Don’t you love getting a better bang for your buck?
If you’re spending money on Facebook Ads for your online business, every dollar counts.
The better you are at decreasing your cost per action (CPA) (i.e. the average price you pay Facebook for your conversion goal—lead, purchase, add to cart, etc.), the quicker you’ll make money from Facebook ads.
But when you’re new to Facebook ads, or to conversion optimization, it’s normal to feel as ill-equipped to decrease your CPA as a deep-sea tube worm performing brain surgery.
Check out this cool 're-dip strategy' to discover how you can launch a Facebook ad campaign and then use it to help you decrease your cost per lead by strategically increasing the conversion rate of the landing page where you send the Facebook ad traffic.
We used it to increase the conversions (and therefore, decrease our cost per lead) of a landing page by 47.74%…
More...
Control Variation - Conversion Rate: 14.14%
Test Variation - Conversion Rate: 20.89%
Test Results from Thrive Optimize
Start With Data…and Luck
Step 0: What You Should Already Have
The 're-dip strategy' requires you to create a Facebook ad campaign to get more leads using paid traffic. Even though I won't be covering these in this post, they are crucial to your success:
- 1Good ad targeting and some experience publishing Facebook ads.
- 2Have a killer opt-in offer and the ability to collect email addresses with an opt-in form. If you're using "Sign up for My Newsletter," please take this free email course and discover how to drastically improve.
- 3A decent lead generation landing page for your offer to send your Facebook ad traffic to. Need to learn how to make one? Take this free Landing Page Building course.
Note: If you already have a landing page getting traffic without running any Facebook ad traffic to it, you should publish a clone of the page to use for the Facebook ad traffic, as it will give you better data.
Step 1: Create Your Ads & Publish Them
Create 18 ad combinations from 3 images, 3 headlines and 2 ad texts for the ad campaign. I use AdEspresso to do this quickly, but you can also do it manually using Facebook's Ad Manager or their Power Editor.
3 images, 3 headlines and 2 ad texts to create 18 total ads.
This many ads gives you variety, but not so much you have to wait months to find the winning ad variations. Try to publish ads highlighting different value propositions of your offer.
Step 2: Find the Winners
This is part data and part luck because a large uncontrollable in your ads’ success/failure is the Facebook algorithm.
Facebook wants to make sure their users have a good experience, so when you click publish on your ads, they aren’t going to just blast them out. Facebook slowly drips your new ad variations to see which ones perform well and then shows those ads more than the ads not performing as well.
For example, one thing the algorithm takes into account is the level of engagement your ads get through “likes” and commenting. An ad with lots of comments and discussion will get preference over one with none.
Once your campaign is launched, you need to periodically check in and pause underperforming ads.
These are the stages you should go through to find the winning 2-4 ad variations (you can find all of this data in the tool you're using to publish your ads, e.g. the Facebook ads manager):
'Re-dipping' Your Ad to Build Your New Landing Page Test Variation
At this point you have 2-4 high performing ads sending visitors to your original lead generation landing page from Facebook.
You’ve optimized the ads to make sure you’re getting the best outcome, now it’s time to try to improve the conversion rate of the landing page by 're-dipping' your ad and using it to create a new landing page variation to further decrease your cost per lead.
Creating your original lead generation landing page is hard enough (your control variation), but if you want to start improving conversions, you’ll need to come up with different angles to test against your control to see if these different angles resonate with your audience more (because the conversion rates are better).
These small wins add up to massive wins for you over time.
To have the best chance of getting a win, we recommend testing “bowling balls vs. paperclips.”
This means you want to test landing pages blatantly dissimilar from each other (like the difference in weight between a bowling ball and paperclip), because it takes less time to find if one is getting significantly more sign ups than the other.
Coming up with winning 'bowling ball' angles is hard, which is why I love the 're-dip' strategy: not only is it a quick and easy win to find a new 'bowling ball' angle to test (by using one of your high performing Facebook ads from step 2 to 're-dip' for a new landing page variation), but it also increases congruence.
Good congruence will (probably) be better for your conversions.
It's when someone has the feeling of “Yup, I’m in the right place. This is what I wanted and expected” after they click your ad and see your landing page.
This is not the experience you want your visitors to have when they see your landing page after clicking your ad.
Choose one of your winning ads and use it as a model for your new lead-generation landing page variation (the re-dip version).
