Designing and Testing Ads on Facebook
There are three different placements for ads: Mobile News Feed, Desktop News Feed, and Right Column of Facebook. The key distinction is that Right Column ads do not show up on mobile devices; therefore, you lose 50% of your prospects. According to Facebook, as reported by SocialMediaToday, 48% of daily active users and 49% of ad revenue come from mobile. However, we’ve been involved in campaigns where 60% of our daily active users and our ad revenue came from users on mobile devices.
News Feed ads are in your prospective customer’s News Feed. In general, they are more social and more effective than the other ad placements. Plus, you can populate the News Feed with great images—it is much easier to grab attention in a News Feed. The reason you see Suggested Post on some of the posts in your News Feed is because the owner of the advertisement has targeted you—either you have an interest in the subject matter and/or you fit its target demographic. The reason you see Sponsored at the bottom of an ad just means that it is a paid advertisement. And, no, one is not more expensive than the other.
This is because as soon as one ad gets a click, Facebook floods that ad with impressions, and the other ads are ignored. The best way to handle this is to pause all other ads in the campaign, and let the first one run for a day or so. Then pause the first ad and reactivate all the others until one of them pulls ahead. Repeat this process until you have at least three ads with decent CTRs. Then pause everything and let those three run simultaneously. You’ll find that all three will get equal impressions, and you can run a proper split test.
Three days is the longest I’ve ever waited, but that was a rare case. It is possible that your landing page is holding things up. You might want to check to see if your page conforms to Facebook’s guidelines. If the Facebook Ads rep who reviewed your ad saw anything questionable there, he might have placed you in a “holding pattern” for a supervisor to review.
These big banner ads are a form of malware and are not hosted through Facebook. If you see these, your operating system or browser has been infected with adware. You should run a virus scan and/or clear all cookies from your browser right away.
No. Don’t even try it. Never use anyone’s image, name, or likeness in an ad without that person’s written consent.
It depends on your market. Some demographics have CTRs hovering around 0.4%, while others are lucky to get above average (0.02%). If you have good competition, it’s going to be tough. In untapped markets, Facebook ads have a great CTR. Personally, we wouldn’t run an ad with a click-through-rate below 0.9%—we find that the ads are no longer profitable when CTRs fall below 0.9%.
If you’re not getting any impressions, you will need to either increase your spend or check your targeting. Maybe you are bidding too low to be clickable. After a certain amount of time, if the impressions aren’t coming, it’s time to test some new ads.
If you’re using Promoted Posts, your results may improve if you change your photo/image regularly. Just because a formula was getting crazy clicks in the beginning doesn’t mean it can always sustain for long periods. New, fresh images can make a dead campaign come back to life. You can go back and edit your post to test headlines, new images, new call-to-action, and continue your multivariate testing to keep those hot results.
There’s no super secret formula. Run a direct Facebook likes campaign. For it, you’re going to need top quality content that is not just well-made, but speaks to the audience you’re targeting. Make sure it is also highly image driven. You’re going to want to especially multivariate test images for this type of campaign. At the same time, cross-promote your Facebook page on all of your other social channels.
Potentially a lot of things. Are you targeting the same people? Are there more than 200,000 people in your target group? What is the reach of your campaign? Has most of your target group already seen your ad at least three times? Can you set up two landing pages for opt-in content, and split-test them? Have you tried setting up a new campaign with new images, headlines, and ad copy, but with the same landing page? Conversion is a complex process with many factors. It’s important to dissect as much as possible to understand what is going wrong. New campaigns can cure many ills. On the other hand, maybe the whole thing was a coincidence, and you’ve already reached all of the people who are willing to participate in your target audience.
There are two ways you can bid on traffic in a Facebook campaign. One is CPM (cost per mille—your ad gets shown 1,000 times) where you pay for impressions, and the other is CPC (cost per click) where you pay per click, or each time someone clicks on your ad. You can also let Facebook automatically bid for you, or you can manually bid CPM or CPC. Facebook gives you a suggested range and you just put in an amount that is the maximum you are willing to pay for the click. According to Facebook, the suggested bid range is there to help you pick a maximum bid so your ad will be successful.
It’s based on how many other advertisers are competing to show their ad to the same audience as you are. You don’t have to pick a bid in this range, but you can use it to help you pick a bid that will result in a decent-sized amount of impressions and clicks. According to Facebook, you will never pay more for an impression or click than necessary, so entering your true maximum is the best way to make sure you don’t miss out on impressions and clicks you could have received. That being said, we always bid manually, because that is how we find our ROI in advertising. However, when you’re doing a “LIKE” campaign (a campaign for “Like Us” clicks only), this is automatically set up as optimized CPM by Facebook.
What type of ad are you running? It’s generally harder to get impressions on a socially recommended ad (so a page post ad or page ad that includes stories about their friends liking it); if you have it socially recommended, try removing that and see if you get more impressions.
