When it comes to online advertising, Facebook would be one of the first places you might think about.
Facebook is the largest social media network with more than two billion monthly active users and it continues to grow every day.
But with all of Facebook’s smart features and huge user base, it can spend all your budgets without any significant results for you.
In this guide, you’ll learn exactly how to launch, track, optimize, and analyze your Facebook advertising campaigns, including:
- Step #1: Set up the Facebook pixel
- Step #2: Create a right campaign structure
- Step #3: Campaign optimization. Chose the Conversion Event
- Step #4: Chose the type of budget management: CBO or ABO
- Step #5: Audience Testing
- Step #6: Chose right placements
- Step #7: Create Ad/Creative
- Step #8: Scaling
- Step #9: Retargeting
Step #1: Facebook’s Pixel. Keep Track of Users on your website
The Facebook pixel is a piece of code you place on the backend of your website to track visitors to your site. The pixel allows you to run highly targeted campaigns, so it’s important to install it before running Facebook ads.
Each ad account gets one default pixel to use. The code is made up of two main parts: the pixel base code and event code. The pixel base code tracks all of the traffic to your site. Event codes are additional pieces of code you can add under the default pixel code to specific pages of your website that allow you to track certain actions on those pages.
Install the base pixel code on your website and Facebook will begin tracking immediately.
Read more about Best Practices for Facebook Pixel Setup
Step #2: Create a right campaign structure
Ads you create in Facebook Ads Manager have three levels (parts):
- campaign level
- ad sets level
- individual ads creatives
Basically, this is called the campaign structure.
Knowing how the parts work together will help you run your ads the way you want to and get them to the right people.
Start at least with 5 ad sets with different audiences to test
Step #3: Chose the Campaign Objective and Conversion Event.
Optimize for the pixel event or objective goal you want to achieve.
If you optimize your campaign for conversions – this might be:
- Complete Registrations
- View Content
- Add to cart
- Add to wishlist
Step #4: CBO or Ad Set Budgets? What’s Best?
So how do we go about testing with CBO?
Below is an easy print and reference cheat-sheet followed by the step by step guide.
Let’s dive in!
If you’ve not been through the introduction to CBO, there are a few things you need to recap on:
- CBO is a learning system and is different from ABO (Ad set budget optimization) – the decisions made on which audiences and ads focus on adapting during the day.
- If you’re constantly using a single CBO for testing and scaling, pausing out ad sets and ads as your testing cycles end, this causes issues with the ML (machine learning) – the clever nerdy stuff that Facebook implement to help you achieve your goals.
- CBO gets better over time, with stability – meaning the fewer edits you make on ad sets and ads, the better the CBO learning algorithm becomes. Stability on average should be there within 5-10 days but it highly depends on the various factors I cover below.
- CBO does not completely, yet, replace the need for ad and ad set level optimization to pause/replace bad performance.
- However, CBO used right gives better stability and reduces the need for constant tweaking and analysis freeing you up to focus more on marketing, offer creation, and your funnel.
- CBO gives faster results with qualified ad sets and ads – which means if they are pre-tested as ‘working’ then you’ll get stability and strong results faster.
NOTE: Campaigns with smaller budgets (below $200 daily) with Ad Set Budget Optimization (ABO) provide better results faster and are more stable
Depending on the amount of time it typically takes for someone to complete a valuable action (conversion) after clicking or viewing your ad chose your conversion window.
I would start with 1-day click, which means that Facebook will optimize ad delivery by using data from conversions that happen within 1 day of someone interacting with your ad.
Step #5: Audience Testing
The purpose of audience testing is to find the winning ones.
NOTE: it is common to make a minimum difference per ad set for accurate A/B testing, for instance, set 1-2 of the below options:
- Physical exercise
- Master’s degree
- New job
- New relationship
- Newly engaged (3 months)
- Engaged Shoppers
- Frequent Travelers
- Facebook Payments users (30 days)
- Facebook Payments users (higher than average spend)
- Different Lookalikes (LAAs):
- LAA of users who visited the website in the past 30 days
- LAA of users who visited a specific page of our website
- LAA of leads
- LAA of purchases
NOTE: You may use excludes based on people that have engaged with/clicked on your ads so that you’re targeting clean, COLD audiences.
To create an exclusion audience go to Audience Create a Custom Audience
Minimum daily budget for each ad set
If your CBO campaign contains 5 ad sets with a minimum daily budget of $100 during the testing phase, set a MINIMUM budget for each ad set of at least $20 per ad set.
This is needed to spend fair across the ad sets in the test phase.
The screen below shows at the ad set level, where to find ad set spend limits (expanded under the Optimization and Spending Controls).
Step #6: Placements
I would usually start with Mobile-only devices and Facebook News Feed placement. In another ad set, I would test Desktop devices only.
