Have you found the perfect Facebook Ads Strategy?
The Internet is flooded with tons of articles describing a proven method to achieve results with Facebook Ads. Unfortunately, most of those articles are only created for SEO purposes. If you know anything about Facebook ads, you know that there is no magic pill. Everyone wants to show off, so they create their own “tiger method”, which usually merely consists of obvious tactics.
We tested most “tiger methods” 🙂 They don’t work. However, there are specific facts you need to know. When you understand the fundamental principles, you can skyrocket almost any Facebook ads account. Tactics always change but the principles remain the same, independent of how big the account is and who manages it.
Fact number 1. You need at least 50 conversions after the last significant edit to complete the learning phase.
Unless you are using Facebook for branding purposes, I assume that you are already using Website Conversion campaigns to drive sales. In the Website Conversion campaign Facebook, machine learning helps to find new customers based on your previous sales data.
To use this campaign, you need to set up conversion tracking using Facebook pixel. Depending on the website, this can take you from 5 min to several hours. Feel free to ask for our advice if you face problems.
What is considered a “significant edit”? To avoid copy-pasting, let me just share a Facebook article on this point.
Fact number 2. Engagement and negative feedback matters.
This may sound obvious, but this is one of the fundamental principles you should know and remember.
You need to know that ads that receive excessive negative feedback may be removed by the Facebook machine learning algorithm. Excessive negative feedback, or poor performance, are often signs that an ad is poorly targeted, contains suggestive images and language, or is related to sensitive topics.
Based on our experience, a successful proportion is 1 negative feedback per 100 positive entries. This formula allows you to keep your ads performance at a high level.
You can check the engagement in the Page posts section.
Fact number 3. Conversions that happened outside of the defined conversion window are not used by Facebook to optimize your ads.
There are two types of conversion windows: duration (for example, 1 or 7 days) and action type (for example, view or click).
By default, Facebook sets the maximum window of “7 days click or 1-day view”.
The advantage is that you will get more events within this ad set (but remember fact number one – you need at least 50 conversions after the last significant edit to complete the learning phase).
The disadvantage is that the system can credit ads that are not really performing well. Let’s say a user viewed your ad on Facebook, but didn’t engage with it because the ad was not effective; however, then the same user converted from Google Organic. If you would use “7 days click or 1-day view”, the system will credit the Facebook ad.
So, should ads be optimized for views or clicks? This needs to be tested in each account. If you don’t have enough conversion events in an ad set to use the click attribution, optimize per micro-conversions or use “7 days click or 1-day view”.
Set the right conversion window.
If your conversion window is a 7-day click, the ~50 conversions you need all have to happen within 7 days of the click that led to them.
Fact number 4. The relevance score doesn’t affect your bid in the auction.
The relevance score is a rating from 1 to 10 that estimates how good your target audience responds to your creatives. This score starts to work after you received more than 500 impressions on your ad.
We commonly use this metric to understand how relevant ads are from this specific ad group and for the target audience. If most ads in the ad set are 1-3, we can see that this might be not the right targeting or that the ads might be wrong.
However, we don’t pause ads with poor relevance scores if they are profitable. This is merely a signal for us that this specific ad can perform much better if we customize it and initiate the next level.
Fact number 5. It takes time until you can show the same ad to the same user again.
In the Facebook News Feed, a person won’t see ads from a single Page or app more than once every 2 hours.
In an Instagram feed, a person won’t see an ad from the same advertiser (whether the ad was initially seen in Instagram feed or in Stories) more than once every 3 hours.
In Instagram Stories, a person won’t see an ad from the same advertiser (whether the ad was initially seen in Instagram feed or in Stories) more than once every 6 hours.
By considering this information, you will understand how many creatives you should have in each ad set. Keep at least two creatives in each ad set.
Fact number 6. Your bidding strategy determines whether you will completely spend your budget.
In the Website Conversion campaign, there are three bidding strategies:
Cost control is the desired maximum CPA. This helps you to control your cost per conversion.
- When you enter your average cost per lead, Facebook will automatically use the “Cost cap” bidding strategy. When you are using the cost cap bid strategy, the system tries to obtain the optimization events with the lowest cost first. As a result, you participate in fewer auctions.
- Bid cap. This strategy focuses on the conversion volume. The bid controls the limit we can bid on every auction. As a result, this bid strategy is less flexible, If you set a bid control of $1 and 50 optimization events to become available for $1.01 bids, Facebook will not try to get any of these (but they would target all of them if your cost control was $1).
- Target cost. With this strategy, Facebook will aim for the average cost to stay within a 10% range of the cost control you entered.
By the way, you can run ads without any cost controls. Especially if the volume is more important than the cost per conversion.
Fact number 7. Audience overlap may prevent some ad sets from spending their full budgets.
In fact, this can also cause other problems. For example, you would not be able to really test different audiences. When you have an audience overlap, your different ad sets compete between each other and Facebook will choose the audience with the best performance history.
To check the overlap, you can go to “Audiences”, check the boxes next to the audiences you want to compare, and click Actions > Show Audience Overlap.
Add relevant exclusions based on these insights. For example, if you are testing a lookalike, you should exclude it from other ad sets.
So, now that you are aware of this short 7-steps strategy on how to build an effective Facebook Ads Strategy, just try it 🙂