Try this: start a new campaign with an identical ad. Facebook Ads doesn’t show an ad to all 155,000 users at once. Your targeted users have to log into Facebook to view the ad. So it will take time to reach all users. You can try raising your bid even higher for a few days; that might get you more impression share. Put it at $1.25 and let it run on a $30 per day budget. A higher bid, even beyond the actual cost, can increase impressions by quite a bit. Facebook Ads wants to see how high of a CTR your ad is going to get, so it will deliver a high amount of impressions until it has an estimate. If your ad doesn’t get a high CTR, Facebook drops your impressions in favor of an ad that does. The higher the CTR of an ad, the more Facebook makes per impression.
This can change drastically from moment to moment. Facebook is guessing the CTR of your ads and weighing it against the average ad CTR. This is no different from other popular ad systems on other networks. It starts at the average CTR in the marketplace, then gives a CPC estimate and uses that to establish a ballpark CPM. The CPC price adjusts as your ad history gathers more data. No data (on a new ad) means an almost certainly inaccurate estimate, so plan on running the ad for a day or so, then adjusting your bid at regular intervals.
If you’re targeting a group that is hard to refine a list of, say you’re trying to target a very audience, say that you are going for wives of military men. You can go and set up a quick, very basic viral Facebook with tons of photos that prompt likers to “share if you agree”. For example, “Share if you married a hero”. Then, use the “likes” list to target to a very homogenous group.
It all has to do with your WOT (web of trust) score, which is a regulated score that Facebook uses to identify potential spam. It boils it down to a score of red, yellow, or green. Anything not green is going to give you problems. Hosting the Leadpage on your own site can often alleviate this problem (if your site doesn’t have authority problems).
You’ll know this only from testing. It’s different for every product, market, service, and offer. It depends on how much your product costs and how much you spend on your ads. Spend a few bucks, and then pause and assess the results. It’s impossible to give accurate figures for CTR and purchases unless you compare against someone in the same vertical.
Expect to see lower CTR and purchases from Facebook than from a search engine ad campaign. When people surf Facebook, they are not intending to buy something, even if they click on your ad. Mostly they want to see what is new with their friends.
If someone searches for a term related to your product or service on Google or Bing, then she has intent and is more likely to click on the ad and then purchase the product.
So rather than focus on expected purchase rate, just focus on return on investment. As long as you have a positive ROI, keep running your ads and landing page, and ramp up your spending while continuously testing variations to attempt to improve CTR and purchases. If you have a negative ROI, make changes quickly or abandon the campaign and move on to another product or media venue.
All markets can benefit from having a solid Facebook presence, even ones you wouldn’t expect. Search engines are giving more weight than ever to social cues. Having said that, the best markets for Facebook ads are: entertainment, gossip/tabloids, employment, fitness, health, travel, personal and household services, self-improvement, ecommerce for physical goods, webinars, and live events.
With today’s advanced targeting and the gigantic Facebook user base, any vertical has potential to benefit from Facebook Ads. Any product or service that can be marketed via traditional media can be marketed via Facebook.
It depends on your target audience. Is this a local business, nationwide, worldwide? Ideally you should find 200,000 people to target. This keeps Facebook sending enough impressions for a while, so your ad stays alive longer without going beyond your targeted audience. However, if you are a local business with a very specific niche demographic, then don’t try to go beyond that targeted audience, and compensate for that limitation by refreshing your ads more often.
Bidding can be tricky because it changes based on day, time, niche, demographic, location, account age, and other factors. Some people have a policy to start low (which you’ve done) and increase the bid until you get a decent number of impressions. In our businesses, we start high. We normally double the low end of the suggested range. We never pay that much, but it helps to ensure that our ads get shown quickly.
If your CTR is above 0.020%, then you are above the stated average. That doesn’t mean that this is as good as it gets or that you should be satisfied. As we’ve mentioned before, in our businesses, we pull any ad that has a CTR below 0.09%—it just doesn’t produce enough of a return. Certainly, keep an eye on your CTR and start creating new ads with different (but similar) photos. Test a few ads at a time, with the goal of getting a higher CTR and conversion rate. Try different photos first, then different headlines.
This is called retargeting, and it can have a huge positive impact. Google definitely does retargeting on Facebook. You just have to put a Google AdWords retargeting cookie on your Facebook landing page; then your ads can follow those people around the Web.
This is excellent for product launches and affiliate offers. It helps create social proof—many images of your product increase your authority and credibility. This helps to ensure that customers buy from you instead of your competitors.
Unfortunately, there is no tried-and-true method that will always work better than the rest. Fortunately, there are ways to find out which works best for you! Facebook Retargeting has been pivotal in our team’s ability to test and analyze sales funnels.
Facebook Retargeting pixel is a specific feature you add to your website; it creates a list of who visits your site from an ad. That way, you can use this info to track conversion for each case: sending the audience to a blog post, to an opt-in page, or to a video. Those stats will show the winner for conversion.
What we do know is that if you’re trying to sell a tripwire, focus on that. Doing a tripwire campaign as well as a Facebook likes campaign or a boosted post often just doesn’t work out. Multi-tasking promotion work leaves more room for making errors.
Try directing the traffic to a content page on your site (one that will likely have a better rating). Then, put a Facebook retargeting pixel here, and drop a link to the seminar there. It is a workaround to be sure, but could be an effective second option.