Facebook Experiment: A Tool In Testing Facebook Ads

Q: A lot of our audience members are navigating Facebook ads for the first time. Can you please do a top-level walkthrough of each of the elements found in the new Experiments section — What are the tests it replaces or combines? What information do they provide? Why is that useful?

A: THEY DO NOT APPLY TO PAST CAMPAIGNS OR ADS

A/B Tests

  • Testing campaigns manually
  • We could do a/b tests and split tests on ads, but not on the campaign as a whole
  • Can only test MOVING FORWARD
  • Campaign Budget Optimization Tests

  • Also called CBO, this kind of test is trying to help you understand if CBO is helping or hurting your ads. I tend NOT to use CBO for local business clients.
  • It has not historically provided excellent results. However, it’s something to continue testing as CBO may one day be the only option we have, and they continue to change and improve these options that we have access to.
  • Holdout Tests

  • This kind of test is not a great idea for local businesses since transactions happen offline, but if you have a precise conversion that can occur WITHOUT someone seeing your ad, you can test it.
  • Holdout Tests is trying to help you understand if your ad is the reason someone took action.
  • Brand surveys

  • Trying to measure how well people remember your brand after advertising, or what their opinion of the company is after seeing it.
  • Have to have spent over 10K in the last 90 days to be eligible.
  • Very interesting, running one for my agency right now.
  • Q: A quote from Facebook states, “When measuring the effectiveness of a campaign, carefully consider the implications of using those results to inform future decisions, while A/B testing can provide relevant short-term insights.” Can you clarify how we should use the data from Experiments?

    A: Use experiments as a way to compare data to something else. From experience, the split testing feature, for example, was never able to achieve the same cost per result after the test completed when running the winning ad. It always went up. Always. They’re saying not to take the results as fact and to be cautious about using that as your only basis of decision making for your future advertising decisions.

    Q: Facebook also notes, “Changes in consumer behavior will impact how people interact with advertising, so it’s important for advertisers to consider these influences and adjust their plans accordingly.” What other influences should marketers be examining here?

    A: ECONOMIC #1 Look at what we’re dealing with. If a business ran a test 1 week before this craziness started, they would likely NOT get the same result today, because the economic landscape has changed completely.

    Election years also impact advertising as there is a lot more noise out there, and politicians buy up lots of ad space. Holidays impact performance, times of year (seasonal businesses), new technologies are becoming available, etc. You have to look at the economy, your customer, and your historical business data in conjunction with your experiment results to have a full picture.

    Q: Is there any other outside data we should be pulling from, or are there any specific tools you recommend for this?

    A: Look at your CRM and tracking. Local businesses need not just look at leads but calls, appointments, conversion rates, purchase value, etc. Sometimes the lowest “cost per result” does not mean that it’s the best client to buy the most or for the most extended period. Your offline conversions are HUGE, and they matter much.

    Q: Your agency specializes in local businesses. How are you advising your local business clients to proceed with their marketing at this time?

    A: CHANGING THE OFFER IS #1.

    Changing our messaging. Acknowledging the situation and making sure they see how our offer relates to it. Supporting, focusing on “support local,” switching to completely virtual options, using webinars as opposed to live events, helping my clients get set up technologically.

    Q: How are they making Facebook ads work for them?

    A: We’ve actually been seeing pretty great success across the board. Even our orthodontic client is doing pretty well given the circumstances, and we’ve been able to get him virtual appointments and in-office consultations as they are still able to do that, just with extra precautions.

    The offer has been vital, not pulling back on ads, being creative, and is committed to staying in business and doing what needs to be done to make it happen. Making changes that maybe are not necessary but are essential to the peace of mind of their clients.

    Have questions? Drop them below. Want to talk about working together? Visit alliebloydmedia.com/book to schedule a time to talk!