especially FBX and website custom audiences.
Are some new placements coming? If you filter by placement in the custom ad reports, you'll get a long list of placements, many of which aren't currently active. Some are old and others are either in testing or on their way. A few that are most interesting are Threads Feed…
One of the primary reasons for ad rejections is the misuse of personal attributes. You cannot assert or imply that you know the viewer of an ad's personal attributes (age, gender, health, vulnerable financial status, etc.). The issue isn't having an ad or business focused on these attributes. Instead, you…
Advertising is simpler than it used to be, but not easier. I often talk about not overcomplicating your advertising. A simpler approach is better. But, that doesn't mean that it's easier to get great results. The past few years have been much more difficult than ever before. That said, complications…
Much of the time you spend worrying about targeting is probably unnecessary. Many times, it just doesn't matter which inputs you provide -- you'll reach many of the same people either way. Once you embrace the fact that your targeting inputs don't do much (and may even hurt you), you…
Here are 10 ways that advertisers overcomplicate things and hurt their results. In most cases, simpler is better. Adding complexity rarely helps. While there are exceptions, you should test to be sure that the complexity that you add is actually worthwhile.
The approach to targeting has changed. If you're wondering which inputs to use to reach your ideal audience, you're overthinking it. There was a need to think this way in the past, but now you might just drive up costs. Here's why...
Do you still use remarketing strategies? Especially if you create separate remarketing and prospecting ad sets, I encourage you to do this. Test it against using one ad set with Advantage+ Audience. Remarketing will happen naturally without needing to create a separate ad set. For more on testing targeting strategies,…
Do you restrict your targeting by age and gender? Some advertisers do this based on the demographics of the typical customer. While I understand, it's possible that this is completely unnecessary -- especially when optimizing for purchases. Create an A/B test compared to Advantage+ Audience without these restrictions. For more…
Do you still use interests and lookalikes? If you do, you're not alone, but I want you to do this. Run an A/B test that compares your results using interests or lookalikes vs. Advantage+ Audience without suggestions. Maybe the results will confirm your assumptions. Maybe not. But you should test…
What is Post-Conversion Optimization? Not many advertisers have it, but PCO could be enormously useful for those wanting to optimize for events that happen further down the funnel. Here's how it works...
You'll find these settings at the top of your Account Overview page. Many of these settings, like managing Auction Overlap and Audience Fragmentation, were previously found within Automated Rules templates. Those templates are no longer available, and these do essentially the same thing.
There are new audience controls in your Advertising Settings. Brand Protection and Employee Controls allow you more targeting control account-wide but with some exceptions. These could be helpful, but they feel half-backed right now. Here's how they work...
When should we use CBO (Campaign Budget Optimization or Advantage Campaign Budget)? CBO is mainly dead now because the primary use in the past involved multiple ad sets by audience. That's no longer relevant due to audience expansion. But, you may instead use it to promote multiple products.
Sometimes your ads need a reset. Don't fear changes that will restart the learning phase, especially if you aren't getting good results. The restarting of the learning phase could act as a reboot. You might actually get better results that way.