Facebook Lead Ads are so simple. When the ad appears on your feed, you fill in your contact information without ever leaving Facebook.
In contrast, with Landing Pages, you have to go to a separate website where load times, intimidating forms, and distractions potentially take you off course from providing your contact info.
Thus, I hastily made a conclusion when setting up our company’s re-targeting campaigns: Lead Ads were the best type of ad to run!
I enthusiastically told my boss, our CEO. His face was expressionless. He disagreed.
“No,” he said. “We don’t know that. Here’s 2,000 dollars. Run an experiment.”
The Experiment Set Up
While we’ve been quiet about it at AdEspresso, our experiments happen in a special place called the AdEspresso University.
The University is a members-only space – filled with video courses, webinars, and a private Facebook community – but feel free to check it out after you see what I did with my boss’ money 😉
For this experiment, I created two “evergreen” Facebook Ad campaigns targeting our website visitors.
The mission of the campaigns was to get a website visitor (warm traffic audience) to download one of our eBooks (provide their email address so we could then offer them more eBooks, and eventually a Free Trial of AdEspresso).
If the user didn’t re-visit our website or download an eBook after being shown a series of eBook ads after18 days, they would be excluded from seeing any more ads.
Campaign Set Up:
The first campaign to run was the Landing Pages campaign. A few days after it finished, I ran the Lead Ads campaign.
The campaign was based on timeframes of website custom audiences. I also excluded anyone that had downloaded an eBook before, had a Free Trial before, or were/had been a Customer.
We wanted to attract new leads only.
The only difference between the campaigns was that one went to a landing page to fill out a form to get the eBook, and the other allowed the user to fill out their information in a form without leaving Facebook via a Lead Ad.
Daily budget of 60 dollars per day, with a total spending cap of 1000 dollars per campaign. So 2,000 dollars total.
Landing Page Ad Example:
Once they clicked “Download” they were taken to this page:
Lead Ad Example:
Notice that it looks identical to the landing page ad. Except once they clicked Download, this form popped up so they didn’t have to leave Facebook:
Ad Set Structure:
Each ad set contains only one ad – for a specific eBook – and this ad was served to a website visitor based on the timeframe they visited the site. WCA stands for “Website Custom Audience.” As soon as someone visits the site, they would enter the campaign(except for the excluded audiences previously mentioned), and begin seeing an ad for an eBook – “Do’s and Dont’s of Facebook Ads.”
- Day 0 (initial visit) – Day 3 since website visit: Dos and Don’t eBook
- Day 3 – Day 5 since website visit : 500+ Facebook Ads eBook
- Day 5 – Day 7 since website visit : 137 Instagram Ad Examples eBook
- Day 7 – Day 10 since website visit: 500+ Twitter Ads eBook
- Day 10 – Day 12 since website visit : Social Proof Marketing eBook
- Day 12 – Day 15 since website visit: Facebook Custom Audiences eBook
As you can see, each ad lasted about 2-3 days.
I also set it up so that if the user clicked on the ad, they would no longer see that particular ad anymore, even if they re-visited the website (I’ll explain how I did this later).
And of course, once the user downloaded an eBook, or if they started a Free Trial without downloading an eBook, they would exit the campaign and not see any more eBook ads at all.
- For Landing Pages, the ads would appear on Desktop, Mobile, and Right-Column.
- For Lead Ads, the ads would appear on Desktop and Mobile.
- Those that downloaded an eBook before, or during, the campaign
- Previous and current customers
- Any website custom audience that wasn’t in the ad set’s timeframe (over 15 days since their last visit)
Also, as mentioned earlier, if someone clicked on an ad to go the Landing Page or start the Lead Ad submission process, they were then excluded from seeing that specific ad anymore.
For example: If someone clicked on the ad for the Do’s and Don’ts eBook, but didn’t give us their email for that eBook, they would be excluded from seeing that specific ad for the “Do’s and Dont’s” eBook. However, they wouldn’t exit the campaign itself. After 5 days, they would then be shown the ad for the “500+ Facebook Ad examples” eBook…(still confused? ask a comment!)
