One thing that I thought was interesting for us to explore a little bit was video advertising on Facebook.

Right now, in the last year or so – actually, it’s been more than a year now – video has been the preferred ad type for Facebook. We’re capitalizing on this in a couple of different ways.

Right now, a lot of people are having a hard time developing video content – almost because they’re getting in their own way. This is sort of beneficial for us as an agency because we know that it actually is a very low barrier to entry in order to produce some video advertising pieces of content for Facebook.

A lot of people think that it takes the TV budget – high resources, high budget, really takes a lot of effort and resources in order to develop something that’s appropriate for Facebook – but, in actuality, there’s a lot of different ways that you can produce a video asset without breaking the bank.

We started out over a year ago developing GIFs and this was at a time where Facebook was pushing video for branding and then they started saying that, with a certain sequence, it could be done for direct response.

Now, my agency just does direct response. We only focus on user and customer acquisition for our clients. And so, that’s always been our focus. And so, we went out the door just testing video advertising with, you know, these really low barrier to entry GIFs. We didn’t want to spend a ton of money. We didn’t want our clients to spend a ton of money on these big branded five-minute videos that we didn’t know if they would move the needle at all.

So, we went out with these GIFs and actually killed it. They were absolutely destroying all of our other ad types – Carousel ads, link ads. It was actually pretty amazing. And so, we immediately started testing it across all the different industries that we served and the response was amazing. You know, we started developing these really sort of low-budget GIFs for a bunch of our different clients – just as a way of testing video – and it was crazy.

Fast forward a year plus and there’s the reason that virtually all of our eCommerce clients are running only video ads on Facebook. I’ll say that again – virtually all of our eCommerce clients are running only video ads on Facebook.

We have clients that are spending a million dollars a month on Facebook advertising and video ads is the only thing that they’re pushing. There’s a reason for that. Basically, because it’s Facebook’s preferred ad type right now, they are able to push video where others might be putting out link ads or your quintessential website link ads. They are pushing video ads because that’s what they see as engaging. That’s what they see people clicking on.

And so, there’s a couple of ways that you can test video with Facebook advertising. You don’t have to immediately spend a ton of money on a big budget production video and create essentially a TV commercial for Facebook advertising.

You can dip a toe in the water by testing Facebook’s slideshow video ad feature. Basically, you go in and it’s about as simple as developing a PowerPoint. You go in, you drop a couple of images in, you can add some overlays, you can add some music, and you create a slideshow. You can do that all within Facebook’s ads manager and you can just go out with a video advertising that way.

Before you test anything, before you spend a lot of money on a big budget TV commercial-type video asset, you can test the slideshow feature and see if that moves the needle for some of your video advertising. Then, you can move on to other things like creating a GIF. Any graphic designer or web designer should be able to create a GIF.

We did a thing for one of our clients where they sold a bunch of the same types of product in one color. And so, all we did was spliced the same images over and over again of the same product but we just changed the color over and over again so it looked like a GIF that was just one product staying in the center and it changed from red to green to blue to yellow. Any designer should be able to do that.

We also took existing video assets from some of our clients who did have, say, a Kickstarter video or they had a TV commercial but it wasn’t sort of by itself really appropriate for Facebook. What we could do is then splice it up into 15-second increments, 10-second increments, and test those as video ads. Just overlay a logo or overlay a call to action and see if that moves the needle at all. You can repurpose any existing video assets you have.

I do have one story about a client of ours that did a longform branding video for themselves. They didn’t have direct response in mind. They weren’t creating this video in order to drive sales of their product. Instead, they wanted to tell their story. This client has very similar to, like, Tom’s or Warby Parker – like, a buy one, donate one sort of thing going on – and so what they wanted to do was tell that story and really hit home what the mission is of the company. They built the company around this mission.

And so, they told the story in a 3.5-minute video. When they posted it on Facebook, they got a lot of organic engagement. There was a lot of people that sort of rallied behind the brand and they were really passionate about it. They found that people’s organic sharing of this video was far and away more than they ever expected and that was a huge signal to us of like, “Let’s put some money behind it. Let’s queue it up in Facebook ads and see if that moves the needle.”

It goes against every direct response best practice that we have – which is to keep things short and sweet, keep things under 30 seconds, make sure that there’s a call to action at the end. We said, “You know what? Throw caution to the wind. Let’s put some money behind this 3-minute branding video and see what happens.”