You want to use similar colors, images and copy from the Facebook ad on your new landing page because this will improve the congruent feeling from your ad to the page.
Here are some examples of the re-dip strategy in action:
Facebook Ad
Re-Dipped Landing Page
Facebook Ad
Re-Dipped Landing Page
Click to Enlarge
You Better Test That
You have your new landing page variation. Nice!
Now, because you shouldn’t blindly take advice from some guy on the internet about YOUR business as The Truth…
You’ll want to run a test to see if the 're-dip' variation you built is ACTUALLY going to perform better than your original. You can’t just assume better congruence will lead to better conversions for you because it did for us.
The best way to test which landing page performs better with your Facebook Ad traffic is with an A/B test.
Here’s how to set one up under 1 minute using Thrive Optimize, so you can see if the congruence hypothesis is true for YOU.
Set up your A/B test between the two variations (original version vs. 're-dip' version) and start it. Now leave it running until you have statistically significant results on which page performs better.
With Thrive Optimize you can enable “automatic winner” mode, which will show the winning landing page variation once there are significant results so you don’t have to worry about interpreting the data.
This is huge because you get to keep your original Facebook ad campaign running the entire time the A/B test on your landing pages is happening.
And if your 're-dip' variation wins, Thrive Optimize makes the switch automatically, which means you don’t have to make a new Facebook ad campaign. You get to keep your winning ads running (the ones already pleasing the Facebook algorithm with good social proof through likes, shares and comments), while also improving your landing page conversion rates at the same time.
Ideally, you get a higher conversion rate on the 're-dipped' landing page, which means you’re getting a better payoff by spending less money for new leads!
And that’s the entire 're-dip strategy'! Here it is again in 4 steps:
- 1Use 18 Facebook ads to drive traffic to a lead generation landing page.
- 2Slowly trim the 18 ads down to the best 2 to 4 performers.
- 3Pick one of the ads to use as inspiration for a 're-dipped' landing page variation.
- 4Test the two landing pages against each other to try to increase your conversion rate and decrease your cost per lead.
What do you think? All questions and comments left below will be answered!
Nice article. I was wondering about how to test combos of pages/ads to see what works best. This method, although not foolproof, has a excellent shot of giving you good results. One small quibble — increasing your conversion rate 47 percent, will not decrease your cost/lead 47 percent. It’s closer to 32 percent.
Hi Keith, glad you enjoyed the article.
Yes, you are correct on that math. I realize how I wrote it is a bit confusing, but I didn’t mean to imply because we increased the conversion rate of the page by 47% we also decreased our lead acquisition costs by the same percentage. I merely wanted to make the point if you increase the conversion rate of a landing page, you will also decrease your CPA.
It was an excellent article. It gave me a much better sense of how to approach testing on a limited budget 🙂 In the promo email, it was “Increase landing page conversions by 47.74%, which translates into a cost per lead decrease of the same amount!” That’s why I called it out here. It’s an easy mistake to make; I’ve made it plenty of times myself. I wanted to help others avoid it.
Ah gotcha! Thanks for clarifying that, Keith 🙂
20% that’s terrible. What were you paying per click? What was your ROI? Did you make or lose money?
Terrible compared to what?
Have you also run ads to cold traffic where the conversion goal requires the visitor to create an account for online marketing courses? If so, I’d love to hear your numbers, see the ads and landing pages so I can learn your magic!
Thank you
You are most welcome 🙂
I’m a bit confused, so I hope you can clarify this:
Do you point each of the 18 different ads to the same landing page? And if so, how different can you make these and still keep congruence? (Would be great if you could show examples of different ads you tested, actually, so we can see how different we can make them 😉 )
I mean, the examples you show alongside the re-dipped landing pages show two very different ads. One is about a single course, the other about 30+ free courses. Were these pointed at the same landing page initially, or are they re-dipped pages from different campaigns?
Came here to post this exact thing. Also what about the FB pixel? How does that come into play with all of this?
Your Facebook pixel should already be installed on your website before you attempt anything discussed in this post. There are some interesting things you can do with audiences you create using the pixel to decide who you include and exclude from your ads, but that’s beyond the scope of this post.
Hi Robert! Let me see if I can help clarify…
Yes, each of the 18 ads are pointed to the same landing page.