I prefer to test in the newsfeed first before expanding to other placements (I prefer FB & IG in prospecting and All placements further down the funnel) but this will depend on your audience and how good your creative is set up to work with those placements.
I don’t include Audience Network or Messenger placements for testing because, in practice, these skew results, often quite badly in cases where Facebook may overspend in placements like Audience Network with very bad performance.
Step #7: Ad/Creative
Create ads that trigger the only 3 things you need a user to do: stop and take notice, engage, and take action.
Don’t over complicate your ads.
Plan of creatives to test – ideally to test 3 angles (or pain points – a specific problem that prospective customers are experiencing), each with at least one ad copy and multiple visuals.
I would usually aim to have 3 different ads with different formats:
However to start my testing, I’d use 3 different ads in my ad set, and use those SAME ads in each ad set.
I’m looking at achieving my pixel goal: add to cart, form registration within around 2K-3K impressions.
If ads show bad performance you have a few decisions to make:
- Is it the ad that’s bad? If you run 3 distinct ads or duplicate the same ad 3 times in an ad set, you can potentially tell if its the ad or the audience: if 1 or 2 ads work well and the third doesn’t – then it is likely the ad
- If ALL ads equally performed badly in an ad set, it could be:
- that the audience is not a good one
- or it could be that while the audience is good, Facebook delivers you poor traffic. In this case, you may give the ads another go – often by duplicating the ads and testing them again which drops them in a different audience allocation with a clear past history
If the ad testing has successfully created new winning ads but you don’t need them in your main prospecting campaigns yet, copy them into a social proofing campaign (via Post ID sharing) – specifically created as a keep-warm Page Post Engagement (PPE) campaign, with a single broad audience.
Page Post Engagement (PPE) campaign creation
- Create a new PPE Campaign.
- Use the largest related audience you can for example a broad or large Interest or LAA – one ad set only required because we’re only after social proof – expect some sales to trickle in too!
- Exclude site visitors and purchasers.
- The budget of $5 min. multiplied by the number of ads, with all active ads accumulating social proof/running.
- Copy ads from phase 1 or 2 testings in via their post ID (to share social proof)
- When you need them from scaling, copy them, again via post ID into the scale campaigns. Leave the social proof campaign running!
Step #8: Scaling
After you start seeing the results of your campaign, you may want to increase your ad spend so your profits will continue to grow. Basically, this is called the scaling.
There are 2 most popular techniques to scale your Facebook ads:
- Horizontal scaling techniques:
- duplicating the winning ad sets
- creating and adding new lookalike audiences based on Pixel results
- expanding the target geo
- Vertical scaling:
- increasing the budgets
- Increase ad set budgets or CBO by 20% every 1-2 days if your campaign is performing consistently well for the last 3-4 days
- increasing the budgets
Step #9: Retargeting
An advertising theory says that a consumer has to be exposed to an ad at least three times within a purchasing cycle (time between two consecutive purchases) to buy your product or service.
Does Retargeting Actually Work?
You might be questioning whether this is just another marketing tactic that promises the world, only for you to still struggle to prove ROI once the campaign comes to a close.
The truth is: Facebook retargeting does work and the statistics prove it:
- Website visitors who are retargeted with display ads are more likely to convert by 70%
- The click-through rate (CTR) of a retargeted ad is 10x higher than the CTR of a typical display ad
- 44% of consumers say that they would likely become repeat buyers after a personalized shopping experience with a particular company
When we’re talking about Retargeting it the context of a profitable Facebook ad campaign we mean first of a sales funnel with several levels from top to bottom.
When to set a retargeting campaign?
- For warmed audience:
- User clicked on the ad, visited your landing page but didn’t convert
- People who have watched at 50-75% of your video
- Viewed or added to cart but not purchased
- For existing customers
How to set a retargeting campaign?
A retargeting campaign starts with audience creation
A source for your audience might be:
- Email list of existing customers which you shall load to the Facebook Customer list;
- Pixel-based events:
- Specific page visits;
- Specific CTA-buttons clicks (View Content);
- Signing up (Complete registration);
To create your retargeting audience for specific page visits:
1. Go to Audience Create a Custom Audience
2. Create a Website Custom Audience Chose the right Pixel Chose the option: People who visited specific web pages Indicate specific URL of your Offer Page Name your audience
3. Create a Website Custom Audience Chose the right Pixel Chose the option: People who visited specific web pages Indicate specific URL of your Thank you page
after the purchase has been done Name your audience
3. Create your retargeting campaign INCLUDE Offer Page Visits Custom Audience EXCLUDE Thank you page visits
In this scenario, you’ll reach people who have visited your offer page in the past 7 days but didn’t purchase your product or service.
P.S. I’d sure love to hear what you think about this post! Just drop a line in the comment section below and share your thoughts or questions.