This is a case study post – As of starting this post, I am 6 days into running the campaign in this case study.
In this post I am going over how to run a profitable ad on Facebook. This is aimed at people who have ran a Facebook ad before but are having trouble getting things working for them.
If you’re new to Facebook ads, don’t be deterred. Read this post and bookmark it because you can put it into use once you understand how to run an advert.
A lot of what you are about to read is mandatory for running ads online. Not just Facebook, this is for running ads on all platforms. Test test test!
I am starting off this ad campaign with a broad lookalike audience. I am targeting women, the audience size is 900,000. I created the lookalike audience from my Shopify buyers.
That’s an important step. I created a lookalike audience using a list of 1000+ buyer emails. I exported them from Shopify and created a custom audience on Facebook. When the audience was ready, I created a 1% lookalike audience for the USA.
Next I created a website conversion ad. I set my pixel conversion event to “initiate checkout”.
Find Your Control Ad and Gather Data
Make sure you have a decent budget to get enough data. Initially you should start off from the get-go split testing around 3-5 ads. You want to find your control ad.
Start off first of all with three identical ads except the ad copy is different in each ad. Let the ad run and gather data to see which performs best.
Then take the ad with the winning ad copy and create a new campaign. Make sure it’s a new campaign.
Now you’re going to do the same again except this time you’ll split test three different images. Duplicate the advert with winning copy three times but use a different image for each ad. Again keep running the ads until you find the best performing advert.
After this you should have a half decent advert to get started. You never stop split testing or trying to beat your control. There’s always something you could be doing better.
The next step in the split test process is to test different landing pages. If this is possible for you then test three different landing pages until you find one that converts the best. Then once you find the best one it’s a matter of testing the winning landing page against a similar page with minor changes.
Once you have a control ad, you should always be trying to beat it by running an advert along side it. Testing one thing at a time against the control ad. Test images, headlines, anything that can improve your ad.
I tested three different types of ad copy and found what one the audience liked the most. One got a 8% CTR compared to the others getting 2% and 3%. So now it’s a case of trying to beat my control. As you will see the audience has reacted well to the ad as I am getting an amazing 0.04 cents per link click.
PLEASE PAY ATTENTION to what I just wrote above… If I did not start off split testing my ad-copy and went with one ad straight away, guessing what my audience would like, then I probably would have got bad results. I would have thought my ad was garbage and stopped the campaign after a few days thinking it didn’t work.
After You Gather Data Analyze The Reports
Tomorrow I want to start scaling my advert but before I do that it’s good to dig into the data you’ve gathered with your ad and look at what’s working well.
If you don’t do this you are probably leaving a lot of money on the table by not cutting out non-performing traffic and not focusing on what’s converting the best.
For the first 6 days of this campaign I targeted a broad demographic, 35-64 year olds. Now I have the data we can look over what’s worked and what hasn’t.
I have been running my ad for 6 days with a small budget of $10 a day. I targeted women aged 35-64 on Mobile devices. I have started off with mobile because it has always performed better in my niche. I will incorporate desktop later.
I have had 1270 clicks to my website, 12 people hit the checkout button and 8 people purchased. There’s probably more I could to do on my website to get that number up but this is cold traffic after all.
Important note: I am sending women to a collection of products aimed at women on my website. I did this because I wanted the products to be relevant to the target audience.
Make sure you send your traffic to the actual product you are advertising! It sounds obvious but I’ve seen companies advertise a specific product BUT when you click on the ad, you’re not sent to the product being advertised.
As you can see, 35-44 is my best performing age group if we’re going to go by conversion data. I spent $15.77 and 8 people went to the checkouts. That’s just $2.02 per conversion. However if we look at 55-64 year olds I have spent $17.59 and only had one conversion.
If I never looked at the data I would not know this. Every morning you should check your reports to get an overview of how your ad is performing. Stay on top of what’s working!
Overall I have spent $60 so far and made $357 in revenue. (Around $165 total profit after ad-spend, products costs and shipping etc).
Moving forward I will be putting more concentration on 35-44 year old women. I won’t stop advertising to the other ages completely just yet although it looks like 55-64 year olds are not looking promising. However that could all change when I retarget them and they get familiar with my brand.
Also I will include 25-34 years olds in my targeting as it looks as though the younger people are converting better with this advert.