For the 18 ads, there will usually only be a few ads within the 18 ads that are congruent with the version 1 landing page. So don’t get tripped up on this aspect of creating your ads.
However, once you have trimmed down the 18 ads to 2-4 that are doing well, THIS is the point where congruence comes into play. Pick one of the winning ads and use it to then create a congruent landing page (this is your re-dipped page). Then you want to set up an A/B test between the two landing pages to hopefully get a conversion boost.
For the two examples, those are for two completely different ad campaigns so sorry if I didn’t make that clear. The point of the examples was to illustrate how you go from a winning ad variation to a congruent landing page so you can see how similar they are.
Hope this helps clarify, but let me know if you have any more questions!
Thanks for the explanation. A few more questions though 🙂
If you are testing 18 ads, and they’re all pointing at the same landing page, but some of them aren’t congruent with the landing page, wouldn’t this affect the result?
Wouldn’t the logical winners be the ads that are most congruent with the landing page? Because they would usually get the most conversions, right?
Or, are you focusing more on which ads get the most clicks, rather than on which get the most conversions?
Or do you keep the original control page purposely generic in terms of color, image, etc. to avoid causing too much incongruence?
I notice that the control variation doesn’t have any of the color, images or copy of the orange-background ad.
If you are testing 18 ads, and they’re all pointing at the same landing page, but some of them aren’t congruent with the landing page, wouldn’t this affect the result? –> Yes, that’s the point.
Wouldn’t the logical winners be the ads that are most congruent with the landing page? –> Maybe, but that’s why it’s important to test. Here’s the idea: you come up with your best shot at a landing page variation, but it’s only ONE option. Then you create 18 different ads highlighting different value propositions and trying out different image ideas. Now you have EIGHTEEN ads running (with the FB algo involved as well) and there are 2-4 ads that will be the best.
Like I said in the post, you start with impressions, then clicks and then conversions to find these 2-4 ads, so at the end you have ads that have good CTR, good conversion rate and the FB algo likes them.
Now you have an ad that you know works because you’ve tested it. Then, because you know the ad works the idea is to create a similar landing page to see if you can get another boost from congruence.
Dave, why are you so amazing?
Probably because my mom rolled me around in amazing sauce when I was born. Powerful stuff that amazing sauce 🙂
You can now use the Dynamic creative tool to create multiple variations of your ad elements like headlines, text and creatives. FB does the work for you and gives you reports on what’s winning. You can turn off combinations that are not winning.
Nice! Thanks for sharing that Aja! It has been a while since I’ve looked into what the native FB ad manager offers.
Many thanks Dave…. Fantastic! This is like gold!
I loooooove gold! Thanks, Mark! I’m glad it was helpful 🙂
Wanted to comment even before watching. I’m SO STOKED to start using the new Thrive Op
Awesome to hear Rob! You’re going to love it, it’s super powerful 🙂
Great teardown of the winning ads. Thumbs up. I am wondering how many variables are recommended to change during a A/B testing. Some say just change 1 variable (eg. Title text) at one time. If we just change 1 variable at a time, it takes many A/B testing though.
Hi Cheefoo,
It’s more about coming up with better hypotheses and not just changing 1 variable at a time. For example, “I think a long-form sales page will perform better than an above the fold sales page because…”
The bigger and more dramatic the change is, the less data you’ll need to come to a significant conclusion. Check out the link in the article about testing bowling balls vs. paperclips as well as our A/B Testing Quick Start Toolkit in Thrive University to help you really nail down your testing strategies.
Amazing post !
Thanks Martin!
Thanks for the article, Dave. When I read Hanne’s email, I buckled up for a very complex approach to testing FB Ads. Your strategy is very streamlined. Will take it for a spin this week.
Awesome, Martin! Let us know how it goes 🙂
Hi Dave,
Thanks so much for such an incredible article. Just brilliant.
I run a very successful Stop Smoking Seminar here in Ireland and the UK.
Besides our Private Consultancy, we host the seminar at various venues.
Most of our clients are recommendations from others who ahve attended.
Additional advertising has been via the local regional papers, however, I am now looking at the Facebook ad set up and learning as I go.
My concern regarding the Facebook ad set up has to do with the OBJECTIVE of such.
I can’t see how one could effectively and specifically ‘target’ as I’m sure Facebook would have no idea who smokes and would therefore require our expertise.
