Read on Twitter
1/26 How We Improved @BambuEarth’s Facebook Ads Performance 250% In Less Than Two Weeks: A Thread
I was looking over the results from a recent win my team had and realized there were some replicable principles for other ecom folks.
So I wrote it up. https://commonthreadco.com/blogs/coachs-corner/effective-facebook-ads-campaigns
I was looking over the results from a recent win my team had and realized there were some replicable principles for other ecom folks.
So I wrote it up. https://commonthreadco.com/blogs/coachs-corner/effective-facebook-ads-campaigns
2/ Let’s start with the problem: the other day I tweeted a graph showing volatile (and recently declining) customer acquisition for Bambu Earth– and also showing that my team appears to have fixed it. https://twitter.com/andrewjfaris/status/1247931294681419776
3/ The turnaround came via a few different efforts, with a big part of it via a @TaylorHoliday-inspired "message the moment" offer & funnel during COVID.
We can break it down in 3 parts (and the order is intentional):
1. The Offer & Angle
2. The Landing Page
3. The Ads
We can break it down in 3 parts (and the order is intentional):
1. The Offer & Angle
2. The Landing Page
3. The Ads
4/ Before I dive in: I teased out the process in the most detail on The Ecommerce Playbook podcast this last week.
The episode is only 18 minutes, and you can check that out here: https://podcasts.apple.com/us/podcast/250-improvement-on-facebook-ads-heres-how-we-did-it-episode-8/id1087933109?i=1000470965777
Back to it...
The episode is only 18 minutes, and you can check that out here: https://podcasts.apple.com/us/podcast/250-improvement-on-facebook-ads-heres-how-we-did-it-episode-8/id1087933109?i=1000470965777
Back to it...
5/
First, the offer & angle:
You won't find a cleaner ingredient list on a skincare product than Bambu Earth. Which is great!
The only problem is that switching to Bambu can cause short-term breakouts as your skin purges the harsh chemicals you used before.
First, the offer & angle:
You won't find a cleaner ingredient list on a skincare product than Bambu Earth. Which is great!
The only problem is that switching to Bambu can cause short-term breakouts as your skin purges the harsh chemicals you used before.
6/ So what better moment to try us out than when you're stuck inside? Thus the angle: "The 6-Week Skin Detox" through our Complete Skincare Collection.
Featuring that product is not an accident, and it's crucial to be strategic about this.
Featuring that product is not an accident, and it's crucial to be strategic about this.
7/ We chose it because it's:
AOV
Margin
LTV
Customer Experience (thus the LTV)
All of this lends well to short and long term success, which is important if it works: you want scalability here.
Now we've got an offer and angle that work both broadly and in this moment.
AOVMarginLTVCustomer Experience (thus the LTV)All of this lends well to short and long term success, which is important if it works: you want scalability here.
Now we've got an offer and angle that work both broadly and in this moment.
8/
Second, the landing page.
Here's where we started driving traffic: http://www.bambuearth.com/pages/skin-detox-1.
The basics are in place: CTA at the top. A features & benefits section. Social proof. Product section and another CTA at the bottom.
But there are also a few subtler keys we like.
Second, the landing page.
Here's where we started driving traffic: http://www.bambuearth.com/pages/skin-detox-1.
The basics are in place: CTA at the top. A features & benefits section. Social proof. Product section and another CTA at the bottom.
But there are also a few subtler keys we like.
9/
(1) The headline is critical for tying together the product and angle.
Remember "thesis statements" from 5-paragraph essays in school? That's a helpful heuristic here: every thought on the page should naturally flow from the big idea you get across there.
(1) The headline is critical for tying together the product and angle.
Remember "thesis statements" from 5-paragraph essays in school? That's a helpful heuristic here: every thought on the page should naturally flow from the big idea you get across there.
10/
(2) The "Order Now" CTA is an on-page anchored link. Click there and the page scrolls down to the product section.
Motivated shoppers can get there without scrolling. The rest can consume info as needed.
This gif shows it on desktop, but it's especially useful on mobile.
(2) The "Order Now" CTA is an on-page anchored link. Click there and the page scrolls down to the product section.
Motivated shoppers can get there without scrolling. The rest can consume info as needed.
This gif shows it on desktop, but it's especially useful on mobile.
11/
(3) The "Buy Now" CTA in the product section currently takes you to your cart, but one of our first tests will make this CTA skip the cart and take you straight to checkout.
Full disclosure: we meant to do that here, but made a mistake.
And I'm excited about that.
(3) The "Buy Now" CTA in the product section currently takes you to your cart, but one of our first tests will make this CTA skip the cart and take you straight to checkout.
Full disclosure: we meant to do that here, but made a mistake.
And I'm excited about that.
12/ That's because of the fourth feature of this page:
(4) The lander was built for ridiculously easy testing.
If you didn't notice, the URL has a number at the end of it: "/skin-detox-1".
(4) The lander was built for ridiculously easy testing.
If you didn't notice, the URL has a number at the end of it: "/skin-detox-1".
13/ With every element change on the page, we'll simply duplicate, change the URL number, and put it in our ad account.
