AliDropship’s Premium and Established Stores mostly make money with Facebook ads. In this article, you will find a detailed marketing strategy used by the team for our most successful Premium Store, The Auto Merch.
Most businesses using digital marketing advertise products or services on social media. As Facebook remains one of the most efficient platforms for advertising and selling products, it’s very important to learn how to make money on Facebook ads.
Here you will find a detailed, step-by-step guide on advertising and remarketing methods used by the AliDropship team to promote The Auto Merch, our most successful Premium Store.
How to make money with Facebook ads: The Auto Merch’s advertising strategy
Promoting an online business on Facebook starts with setting up Facebook Pixel which tracks users’ activity. Read this article to learn how to install Facebook Pixel.
Step 1: Preparing a product and a product page
First, the team chooses a product for advertising. No matter how many goods you have on the site, you can’t promote all of them at once. It’ll be too expensive. When picking a product to advertise, the team of The Auto Merch uses the following criteria:
- The product must be popular. Even if you haven’t sold a single item, there are more or less reliable ways to tell whether it’s popular on the market. We’ve discussed a great method of finding dropshipping goods earlier, but the idea is to check how many likes, comments and shares similar products get on social media or how many sales they make on AliExpress or Amazon.
- The price must be affordable. Don’t forget that running Facebook ads and advertising on other platforms costs money. According to our own experience, the average cost of conversion on Facebook varies from $15 to $30. Since you’ll have to include this figure into the price, the product will get more expensive. So, make sure the final price won’t scare away potential customers.
- The product must be useful and valuable to a buyer. The Auto Merch is built around practical car accessories for a reason. Such goods are much easier to advertise.
- The original product on AliExpress must have suitable video materials. In most campaigns, our team uses video ads since this format shows much better results. It’s also important to find videos that demonstrate how this particular product can solve the buyer’s problem.
After that, the team prepares the corresponding product page. They add customer reviews with photos of the item, write a description with GIFs (if possible), improve the page’s loading speed, etc. The idea is to make a perfect product page that won’t disappoint visitors who come after seeing the ads.
Step 2: Creating video posts
Next, the team creates a number of video posts and publishes them on the store’s page. We usually make 7 video variations and 3 text and thumbnail variations for each of them. As a result, we end up with 21 post variations in total.
Why so many? At this moment, we don’t know which of them will attract more attention, which is why we need to test them first. Besides, we are not going to turn off this campaign completely. As it keeps running, the posts will get more likes, comments and shares, thus becoming more noticeable in the feed and getting more reviews.
We use posts instead of standard ads because Facebook users tend to click more often on ad posts with a high number of likes and comments. Everything these posts earn during this PPE campaign will work for our conversion campaigns later.
Step 3: Launching a PPE campaign
Now the team launches a Page Post Engagement campaign to promote these posts. Such campaigns cost cheaper than conversion campaigns and are great for testing ads. We set the budget to $10-15 for each ad (post).
The budget may seem like too large, but as you will see, we are going to turn off many of these ads within less than 24 hours. So Facebook won’t have enough time to spend all this sum.
As for audience targeting, the team of The Auto Merch uses the following settings:
- Countries: The US (as of now, the team focuses on the US only, but generally we also advertise in the UK, Canada and Australia)
- Language: English
- Placement: Automatic
At this stage, you can ignore target audience demographics unless you’re absolutely sure about your customer’s age and gender. For some product types such parameters are obvious even without any tests or analyses. But we usually specify demographics later when we collect more data from this and other campaigns.
Since PPE campaigns raise engagement, your posts are going to get lots of comments. Make sure you reply to all of them (or at least as much as you can) and answer personal messages as well. It goes to all your campaigns (not only PPEs). Don’t forget that social media are places where users communicate!
Step 4: Testing video posts
The team usually launches the PPE campaign at midnight and checks the results in the morning. The choice of time depends on when your Facebook Ad Manager starts working.
We analyze the posts’ Click Through Rate, their Average Play Time and Cost Per 1,000 Impressions and shut off 1 or 2 worst-performing ads from each group of 3 variations.
I mean those text and thumbnail variations we created for each of the 7 videos. So after this we should have 1 or 2 variations left.
After that the team runs the campaign till evening and turns off every post that didn’t perform well enough. We end up with a maximum of 3 or 4 best-performing ads no matter what variations they are.