I have used the REACH objective with some good success, but can only put in the AGE and LOCATION of where we are conducting the upcoming clinic.
Would you have any other suggestions in light of our difficulty of being able to specifically target the smoker? Or should I just keep going with the Reach objective?
Thanks so much.
Warmly,
Geoff
Hi Geoff,
There’s lots of options for you. This post that I linked to in the blog: https://staging.thrivethemes.com/facebook-targeting-hack/ might give you some more ideas on how to target a bit better.
However, because you’ve already run the seminar a few times you could use the email addresses from your previous attendees. You’d upload the list to Facebook and use it to create a lookalike audience from, this would be the simplest method.
Thanks Dave for all your kind assistance. 🙂
No problem, Geoff! Glad I could help 🙂
Hi Dave,
What if we have our own custom 2 step (Gravity) form on the landing page? Can Thrive A/B testing still be used?
David
Hi David,
Yes, you would still be able to use Thrive Optimize for this. I’m not sure what your conversion goal is, but let’s say it’s subscriptions. Since you use a custom Gravity form you’d use the “Visit goal page” in your Optimize test and set it to the thank you page. This data is what would be used by Thrive Optimize to determine the winning variation.
Thank you Dave for this strategy. I would like to ask you for your advice. What would be the best way to monetize a landing page on my blog using FB ads: Selling products directly or generating leads with an opt-in offer and later monetize the list through email marketing?
Hi Juan, unfortunately, there’s not an easy way to answer this question because it depends on so many factors like: what you’re selling, what you’d offer for free to get leads, how compelling your ads are, how compelling your landing page is, how easy it is to give you money if someone wants to buy, how good your email marketing is, etc.
So, my advice to you is to pick one, try it and gather data. This will give you a benchmark to improve upon. The quickest thing to ship would be to sell products directly (usually best with lower priced products) from your Facebook ads.
Thank you!
No problem!
What would a minimum viable test budget be for this strategy?
I recommend $10 per ad, per day. So that would be $180/day for 18 ads. You can reduce the daily spend and the number of ads you’re running but I wouldn’t go any lower than $7 per ad (otherwise it will take too long to gather data) and I wouldn’t go any lower than 8 ad variations.
That would be 8100 USD for a 45-day campaign!!!! That’s a lot of money! Luckily, I’ve seen in a FB marketing vid that in Europe, FB ads are much cheaper (I live in Belgium). I hope it is so, because I’d like to start my campaign immediately when my website is ready (I’m working on it). Anyway, can you recommend some cheaper version strategy? Just in case…
I wouldn’t recommend jumping into FB ads immediately. Facebook ads is a strategy to scale a business that already has money coming in the door…you have a proven idea of what works and so FB ads is a way to grow.
The idea is proven already: it is a patented alternative healing system, one of the best currently available in the world. I want to take it to Belgium and I need online marketing for this. Maybe it is better to start with simple Google SEO and Youtube Funnels, test how that works and than turning to FB ads, if necessary…. I don’t see this clearly.
Awesome! Then yeah, I’d work to understand your numbers as we were talking about in a different thread and then you can start playing around with FB ads and the strategy outlined here!
If you are gradually reducing the nr. of your ads, the total budget will be less than 8100 USD. But still, it would cost thousands of USD. That’s a bit too much for a starting one -person business.
Yes, it’s important to know the costs associated and your numbers (like how much a lead is worth to your business). This allows you to know when to pause an ineffective campaign or to start scaling it.
Thank you. I will think about this.
Dave – this is an excellent article and I would love to A/B test my landing pages using Thrive. But when I tried it, the test didn’t work due to a caching plugin. I need both Thrive AND the cache plugin, so using your A/B testing feature is a non-starter. 🙁
Hi Maria, sorry about this. Did you open a ticket with our support team to see if they could figure out this problem? If you haven’t, here’s how to open a ticket: https://staging.thrivethemes.com/tkb_item/how-to-raise-a-forum-support-ticket/
Hello!
First of all I would like to thank you for the article.
I have one doubt:
When you say that we test for impressions, do you mean that the campaign objective is reach?
When you say that we may test for clicks, do you mean that the campaign objective may be traffic?
Or do we set the campaign objective as conversions for all tests, since we may have data on impressions and CTR?
If the last option is the correct, then incongruence may be high (I have read the other comments and I understand that this will happen).