What were the results?
Here’s a Google Sheet with the results. You can also see the images below.
Notice that the “Cost Per Action – Lead” (far right column), dramatically increases after 7 days.
We had more clicks and leads via Desktop!
Overall Landing Page results:
Of those that came to the landing page, we had 1,077 become leads. This was a conversion rate of 50%.
Similar to the Landing Pages campaign, the cost per lead dramatically increased after 7 days.
In this case, we had more Leads from Mobile than Desktop.
There were 1,569 clicks on the Lead Ad Form, and 1,057 people completed the Lead Form. So a conversion rate of 67%. Nice!
Overall Lead Ad results:
So who won?
Technically, from a cost-per-lead perspective, Landing Pages. But just barely! You can hardly call it a “winner.” It had a slightly lower cost per lead at 93 cents (.93 USD), while Lead Ads had a slightly higher cost per lead at .95 (95 cents). It is worth noting though, the Lead Ads cost-per-lead dramatically spiked after 7 days relative to how the Landing Pages cost-per-lead performed in the same time-frame.
However, Lead Ads performed better than the Landing Pages on mobile placements, while Landing Pages performed better than Lead Ads on Desktop.
This brings us to the key takeaway in this experiment:
When re-targeting website visitors, Lead Ads are more likely to win on mobile, while Landing Pages are more likely to win on Desktop.
Landing Pages might also have a further edge since it took more effort to fill out the form to get the eBook. So they might be higher qualified leads. Nonetheless, everyone who saw the ads already knew AdEspresso (website visitors), so in a sense, all were pre-qualified.
For cold traffic campaigns, there’s a hypothesis floating around that Landing Pages blow the socks off of Lead Ads, since Lead Ads only required a “name” and “email” (very little commitment). I’ve seen this happen in previous campaigns I’ve created, and that’s why I didn’t test it in this experiment.
Right now, based on the results of this experiment, I’m adapting our Lead Ad campaigns to run on Mobile-only, and Landing Page campaigns to run on Desktop-only. Stay tuned for an update at some point in the future as to how this turns out along with the quality of these leads.
As mentioned, with Lead Ad Engagement Custom Audiences, I can also re-target those who engaged with the Lead Ads Forms, but didn’t submit their contact info (approximately 33% of people). In the same fashion, I can re-target those who visited a Landing Page during the campaign, but didn’t submit (approximately 50% of the people). Gotta love re-targeting!
In both campaigns, the Social Proof Marketing eBook performed very poorly as a lead magnet. The Twitter Ads eBook also performed poorly – with a cost of 4.11 USD per lead in the the Lead Ads campaign. I am going to replace these with different eBook offers to see if I can have a better fit for our audience.
Additionally, in both campaigns, the frequency of the later ad sets went over 5 (on average, the user saw the ad 5 times or more per ad set). This means that, by the end of 15 days, some users may have seen AdEspresso eBook ads for upwards of 25 times – offering an explanation to the higher cost per leads as the campaign went on. For future ads, I would lower the budgets for the ad sets later in the campaign.
Any errors in the Experiment?
If someone re-visited the website, and hadn’t downloaded an eBook, they re-entered the campaign at the beginning. However, if they re-visited because they clicked through on a landing page ad – and were of the 50% that didn’t download the book on the spot – they were then re-entered into the campaign, but excluded from seeing that specific ad again.
I still don’t know how to prevent those who didn’t click any of the ads, but revisited the website, from re-entering the funnel at the beginning and seeing all the same ads again.
Want to see these types of experiments first-hand? Members of our University gain access to our experiments one month before we publicly share it on our blog. Starting in 2017, University members also get to choose what to do with our CEO’s money – every month – with a community-wide vote.
Yup, you’ll get to decide where the money goes!
Without joining our University, you’ll still get to see the results of the experiment…it’ll just be a month after everybody else. So if you want early access to the experiments – as well as membership to our private Facebook group, webinars, and courses – I highly suggest looking into becoming a member.