Now that we have winning video posts, it’s time to start a Facebook ad campaign aimed at conversions. But as I mentioned before, the team keeps the PPE campaign running so that the posts could keep raising their engagement level. Only now we increase the budget to $15-20 per each post to gather data as well as likes, shares and comments faster.
It’ll help us make money with Facebook ads later.
Step 5: Launching a conversion campaign
The team uses 1-3 best-performing ads to launch a conversion campaign. This one is going to lead Facebook users to our product page in order to sell them the item being advertised. Still, at this point we don’t expect a lot of sales and want to test the audience by interests.
The team uses Facebook interests for targeting. We discussed how to find and test Facebook interests for targeting in one of our previous articles.
Each conversion campaign initially targets at least 5 interests. Later we can add more interests to test them. So, the team creates a corresponding number of ad sets, with each containing at least one video ad and targeted at one interest. We choose countries to target based on the destinations’ CTR in our PPE campaign, but targeting the US is a must.
The team sets the budget for each ad set to half of the product’s price but not less than $10. For example, if the product costs $40 on The Auto Merch, the daily budget of each ad set should equal $20.
You can set a different budget for your PPE and conversion campaigns. Just remember that you can’t make money with Facebook ads unless you pay for them. Campaigns with a small budget collect data slowly while a big budget boosts the tempo. So, if you want to get results faster, feel free to pay more. These are just average numbers.
Just like we do with our Page Post Engagement campaign, we launch the first conversion campaigns at midnight.
In the morning, the team turns off the ads and ad sets that show terrible results. As for the remaining ad sets, we keep them running for several days even if their performance seems to be not so good. Such campaigns (which we use for testing Facebook interests) often take up to 7 days before you get reliable results. Of course, the team monitors their performance.
A few tips related to checking your Facebook ads
Ideally, you should test your ad sets for at least 3 days. Making conclusions based on shorter periods is often premature because sales always come in peaks. Sometimes you’re simply having bad days.
If an ad set behaves clearly worse than others during all this period, there’s no doubt you should either turn it off or experiment with ads within the set. Sometimes one and the same ad works great with one audience but doesn’t work at all with another one.
Audiences and ads also tend to “fade away”, become less effective as time passes. After a while, they simply stop converting users as well as they used to. In this case, the team changes the ads’ texts and/or featured images/thumbnails. Ideally, you should edit the existing creative or make a completely new ad.
Step 6: Collecting demographic data
The conversion campaign keeps running. At this stage, we leave it alone for a while. Two or three days after we launched our PPE campaign, it’s time to check the demographic data both our campaigns have collected so far.
To see these data, we break-down the audiences of all the existing campaigns including PPE into age and gender segments.
The team uses the new data to optimize the targeting settings of the existing campaigns – both the PPE and conversion campaigns. We simply duplicate the existing ad sets and use the new targeting settings.
Sometimes, when the initial targeting happens to be completely wrong, we turn off the original ad sets. If not, they keep running along with the duplicated versions.
Step 7: Creating a lookalike audience
Now it’s time to get back to our conversion campaign that tests Facebook interests. In order to make money with Facebook ads, we are going to create custom audiences based on these results and use them to generate lookalikes.
First, the team creates a custom audience consisting of users who watched 95% of the video ads for the last 7 days. These data come from all the videos that advertise one particular product. To make this work, we need at least 2,000 users. Depending on your budget, 3-7 days is usually enough to collect this many.
If the video is very long (about 40 seconds or longer), it’s Ok to select users who watched at least 75% of the video. People usually don’t watch long files till the end.
Now that we have more than 2,000 users who seem interested in the product, we create a lookalike Facebook audience and break it down into six audience size segments: 1%, 1-2%, 2-3%, 3-4%, 4-5% and 5-6%.
Step 8: Launching a conversion campaign targeted at the lookalike audience
Now the team uses this lookalike audience to create a new conversion campaign. We create one ad set per each size segment I mentioned above. The budget is set to half of the product’s price per ad set just like previously.
The team places not one but three or even four best-performing video posts from the PPE campaign in each ad set. By this time, they usually gather quite a number of likes and comments.
The campaign usually runs for 2-3 days before we check the results and turn off the worst-performing ads. When the campaign gets at least 20 add-to-carts, checkouts or purchases, the team moves on to creating another range of custom and lookalike audiences.
Step 9: Creating new lookalike audiences
As our campaigns collect data, we can use it to create new lookalike audiences for another conversion campaign.