If first and second option are the good ones, incongruence will be reduced since we can design the landing page according to the last 2-4 winners. Am I right?
Thanks!
Hi Tania, glad you enjoyed the article.
I set the objective to be “optimize for offsite conversions” where the conversion, in this case, is a lead (getting someone to sign up for our free Thrive University).
When I’m looking at the data I first look at impression numbers to eliminate ads that FB isn’t favoring, which is included in the data.
Then I look at CTR data, which is again included in the same data, to find ads that are getting good engagement from viewers.
And yes, this might mean that incongruence is high between the 2-4 winning ads and the landing page. Which is why you then want to re-dip, to try and increase the landing page conversions with increased congruence.
Thanks Dave,
Have a nice weekend!
You too!
That is amazing
Thanks Gurdeep!
Hey Dave,
Thanks for the insight. This is very helpful.
A few quick questions. Do you ever use the new Facebook ads split test feature in conjunction with Thrive Optimize? Would that screw up your data any if you were running both an A/B test on the ads in Ads Manager to find the best ads while simultaneously running another test on the site using Thrive Optimize to find the best landing page?
If so, what do you recommend doing in the event that one ad performs better in Ads Manager but it’s corresponding landing page performs worse? Where’s the priority in your mind and how does this affect the congruence that you mentioned?
Thanks!
No problem, Bryce. Glad it was helpful for you 🙂
I haven’t used the new Facebook ads split test feature at all. But if you’re using it I recommend only running one test at a time. Run your ads test to find the best performing ad and once the test is done, use Thrive Optimize to split test the landing page.
That was very much helpful Dave
Things that helped me to scale up facebook leadgen campaigns is using landing pages with call schedules such as appointlet
Glad you found it helpful! And that’s an awesome strategy, thanks for sharing 🙂
The 4 points at the end explain it well. The other stuff was confusing.
Well I’m glad the wrap at the end was helpful! What could have made the other stuff more clear for you?
Hi Dave, great post. Thanks.
Question: When choosing subscription as a goal, will it also work with ‘normal’ forms as Gravity Forms or even the new forms in Thrive Architect?
Hey Tom, glad you enjoyed the post. Am I assuming you’re talking about subscriptions as a goal in Thrive Optimize?
If so, yes, you would still be able to use Thrive Optimize with gravity forms. If you’re using a custom gravity forms with subscriptions as the goal, You’d use the “Visit goal page” in your Optimize test and set it to the thank you page. This data is what would be used by Thrive Optimize to determine the winning variation.
Ok, thanks Dave. Will it work as well with the new forms option in Architect itself?
Hi again Tom, Thrive Optimize will work with any Thrive Leads opt-in form.
If you’re talking about the new forms as in the new contact forms available in Thrive Architect, you can use Thrive Optimize to test using the same strategy I mentioned above. Redirect the user to a thank you page after submitting the contact form and use the thank you page as the submission goal in Thrive Optimize.
Thanks, this was a helpful article, I’m going to apply some of these learnings to my business.
Awesome to hear, David! Let us know how it goes 🙂
Very Informative!
Thanks, Sanju! Glad you found it helpful 🙂
Awesome and Highly informative article, Dave…
Glad you enjoyed it, Aneesh!
Hey this may be a silly question but If I want to run FB Ads for multiple clients with different funnels/landing pages, will I have to do it from my website domain in WordPress? How would I go about doing this? As i know with Clickfunnels you can just create multiple landing pages and run ads straight to them. Would love to hear back!
Hey Cameron,
You have 3 options:
1) Get access to your client’s website and create the landing pages for them within their WordPress environment.
2) Create subdomains on your WordPress environment for each of your clients (this is basically what ClickFunnels does) and create your funnels/landing pages within each separate subdomain.
3) Create your different funnels/landing pages within your WordPress environment.
Hi Dave,
Thank you so much for responding, is there a guide on how to create subdomains on my WordPress environment? Thanks
It can be slightly different depending on what hosting you use, so go to google.com and search “how to install a subdomain on [YOUR HOSTING] (bluehost, godaddy, wpx, etc.)” and you should be able to find a guide on how to do it.
Here’s a decent one if you’re using cPanel to install the subdomain: https://wpsites.net/wordpress-tips/what-is-a-subdomain-how-to-create-sub-domains/