We did that and the ripple effect of people engaging with this video was driving sales more than anything we could have ever imagined. Even though the video was not like a low-budget asset, it was 3.5 minutes, it tugged on your heartstrings, it made you laugh, it made you feel for the mission of the company, and there was no direct call to action like, “Buy today! Buy now!” There was none of that yet we experienced all of those direct response benefits by engaging people on a really human emotion level. So, that was a really exciting client to work with because it went against everything we knew about direct response for video advertising.

Now, we’re sort of pushing our clients more towards that realm of let’s test things that aren’t typical. Let’s test things that are about your brand and not about direct response and let’s see if we can keep this momentum going. That was definitely something exciting to learn and then bestow across all the rest of our agencies.

Another advertisement that I’d like to talk about is one that was actually the complete opposite. For that example I just told you, it was a client that spent a hefty chunk of change on this branding video – this 3.5-minute branding video. On the flip side, we had a client that came in and they had never run Facebook advertising before. They wanted to test video advertising. At the time, Facebook didn’t have that slideshow effort and they didn’t even really have much in the way of graphic design resources internally.

What happened was the client took his cellphone out and he taped his toddler daughter talking about his product. What he did was he shot his daughter – who was one or two years old at the time – crying for more bacon. As it turned out, it was sort of like, “Don’t let this happen to you. Don’t let yourself run out of bacon.” He shot it with his cellphone in his kitchen. Same as the other story, when he put it up organically on Facebook, people were laughing and loving it and sharing it and engaging with this video asset. He was like, “Huh, there’s something here!”

So, we put some money behind it and – lo and behold – that was like a huge video that we ran for months. I mean, if it has momentum, let it keep going. You don’t have to turn off an ad just because you think it’s sort of run out of steam. You have to look at those metrics and see if it’s still killing it. If it’s still resonating with audiences, let it keep going. So, we did that as well for a client and he literally just took out his cellphone and shot his daughter crying for his product so that was kind of a cool story.

That can be sort of the easy part of running video advertising. You can make it what you want, but just test anything and everything and don’t worry about something being perfect. Don’t worry about your video ad being like TV quality. That’s definitely a lesson learned for advertisers that can’t get out of their own way and are afraid to run video advertising because it’s too hard or requires too many resources. It really doesn’t have to be that way at all.

In terms of video advertising, like I was saying, it’s Facebook’s preferred ad type right now. Obviously, we don’t know how the algorithm works internally within Facebook, but we know that, when we are competing for certain eyeballs and certain impressions, Facebook really prefers putting video in front of their audiences right now because that’s what people are engaging with. And so, we’ve been able to take that ball and run with it across all of our eCommerce clients and push video into their newsfeeds and even on Instagram because that’s what people are engaging with.

We’ve actually benefited from this because there’s sort of lower competition right now for video ad units. We’re able to get in there and lower costs by running video compared to Carousel ads or link ads because there’s more competition for those ad types. Those seem to be a lower barrier to entry for some advertisers but we find that video is really the place to get your most efficient ad spend and really drive sales up by.

In terms of segmenting your targeting, it’s funny – you know, we worked in this field for quite a long time. We work with direct response advertisers specific to Facebook for like four years now. There’s a lot of clients that come to us and they have an idea of who their audiences are. They use demographic information. They use your typical location, age, gender, all that sort of thing to sort of segment their audiences. But there’s literally hundreds of ways that you can slice and dice your audiences.

One of the best ways that we like to segment our audiences for the most relevance is to build lookalike audiences. A lot of our clients come to us and they have email lists or catalogue lists of thousands of people and, by uploading these lists to Facebook, Facebook then takes the information and matches it to people’s accounts and people’s profiles on Facebook.

Now, as an advertiser – full disclosure – we don’t see that information. We don’t see so and so in Wichita, Kansas is this person on Facebook. Like, we don’t get to see that data as an advertiser. Facebook does it on the back end. But, if we have address information, email information, phone information, you can upload that and create a custom audience on Facebook and then build lookalikes.

If you do a one percent lookalike, Facebook will give you one to ten percent to sort of match your audience – to find the audiences to the seed audience that you put up there. If you upload a list of your best customers, your high lifetime value customers or customers that have higher average order value or that are just most valuable to you, you can then build a one percent lookalike and Facebook will match that custom audience with the people that they have on the platform that most look like your custom audience and that most look like your more valuable audiences.