Here's a screenshot of where we're at for a lander for @FCGOODS. I expect that to get to at least 20. Maybe 100. Why not?
Here's a screenshot of where we're at for a lander for @FCGOODS. I expect that to get to at least 20. Maybe 100. Why not?
14/ Crucially, none of this requires either:
(1) The subjective CRO opinions of a million self-proclaimed experts. (This is maybe kinda just maybe a little bit of a sore spot for me...could you tell?)
(2) CRO software.
Let's leave the first one and go to the second:
(1) The subjective CRO opinions of a million self-proclaimed experts. (This is maybe kinda just maybe a little bit of a sore spot for me...could you tell?)
(2) CRO software.
Let's leave the first one and go to the second:
15/ We build these in a wysiwyg editor (we use GemPages now but are testing Shogun soon), then start duplicating and changing.
That makes it easy to duplicate a winning ad in a given ad set, change the destination URL, and presto, you've built a landing page test.
That makes it easy to duplicate a winning ad in a given ad set, change the destination URL, and presto, you've built a landing page test.
16/ Now you're built for lots of testing at every level. You can:
• Run one ad to multiple landers
• Run multiple ads to one lander
• Any combo of the two
The process works on any ad network in the world, by the way. Here's a snapshot of the early setup in FB.
• Run one ad to multiple landers
• Run multiple ads to one lander
• Any combo of the two
The process works on any ad network in the world, by the way. Here's a snapshot of the early setup in FB.
17/ Speaking of...
Third, the ads.
We started with these three.
The images here were already existing assets. They’d worked elsewhere as ads, so we changed the headline to reflect the angle as a starting place for this offer.
That was the easiest thing to test the fastest.
Third, the ads.
We started with these three.
The images here were already existing assets. They’d worked elsewhere as ads, so we changed the headline to reflect the angle as a starting place for this offer.
That was the easiest thing to test the fastest.
18/ We've also got some new ads coming with a totally different approach (featuring @A_at_admission, who has now sold a lot of Bambu Earth– thanks Adrianne!).
These ads are video with much more information at the ad level:
https://www.facebook.com/115977923900/posts/10158495926463901 https://www.facebook.com/115977923900/posts/10158495926478901
These ads are video with much more information at the ad level:
https://www.facebook.com/115977923900/posts/10158495926463901 https://www.facebook.com/115977923900/posts/10158495926478901
19/ At the ad level, I still believe everything I wrote below: generate interest, follow with info.
In those still image ads, we're leaning on photography and headlines for interest and the lander for info. The videos bring the info in earlier. https://twitter.com/andrewjfaris/status/1210722181140738053
In those still image ads, we're leaning on photography and headlines for interest and the lander for info. The videos bring the info in earlier. https://twitter.com/andrewjfaris/status/1210722181140738053
20/ OK, the results: what about that 250% improvement I mentioned?
On the podcast, I mentioned $5k in spend. We're up to $13k now on just this funnel (and still spending) with a 1.29 click-only ROAS.
Not great, right? Did I just make you read 20 tweets for a crappy result?
No.
On the podcast, I mentioned $5k in spend. We're up to $13k now on just this funnel (and still spending) with a 1.29 click-only ROAS.
Not great, right? Did I just make you read 20 tweets for a crappy result?
No.
21/ For one, Bambu Earth gets big delayed attribution. That 1.29 (before any remarketing value) will end around a 1.9.
But that's not even the main thing.
The main thing is that this has been a big part of nearly tripling our daily ad spend while INCREASING our ROAS ~20%.
But that's not even the main thing.
The main thing is that this has been a big part of nearly tripling our daily ad spend while INCREASING our ROAS ~20%.
22/ So "250%", I admit, is a back-of-the-napkin number. And it undersells the result. Because if you've read for this long, you probably know that scale and efficiency are inversely related.
That is:
Spend =
ROAS.
So this is actually a huge result.
That is:
Spend =ROAS.So this is actually a huge result.
23/ Plus, we're actively taking as thin of a margin as possible. Why?
Because of these charts, which show returning customer revenue week-over-week and move-over-month since we really started pushing last year.
So we're highly confident that our customers will come back.
Because of these charts, which show returning customer revenue week-over-week and move-over-month since we really started pushing last year.
So we're highly confident that our customers will come back.
24/ Side note: if you want know what's going on with that spike in March on the weekly chart, that's when we sent the email @TaylorHoliday talked about here: https://twitter.com/TaylorHoliday/status/1242909552036556800
25/ Finally: content like this can make me look like I'm the marketing hero who snaps his fingers and fixes it.
But that's just not reality. You only win this game with great teams.
But that's just not reality. You only win this game with great teams.
26/ My part was pushing some team members to try something different and offering maybe three opinions along the way.
How did it actually work work?
@KohlmanVerheyen, @vincent_wu_817, @RobbieAronson, Chris Johnston, and Amber Hawthorne actually made this happen.
My team slays.
How did it actually work work?
@KohlmanVerheyen, @vincent_wu_817, @RobbieAronson, Chris Johnston, and Amber Hawthorne actually made this happen.
My team slays.