Build A Custom Audience For Retargeting
You cannot miss this step. I won’t go over how to create a custom audience in detail, there are hundreds of websites/youtube videos that cover this topic
You need to create an audience for all the people that land on the URL you are advertising.
As you can see after one week my custom audience is showing 1,100 people have hit my landing page. Now my next step is to start a retargeting campaign.
Now they have seen my website I am going educate them a little bit about the product. I am going to show them a very similar ad to one they have seen except this time I will send them to an pre-sell article on my website. It will have a story about my product and I will educate them on the features and benefits of the product. I will also throw in typical things you might see on a sales page like our guarantee and testimonials etc.
Notice I am not sending them BACK to the page I sent them to in my original advert.
This is for two reasons, one is to make them more familiar with my product and brand. This is cold traffic so I want to build a relationship. The second reason is so I can put them through a retargeting sequence.
I will create another custom audience for everyone who reads the pre-sell article. Then I am going to retarget my retargeting audience (if that makes sense). Using the new custom audience I am going to send them BACK to the original URL they saw in the first advert. This time with a 10% discount to help them make a purchase
The reason I am going out of my way is because this is cold traffic and they don’t know me. 99% of the 1100 people who landed on my website looked but didn’t buy and a big reason for that is because they don’t know who I am! So the goal with all this is to build trust and get them familiar with me and my product.
Here’s an overview:
Cold traffic see’s my advert, clicks through to website > Retarget non-buyer cold traffic with education pre-sale page > Retarget the people who landed on the pre-sale page and offer a discount > $$$
An alternative step you could do is retarget the pre-sale audience with a giveaway. They are familiar with your product now so they will opt in to a giveaway. Get them on a list so you can continue to market to them. This is no longer cold traffic, they are good leads and are more likely to buy in the long run.
This is where I am at so far and here’s what I plan to do this week…
- Create new campaigns for my best performing age-groups.
- Scale up my initial ad but stay within budget
- Exclude people who already land on my page
- Write pre-sale page
- Send custom audience to the pre-sale page
- Keep split testing against the control ad for the main advert
I will report back next week with an update and the results. This is more than a case study, it’s a step by step guide into running ads on Facebook. It might not be the best way to do things but it works for me. Keep your eye on this post and put it into use.
QUICK UPDATE – Tuesday, 17th Jan 2017
Yesterday was spent creating the pre-sale page for the retargeting audience. I opened with a story, then spoke about my most popular products, ran through features and benefits, threw in some testimonials and a guarantee.
I am using clickfunnels for the pre-sale article because I can create a much better looking article for mobile users. I created it first inside a page on my shopify store and it was ugly even on my desktop. Far too small and too many distractions going on around the page.
With Click funnels I could create a page and make sure they could only see and click where I wanted them to.
Yesterday we also made another $136 from $10 spent. Someone purchased four products in one transaction.
Early this morning I put the finishing touches to the sales letter which has allowed me to finally scale out my ads.
The budget has now gone from $10 a day to $75. I kept the original ad going at $10 a day. I then broke down ads into their own age group; 25-34, 35-44, 45-54. Each with it’s own budget of $10 a day.
I also added an ad for 35-44 years olds targeting desktop with a $10 budget. To see how desktop performs and if it goes as well as mobile, then I will create ads for desktop as well.
As you can see I did not put 55-64 year olds into their own ad group. So far I’ve spent $22 at that demographic in the original advert and it’s still showing only one conversion.
However I have created a retargeting ad for 55-64 year olds. If I get conversions through retargeting then I will target them directly in their own ad.
I will also try to target 55-64 on desktops because I think maybe that age demographic are more likely to feel comfortable buying on a computer than on a mobile.
What you have to be prepared for with Facebook is it can take 24-72 hours for yours ads to start performing. It’s frustrating as hell and no one knows why but that’s just the nature of the beast.
*IMPORTANT* Many people have a successful ad and then they scale up their budget to reach more people and the ad stops converting. They turn it off complaining about Facebook has screwed them over… However if they waited and let Facebook do it’s thing, it would have come back to life after a couple of days. Keep this in mind when scaling up a campaign.
Hopefully we see results kick in right away otherwise the $75 will eat into our budget.
I will update again in a few days.