This is one of the first things that we do for a lot of our clients. It’s the best way to get in front of most relevant audiences first and foremost. And then, what you do is you take that one percent and then you can over-qualify them. So, segment them by your demographics – your location, maybe some interest targeting – you know, you just want people that like to travel or go on cruises or you want people that only buy luxury goods.

Facebook has these categories where they’ve been able to identify audiences that have certain purchasing power or certain purchasing behavior or they are searching for a new home or they just got married – all these different categories that you can sort of layer over your own custom audiences to really over-qualify these audiences and you’re more likely to find segments of people that drive your costs down because your advertising then becomes a lot more targeted to them. I think that that’s something that people miss. They sort of just scratch the surface with their targeting but there’s definitely hundreds of layers into targeting on Facebook advertising that involves both things you know about your audiences and things you don’t. It’s important to do any and all of it.

For a business that doesn’t have a lot of qualified clients, a list of a thousand or a few thousands or a few hundreds of people that they know are really good and they want to retarget, create a lookalike audience in Facebook; if they don’t have that already, do you recommend using video ads to figure out who their audience should be?

Yes, I mean, that’s a great way to sort of marry the things we know with the things we don’t. So, we know that video does really well and we know that video resonates and it’s engaging. But, if you’re looking for new audiences and you don’t have a place to start, that’s a good place to start. Particularly, what you’re going to want to do is sort of prime your audience. You can run video advertising and target people for video views.

Even if you have direct response metrics and goals in mind, you might want to start and prime your audience by putting out one video. And then, what you can do is retarget people that have watched 10 seconds, 30 seconds, a minute of that video. Retarget them with another video and then retarget them again with, then a call to action to sort of over-qualify them over the course of a couple of different steps. That’s a good way to sort of build a sequence.

You know, you may get a small audience, but you’ll start to understand the type of things that trigger people to click off and convert. And so, you can do that with video, definitely. You can also do that with Facebook has these partner categories that are sort of self-defined. Even if you might have an idea of your demographic but you don’t have a whole lot of characteristics to go off of, Facebook has hundreds and hundreds of categories in terms of, like I said, college educated, empty-nester, has kids that are teenagers, buys luxury goods, travels and goes on cruises specifically.

Facebook has done a really wonderful job building up these partner categories. And so, I think that’s also a great place to go to if you’re sort of starting from scratch.

Do you have any tips in terms of the actual video content? When you’re doing a video ad, do you prefer if it’s a selfie style, very authentic, raw video? Or is it better to have a planned-out video ad? What do you recommend?

I have to say it’s definitely in one of those “it depends” sort of answers.

There’s sort of a rule of thumb. If something is going to resonate organically, it’s going to resonate in advertising. That’s why, if you’re doing sort of the selfie style iPhone, very sort of low-budget video, if your product or your service can be translated that way and you feel good about the actual content, then there’s no problem running an advertisement that way if you’re just starting out and you want to check it out.

I think, also, if you need to sort of start somewhere, you can dip a toe into that slideshow format that I talked about and just sort of build. Like I said, it’s about as easy as building a PDF. You can plop in some imagery, add some text overlays, add a voiceover or any other sort of music file, and you can test that, too.

I do think, if you have the budget and you can go sort of big TV audience-style advertisement, why not? Try that, too. If you’ve invested some budget in something that’s longer – 3 minutes or 30 seconds or 90 seconds – slice it up into 10 seconds here, 30 seconds there, and just test anything and everything. Like I said before, we’ve been surprised at the performance of some videos. There were some videos that we thought were going to do absolutely amazing and they failed. You really have to test anything and everything to make sure you’ve sort of crossed off everything on your list to find the type of content that works for you and the type of video or ad type that works for your audience.

How many tests are you usually able or do you recommend clients to do within a week or within a sprint of a couple of weeks or a few weeks? Do you recommend doing a high volume of tests?

That depends on your budget, of course.

We have clients that are running a million dollars a month. They set aside 20 percent of their budget just for testing and they can do that sort of higher volume because they have a multitude of audiences to go after. Any given time, we’re running – it obviously depends on the client – we have clients that are very particular who they go after and they have maybe five audiences that they routinely target. Or we may have twenty audiences that we routinely target with two different ad types per audience – I don’t know.

It definitely depends on your budget. I think, if you’re working with a smaller budget, you’re going to want to be really judicious. Maybe find that audience that seems to always really engage with your advertising and test new creative against that high-performing audience – that way, you know the audience is high quality but you just want to see what sort of ad and make sure of the next.