First, the team starts making custom audiences consisting of the following categories of users (these are the abbreviations the team uses to name these audience segments):
- 95% VV7 (95% video views, 7 days): users who watched 95% of our video posts over the last 7 days;
- 75% VV7 (75% video views, 7 days): users who watched 75% of our video posts over the last 7 days; used for long videos;
- WV 7 (website visitors, 7 days): users who visited the website over the last 7 days no matter how long they stayed;
- VC 7 (view content, 7 days): users who visited the website over the last 7 days; it’s different from the previous segment in that these users spent more time there;
- VTS 25/10/5 7 (visitors by time spent, top 25, 7 days): top 25/10/5 users who stayed on the site longer than anyone else over the last 7 days;
- ATC 7 (add to cart, 7 days): site visitors who added the product to the shopping cart over the last 7 days:
- IC 7 (initiate checkout, 7 days): site visitors who initiated the checkout process over the last 7 days;
- PUR 7 (purchase, 7 days): site visitors who purchase the product over the last 7 days.
To generate a really useful lookalike audience, Facebook needs examples. The larger your customer audience the better.
For lookalikes based on “video views”, it needs at least 2,000 users and at least 200 for “website visitors” and “view content”. “Add-to-carts”, “initiate checkout” and “purchases” require at least 20 users, but it’s best to have about 50. Otherwise, Facebook will have too little information for finding similar users.
The time period is also important. While it’s Ok to set it to 7 days for most of these audiences, ATC, IC and PUR require a different approach.
At first, we recommend creating these three audiences based on users who performed the corresponding actions over the last 30 days instead of 7. Only after they start generating at least 100 purchases, add-to-carts and checkouts within a 7 days period, we change the period to 7 days.
As the campaigns keep gathering more users, we create these audiences one by one.
Step 10: Creating value-based audiences
It’s also a good idea to experiment with value-based lookalike audiences to make money with Facebook ads. Let me explain what they are in simple words.
Facebook generates standard lookalikes based on a custom audience. It analyzes their behavior and interests to find people who act in a similar manner.
When Facebook generates value-based audiences, it also takes into account how much money these people tend to spend. In other words, it tries to find users who will most likely spend more money on your store.
Such audiences can be based on custom audiences that consist of users who viewed the site’s product pages, added something to the shopping cart, initiated checkouts and purchased something. Facebook algorithm analyzes the prices of the products they viewed, added to the cart, tried to purchase or purchased and looks for people who are ready to spend similar sums.
Now the team creates value-based lookalikes (based on the custom audiences I just mentioned) and breaks them down into size segments like we did with our previous lookalike audience.
Step 11: Launching and running a SANDBOX conversion campaign to make money with Facebook ads
The team launches another conversion campaign targeted at the audiences I have listed above. The team uses one campaign, but if you find it more comfortable, you can launch several campaigns.
We call this the SANDBOX campaign because it’s going to be our main testing arena.
Each ad set is targeted at one size segment of one of the listed lookalike audiences. The budget is set to half of the product’s price in the store but not less than $10 per each ad set. Also, the team excludes the custom audience users who have already purchased the product over the last 180 days from each ad set.
Next, the final testing begins. We give the ad sets at least two days before making any decisions.
Any ad sets in the SANDBOX campaign that perform badly can be turned off. If it performs well, the team duplicates the ad set into a new campaign and doubles the budget. If this new ad set performs well, we duplicate it and increase the budget again.
For example, if the budget for an ad set in the SANDBOX campaign was set to $15, we duplicate it into a new campaign with the budget set to $30. If everything is good, we make another copy with a budget of $60 and so on.
Why duplicate the ad sets each time they perform well instead of just raising their budgets? There’s an ideal budget for each ad set which generates the maximum ROAS. You simply need to find it.
But each time you change the budget, Facebook has to start over, so to speak. On the other hand, if you duplicate the ad set, your “previous” ad set keeps running. If the copy performs worse than it, you simply turn it off in one click.
How to make money with Facebook ads: The Auto Merch’s remarketing strategy
As you can see, the advertising strategy we use on The Auto Merch starts with targeting huge audiences of random people. It lets Facebook define which audiences convert better on its own, which is really useful for broad niches we use in our Premium Stores.
As we test different settings and ads, we collect more data and narrow down the audience. Facebook also manages to find people more interested in our products, thus improving the quality of its audiences. As a result, we target people who are more likely to buy.
The more money you spend on advertising, the faster you can collect the data and the more people will become actual buyers.
The warm part of this audience who didn’t purchase anything can be retargeted with remarketing campaigns. And that’s another way to make money with Facebook ads.
