You Ask, I Answer: Determining Facebook Ads Effectiveness?

In today's episode, Jen asks, how can brands find out which kind of Facebook ads work best for them? So this is an important question for not just Facebook ads, but any kind of ad. The advantage of Facebook is that it's a closed ecosystem so that you can do uh apples to apples comparison of the different types of content. The starting point for this is content analysis is understanding what resonates with audiences. So you're going to need to uh brush up on uh your Facebook data analysis skills and getting data out of Facebook so that you can analyze it.

There's four different buckets of content that you need to understand. The first is audience content themselves. If you have a list of people that are customers at Facebook uh that are customers of your company on Facebook, looking at at their stuff, what do they share? What do they post, what do they talk about? And this can be qualitative.

Uh this is not have this step does not necessarily have to be strictly quantitative. It's just trying to get an understanding of what stuff are do people find compelling? What stuff do they post on Facebook? Um what are the things that they share and then reshare and comment on? Getting that sense of who the audience is uh is really important.

And you can also get some of this information, not all of it, but some of it from Facebook audience insights. If you have a uh a page and that page has at least a thousand likes, then you can start to really dig into the data and understand what are the the aspects of the audience that uh are important. What other pages do they like? Um what uh whether demographics, things like that. If you have access to um Facebook data tools like uh Facebook audience insights, uh Facebook Crowd Tangle, etc.

You can pull a lot of this data and start doing you know very large scale topic modeling to understand, for example, uh if your audience in Facebook Audience Insights uh if your audience is there, um you can look at the top 50 pages that that audience also likes and exam again examine their content. Second step is you have to understand your own content. So again, this is this is uh going into Facebook Analytics and doing an analysis of the stuff that you've posted, what content has gotten an engagement, what content has got has performed well. Um is critical for your ad content because you what you don't want to do is just kind of create ads randomly. You want to create ads on things, topics, concepts, ideas that have already done well for you for your Facebook page.

If you haven't done that, and like if you you have no content that resonates well, well, Facebook might not be the place for you then. Um if you're posting and posting and posting and just getting nothing, um maybe do a a week of boosting posts just to see if anything uh resonates with your audience. You might have to pay to play just to even be seen, um, and then make some judgments based on that. The third uh uh data segment you need is competitive content. So what are your competitors sharing?

What are your competitors, what ads are your competitors running, and what kind of engagement do they get on them? This would be direct competitors, uh, and again, you want to see what's popular. The fourth bucket is what I would call landscape content. Now, this is looking at companies that are functionally similar to yours, uh, but are not competitive. So let's say um you own a Chinese food restaurant that delivers, you m uh you might look at um other Chinese food restaurants, you might look at other pizza restaurants, uh restaurants that deliver in general.

You might look at Instacart, um, you might look at the coffee shop, uh, you might look at Whole Foods. Uh basically for people who get consumable items delivered to their house, what content is engaging with them? Are there things that are relevant to the product or service that is being shared in this landscape of behaviors? If you're a B2B company and you sell um SaaS-based accounting software, uh, what do other SaaS companies not in accounting software doing? What is working for them?

Um what's uh like SaaS-based um email marketing software, marketing automation software. Again, we're not necessarily looking for the exact uh topic for those non-competitive but similarly structured companies. What we're looking for is audience behavior of a group of people who are receptive to the type of service that you offer, and then trying to figure out what resonates with those people. Once you've done all four analyses, um you have a you should have an idea of the different concepts, uh, the different media types, the different topics that your audience is interested in. I would also suggest using a social listening tool of some kind.

We used uh Talk Walker for all of our stuff. And again, getting a sense of those topics and keywords and and related conversational things uh in order to uh uh understand your audience better. Once you've got all that down, then you can start creating ads. You should be able to see if in your Facebook data that videos about I don't know, cats, uh, always seem to do well, or videos, uh behind the scenes videos always seem to do well, even if they're not yours, even if they're competitive, that should give you a sense of okay, if we're gonna run ads on a video, the video that we run ads on or the video that we put in our ads should be um a behind the scenes, or maybe even a behind-the-scenes video with cats in it, right? Um if content that talks about uh how your product works has resonated well with your audience, try running some ads on that as well.

The key is to try and align your ads as best as possible with what the audience already likes. Because with Facebook, you don't get many shots at engagement, right? The algorithm works in such a way that content that gets engaged with does better, and content that doesn't get engaged with get does worse. And it becomes very much kind of this uh virtuous or vicious circle of feedback where things simply you know you get less and less engagement, which means you get shown less, which means you get even less engagement. And obviously the only way out of the the vicious cycle is to pay your way back to getting some eyeballs.

That's why the competitive analysis and the landscape analysis is so important because your own data may say, hey, nothing's working. And if that's the case, then you need to look at other data sources to calibrate your ads on. So that's how you can find out what kinds of ads uh work best for you. I would strongly encourage, as you do that, um, to test, to test a lot of different stuff. Um, if you have the budget to do so, uh, test a lot of stuff to see what different types of content are.

So when you do this analysis, if you have four or five or six major topics, if you got the budget, run you know, two to three ads in each topic and see again which one resonates the best, uh, which one gets people to engage even with the ad itself. If after all this, you're still not getting ad performance, it's probably because Facebook isn't the place for your for your audience. Even if they're there, they may not want to engage with your kind of business on Facebook because that's not why they're on Facebook. They're on Facebook to stay in touch with friends, uh, to complain about politics, you know, all the things that we usually use Facebook for. Um and so doing business with you may not that may not be the place.

You may be better off on a LinkedIn or a Twitter uh or YouTube, uh, or even an email newsletter. Right. So do the analysis, but exp understand that Facebook may not be the place for you. The easiest way to make that determination is check how much unpaid traffic you're already getting from Facebook. If it's zero or very close to zero, or you know, less than one percent of your site's traffic, Facebook might not be the place.

Run some tests to verify it, run some ads to test that assumption, and if the ads don't perform, you know it's not it's not the place for your audience. You have follow up questions, leave them in the comments box below. Subscribe to the YouTube channel and the newsletter. I'll talk to you soon. Take care.

Want help solving your company's data, analytics, and digital marketing problems? Visit TrustInsights.ai today and let us know how we can help you.