A Crash Course on Incrementality
With Q4 around the corner, advertisers are looking to perform their final testing agendas to ensure a wonderful holiday season. Incrementality has become more and more important as costs have risen year over year. We thought it beneficial to provide a quick refresher on incrementality to help out with your holiday and testing agenda prep.
The goal of this article is to not provide an exhaustive study of conversion lift testing. Rather, what are several overlooked aspects of Facebook’s conversion lift testing that may provide an edge on your analysis.
What is Incrementality on Facebook?
Facebook leverages an Intent To Treat model for their *incrementality testing. After deciding on a control size, Facebook randomizes the entire platform to be either in the control cell or in the test cell.
Conversion lift testing is then taking the total sum of activity that occurred in the test cell and subtracting everything that happened in the scaled control from the test. The only difference between the cells is the spend that occurred on Facebook. Thus, this subtraction shows the incremental impact of Facebook on your business.
Scaled control is an important term as it allows an advertiser to not run tests with a 50% hold out. For example, if an advertiser ran a lift study with a 20% control it would sum all the activity in the control and multiply by 4 to compute the scaled control.
If 10 conversions happened in the control, then the scaled control would have 40.
Why multiply by 4? Because the test cell had 80% of the population and multiplying 20% by 4 equates to 80%.
If we assume that the test cell above had 100 conversions then this study produced 60 incremental conversions.
What Does Opportunity Mean?
Technically speaking, conversion lift testing on Facebook is not an A/B test. (Colloquially people tend to use them interchangeably.) With an A/B test, an experimenter takes the results from everyone who was exposed to a stimulus in cell A and compares it to cell B. (Cell B could be a control or have a separate stimulus.)
For conversion lift studies, it is the summation of all activity which occurs in the test cell. Everyone in the test cell has an opportunity to see an advertisement, but not everyone does.
As exposure is not 100% in the test cell, lift studies are by their nature a conservative measurement tool. Furthermore, incrementality on Facebook does not have to look positive for the platform. We have seen many studies that looked less than favorable for the platform which is one of the best reasons to trust it as a tool.
When choosing Basic Results, Facebook is simply reporting the pixel fires tied to people in the test cell versus those that occurred in the control cell. If you have verified the pixel, then the lift study will be accurate. In short, the veracity of the lift study is controlled not by Facebook but by the advertiser.
Wait, What Are We Studying Again?
One of many pre-test checks we perform at dysrupt is confirming that the question being asked is the one the advertiser wants answered. Whether the holdout is at an account level or campaign level, which accounts are included in the test, etc. all impact the question that the test is asking.
Let us suppose that an advertiser has two ad accounts, A and B. If they place both accounts in the test, then the question they are asking is “What is the incremental impact of Facebook advertising on my business above all other marketing efforts?”
If the holdout is only placed on account A and account B is not included in the study then the question changes. It becomes “What is the incremental impact of activity from Account A on my business versus all other marketing efforts including ads on Facebook from account B?”
We can suspect that since the control is now exposed to ads from account B, the second version of the study would have a lowered version of incremental impact than the earlier, universal hold out question.
Now What?
We’ll write about how we analyze studies at a later date. After running 257 lift studies for over 30 companies, there are lots of overlooked aspects in the conversion lift UI. One of the largest mistakes we have written about in a previous article. Don’t worry...there are many other nuances to consider.
We hope this article provides a quick primer of salient features around conversion lift testing.
And for the non-analytical here is a brief summary of the points.
- Incrementality is different than an A/B Test
- Take time to define the question being asked before testing
- 50% hold outs are not mandatory due to the Scaled Control
- Incrementality is based on a person’s opportunity to see an ad
- Lift studies are as accurate as your pixel fires
Reach out to info@dysrupt.com if you ever want to talk measurement around your specific business needs!
* For this article, we will interchange lift, conversion lift and incrementality testing throughout to mean the same thing.
---
Thanks to Prateek Katyal for sharing their work on Unsplash.
Read next
Facebook Ads Bidding Strategies can either make or break your advertising campaign. If you've been struggling with getting the best results, understanding the benefits and drawbacks of each strategy can save you both time and money. The right bidding strategy can help you reach your target audience more effectively and get the most out of your advertising budget.
Each bidding strategy has its unique benefits and challenges. Some are great for maximizing visibility, while others prioritize cost-efficiency. Choosing the right one depends on your specific goals, whether that's more clicks, better engagement, or higher sales. Knowing the pros and cons of each strategy will help you make informed choices that benefit your business.
This article will guide you through five key Facebook Ads Bidding Strategies. You’ll learn about their benefits, drawbacks, and how to pick the one that suits your campaign objectives. By the end, you’ll have a clear understanding of which strategy will help you achieve your advertising goals effectively.
Understanding Facebook Bidding Mechanics
Facebook bidding is essential for advertising success. It involves auctions where advertisers compete for ad placements. Understanding key elements like Auction Dynamics and Different Bidding Strategies is crucial.
Auction Dynamics and How Bids Work
In Facebook's auction, ads compete based on bids, estimated action rates, and ad quality. Bid represents how much you're willing to pay for a specific action (like clicks, views, or conversions). The Cost Per Result adjusts based on competition.