Remarketing is another important part of The Auto Merch’s promotion strategy. While the standard ads are targeted at cold audiences, unfamiliar with our store, remarketing targets people who have already interacted with the website or our ads. This makes remarketing much more effective.
Here is the strategy the team uses to retarget audiences.
Step 1: Creating custom audiences
The work usually begins 7 days after launching the initial PPE campaign. By this time, we should have enough users who have interacted with the ads or the website to start remarketing.
The team creates two custom audiences from all users who got into our previous campaigns.
The first audience are Facebook users who watched 95% of the video posts over the last 7 days. Remember: if the video is about 40 seconds long, it’s Ok to set the limit to 75%.
The second audience are users who visited the product page (“viewed content”).
Step 2: Launching a remarketing campaign and making exclusions
When the audiences are ready, the team makes another set of ads. For users who watched 75-95% of our previous video posts (“video views”), we can use the same videos. But we also create new ones designed specifically for retargeting. Sometimes the team also adds coupons or photo reviews from customers in remarketing ads.
In addition, our team creates image ads. We recommend adding the images in 3 formats – vertical, horizontal and square for each group of placements. We also make up to 5 text variations for each ad and let Facebook define which one performs better.
Then the team launches a conversion campaign and sets up exclusions.
For “video viewers”, we exclude the users who visited the product page (“viewed content”) and the ones who have already purchased the product. For “content viewers”, we exclude only the ones who have purchased the product.
In theory, you don’t have to exclude “buyers” from the first audience because people can’t buy anything unless they visit your store. Excluding “content views” should be enough. However, sometimes Facebook doesn’t mark buyers as site visitors, and they get into the audience. That’s why we recommend excluding buyers “manually” – just in case.
The team creates one ad set per each of the two audiences. As for the overall remarketing budget, we recommend setting it to 10-20% of the budget you spend on advertising. For example, if you spend $1,000 on advertising, set your remarketing budget to $100-200. Thus, the more you spend on advertising the more you should spend on remarketing.
If you’re not sure how to split the money between the two remarketing audiences, use Facebook’s Campaign Budget Optimization. This function sets the budget for the whole campaign and allows the algorithm to make the decisions based on how the ad sets perform.
If you want to do it manually, set the initial budget to $10-15 per each ad set. As the budget of other campaigns keeps growing, you should also spend more on remarketing.
Step 3: Start making money with Facebook remarketing ads and monitor the campaign’s performance
After the campaign is launched, the team monitors the “Frequency” metrics. It reflects how often a user sees your remarketing ads. It should be between 2.5 and 5 as long as it brings sales. We also monitor Click Through Rate and Cost Per Click.
If a creative or ad set performs well, the team raises its budget by 50% of the current value at a time until its revenue stops growing. If the ad set’s performance starts to get worse, we cut the budget.
In this case or if a creative or ad set performs not well enough from the very beginning, the team cuts the budgets by 33% of the current value at a time. If the performance doesn’t get better, we keep decreasing the budget until we have to turn off the ad set completely and try something new. The team usually shuts off the creatives with the lowest CTR and highest Cost Per Click.
Clicks here are more expensive compared to our previous campaigns, but don’t worry. Click Through Rates and Return On Advertising Spend (ROAS) in remarketing campaigns are always higher, so you can make money with Facebook ads despite the cost of remarketing.
Note that you shouldn’t change the ad sets’ budget more often than every two days as Facebook needs time to complete its learning phase and optimize its performance.
Since retargeting people who have interacted with your store is more efficient than targeting people unfamiliar with it, the remarketing campaign keeps running all the time. However, we still experiment with new creatives and audiences with different time periods.
For example, if one of your initial ad sets was targeted at people who watched 95% of the video ad over the last 7 days, you can create an ad set targeted at the same segment but collected over the last 14 days. We also recommend launching ad sets targeted at the same audience collected over the last 30 days as soon as enough time passes. Also, don’t forget to exclude all previous segments from new audiences.
The team always uses new ad creatives for these new ad sets.
This is the basic method used by our team for Facebook advertising. However, there are more nuances and secrets, which you can learn from AliDropship’s free bonus course available after purchasing any Premium Dropshipping Store as well as from upcoming blog articles and YouTube videos. So stay tuned!
Having these insights on how to make money with Facebook ads, you can easily start your own successful dropshipping business by purchasing a copy of any Premium Store and putting your marketing knowledge into practice!
Learn how to make money with Facebook ads.