Bid Cap lets advertisers set a maximum bid. This ensures spending control but may limit campaign reach. Meta bidding strategies, like Lowest Cost and Target Cost, help optimize for specific goals, balancing cost and performance.
Factors influencing the auction include:
- Bid amount
- Ad relevance
- Estimated action rates
Exploring Different Bidding Strategies
Advertisers can choose from several Facebook bidding strategies. The Lowest Cost strategy aims to get the most results for the lowest price but may lack spending control. The Cost Cap strategy helps maintain an average cost while driving results.
The Bid Cap strategy is useful for high-control needs, letting you set the max bid per action but it might restrict delivery. Target Cost aims for a stable cost per action, ideal for steady budget planning.
Choosing the right strategy depends on your campaign goals, budget, and desired Cost Per Result. Evaluate each option to find the best fit for your needs.
Implementing Bidding Strategies for Campaign Success
Successful implementation of bidding strategies can drive better results and optimize ad spend. Key factors include setting appropriate bid caps, maximizing returns using ROAS goals, and balancing volume and value.
Setting the Right Bid Cap for Your Campaign
Setting the right bid cap involves determining the maximum amount you are willing to pay for a result. This ensures costs don't exceed the budget. Bid caps can help control spending and improve efficiency.
- Analyze past performance: Review historical data to identify the highest bid that achieved desired results.
- Adjust as needed: Be flexible to change bid caps based on real-time campaign performance.
- Consider the competition: Higher bid caps might be necessary in competitive markets.
Maximizing Returns with ROAS Goals
Use the Return on Ad Spend (ROAS) bid strategy to drive maximum returns. ROAS goals ensure that every dollar spent on ads generates a specific amount of revenue.
- Calculate target ROAS: Set a realistic ROAS based on past campaigns.
- Monitor and tweak: Regularly check ad performance and adjust your ROAS goals to meet revenue targets.
- Balance quality and cost: High ROAS might limit reach, so find a balance between cost and quality.
Balancing Volume and Value in Bidding
Balancing volume and value helps achieve the right mix of reach and profitability. Consider using both Highest Volume and Highest Value strategies.
- Highest Volume: Bids are set to get the most conversions, good for awareness and large-scale campaigns.
- Highest Value: Focuses on getting the highest-value conversions, suitable for targeting high-value customers.
By carefully implementing these strategies, advertisers can meet their campaign goals effectively.
5 Facebook Bidding Strategies: Benefits, Drawbacks, and Selection Guide
Static ads and dynamic ads serve different purposes in the world of marketing. Static ads are simple and stay the same at all times. They are easy to create and can be effective for straightforward messaging. But dynamic ads offer customization, changing their content to fit the audience's preferences and behaviors.
Dynamic ads might seem complicated, but they bring better results by targeting specific groups with personalized messages. This means higher engagement rates and more conversions. Static ads, on the other hand, are less effort to produce but may not capture attention as effectively.
Deciding between static and dynamic ads depends on the brand's goals and resources. Each has its strengths and can be powerful if used appropriately in a marketing strategy.
Understanding Static and Dynamic Ads
Static ads and dynamic ads serve different purposes in digital marketing. Each has unique features and benefits that cater to varied marketing needs.
Exploring Static Image Ads
Static image ads are straightforward. They are typically still images that do not change once created. These ads are ideal for conveying a clear, unchanging message or brand image.
A static image can include text, graphics, and logos, and is often used on websites and social media platforms.
Advantages of Static Images
- Consistency: The message remains the same, which can be useful for brand recognition.
- Simplicity: They are simple to create and often cost less than dynamic ads.
- Predictability: Once the ad goes live, what you see is what you get.
Unpacking Dynamic Advertising
Dynamic ads are more complex. They can change content in real-time based on user data and behavior. Unlike static ads, dynamic ads can alter images, text, and calls to action depending on who is viewing the ad.
Benefits of Dynamic Ads
- Personalization: Content can be tailored to each user, potentially increasing engagement.
- Flexibility: They can show different messages to different audiences without creating multiple ads.
- Efficiency: They adapt to user preferences, making the ad experience more relevant.
Comparative Analysis and Use Cases
Static and dynamic ads offer different benefits and limitations. This comparison will help you understand where and how to use each type effectively in your marketing strategy.
Static Images Vs. Videos
Static images are simple and quick to create. They load faster than videos, which is great for mobile users and slow internet connections. They allow for clear, focused messages without distractions.
Videos, on the other hand, capture attention better with motion and sound. They convey more information in a short time. Videos are more engaging and can demonstrate products or services in action.
Leveraging Opportunities for Static Ads
Static ads are useful in various scenarios. Billboards are a great example, as they need to be read quickly. Print ads in magazines and newspapers also benefit from static images. Online banners are often more effective when static, as they load quickly and are less intrusive.
Static ads are best when the message is straightforward. They work well for short calls to action like "Buy Now" or "Sign Up." Visually, they should be clean and uncluttered to convey the message quickly.
