For many advertisers, the most important thing in the operation of advertising campaigns is the effectiveness and conversion rate. Today we will not talk about types and about the process of setting up conversions. Instead, we will compare and analyze the relationship between Facebook attribution windows.
We have been saying for a long time that it is wrong to analyze campaigns only by clicks. Especially if the campaigns are media. For example, you are the owner of a business with a long product turnover life cycle: selling cars, real estate, equipment, etc. You don’t always get a conversion from the first day in such topics. Therefore, deferred demand should not be overlooked in data analysis. Speaking in digital metrics, these are post-view and cross-device data. This is an important aspect in the analysis of advertising campaigns, and without it, you will draw the wrong conclusions about the results of a media campaign.
Read more about this in the article ”Comprehensive Analysis of Display Advertising”.
What is post-view on Facebook and how to work with it?
The attribution window in Facebook is the period of accounting for conversions and client clicks from the moment the user contacts the ad. The standard attribution window (“last click” model) takes into account the conversion if it is made on the site after 1 day of viewing and 28 days after the click.
Optionally, columns for various attribution windows can be included in the performance report.
Facebook post-view key metrics
View-through conversions (days after viewing) – the user saw the creative, did not click on the link (ad). After some time, he visited the site and made a conversion.
Click-through conversions (days after click) – the user clicked on the link and made a conversion within a certain number of days.
Here it is important to remember and understand that conversion data may overlap with Google Analytics. If a person saw the banner, went to the site, and made a conversion, then Google Analytics (according to the standard “last indirect click” attribution model) will count this conversion as the source “Facebook | CPC”.
But if a person went to the site, left, and a day later returned to the source of search advertising, then this conversion will go to the google / CPC source, and the conversion from Facebook can be seen in the associated ones.
Important Point: we recommend you check the integrity of the data. This can be done by expanding all attribution windows into one table. The data in the “1 day after view” and “28 days after click” columns should equal the standard attribution window. It should be remembered that when we launch several campaigns (even if they have different interests), there is always a moment of audience intersection.
Note that on Facebook, view-through conversions also include cross-device data. They can be seen if you make a special report on advertising and set 2 indicators in the “breakdown” column: “impression device” and “conversion device”.
The impact of data on the evaluation of the effectiveness of an advertising campaign
Let’s now analyze this data in a real case and see how it affects campaign analysis.
Category “Lending”
Campaign dates: 01/15/2020 – 01/27/2020.
The case data is sampled to maintain the confidentiality of information. At the same time, they allow you to see how the cost per conversion varies depending on the standard combinations of the attribution model.
Conversion – Thank You Page
- There were 45 conversions in the 1-1 model, among them post-view – 33%, and post-click – 67%.
- In model 7-7, there were 121 conversions, among them post-view – 64%, and post-click – 36%.
- In model 28-28, there were 231 conversions, among them post-view – 72%, and post-click – 28%.
We can conclude that there is a significant increase in post-view conversions from the first attribution window to the second. Such growth can be in the range of 50-200%. It all depends on the product and topic. If we analyze Facebook ads only in the standard attribution model, then we will not see the whole picture of how the advertising campaign worked. At the same time, the share of such conversions is critically large.
If we walk through different levels of the campaign, we can analyze the effectiveness of individual groups and creatives. This will make it possible to evaluate which creatives generated more post-view conversions.
- The conversion cost from model 1-1 (1 day after click – 1 day after view) to 7-7 dropped by 63%.
- The price from the 7-7 to 28-28 conversion model fell by 48%.
- You can analyze the share and increase in conversions in each of the campaigns depending on placements (types of targeting).
- The conversion cost from the standard attribution window to the widest attribution window drops by $6.42.
Key takeaway: regardless of the product and your theme, there will be a pent-up demand effect. The main thing is to understand how big the demand is and whether you can estimate it. We can only analyze pure conversions if we look at 1-1 attribution without including data from other models.
Attribution models can be combined, for example, 7 after click and 28 after view. So we can find out the number of conversions, which is auxiliary. In analytics, as a rule, they are also called assisted conversions.
If the issue of creative for advertising is relevant for you, read our article “How do you know which creative works best? Test.”
Influence on the evaluation of the effectiveness of targeting
Let’s take a closer look at targeting performance.
For example, the “Look-alike 4%” campaign, which was aimed at a very cold audience, gave few conversions in the standard attribution window, but after a while, this campaign caught up with the main ones, as the “deferred demand” effect worked.
The look-alike 4% 2.0 campaign, as you can see from the statistics, gave only one conversion for $37. At the same time, having received the post-view conversion data, the conversion cost fell to $4.12.
This example is shown on a relatively small number of targetings. In our experience, the broader your media plan, the more valuable insights you can draw on audiences.
In addition, a very important nuance: some targetings work well for micro-conversions for a long time, while others work well for sales. Depending on your task for an advertising campaign, you need to select and analyze different actions for each stage of the funnel
We conclude: in most cases, the final effectiveness of all targetings is visible only when evaluating the full range of data. If you’re only evaluating a click, you’re more likely to make a survivor’s mistake.
How else to use post data for your campaigns
Determine the optimal frequency
In marketing books, there is a well-known “7 touches” rule, which states that a customer needs to be reminded of a product at least seven times through different sources to make a conversion or increase brand loyalty.
For each product and advertising format, this figure is different and our holistic analysis of media advertising, we can say how many touches are needed through different sources for the user to convert into a certain period
In the case of analysis outside of Facebook, where it is possible to use Google Campaign Manager, we can answer at which specific frequency the cost of contact or conversion is the most effective.
If we are talking about Facebook, then here the analysis by frequency is possible in a somewhat truncated form with the help of frequency analysis by targeting.
This analysis can also help answer the following questions.
- How long does the user remember your message and how often in time do you need to communicate with him?
- What kind of creative works for a long time and is most effective?
- What are the conversion chains by source?
Choose an attribution window for your business
We have conducted hundreds of campaigns in different directions and, frankly, there is no golden rule, there are only basic principles depending on the type of business and product, as well as historical data with the correct analysis on which to conclude. We will share with you our best practices for determining the attribution window for a product.
- In the case of a new task, the client, as a rule, assumes the life cycle of his product and the time for making a purchase decision. This allows us to understand in dialogue how to optimize advertising in the first place.
- After the first advertising campaign using post-view data, cut across all possible periods. Calculate at what point the share of user return growth starts to decline.
- With campaign analysis in hand, consider what percentage is “quick demand” and how much goes to other conversions. After analyzing 2-3 campaigns, you will have a complete picture.
In product topics (jewelry, cars, equipment) they have a large share of post-view conversion. We recommend using a wide conversion window of 14-28 days.
Please note that Facebook data allows you to view view-through conversions up to 28 days from ad contact. And Google Campaign Manager allows you to set this indicator for 28, 30, 90 … days and track conversion data in more detail.
How to Do Post-campaign Analysis
Don’t forget to conduct post-campaign analysis for your advertising campaigns! It’s a critical step in evaluating the success of your campaign and identifying areas for improvement. By analyzing campaign metrics and data, you can gain valuable insights and optimize your advertising strategies for better results. Make sure to prioritize post-campaign analysis to get the most out of your advertising efforts.
Post-campaign analysis is a crucial process in marketing, as it allows businesses to evaluate the success of their campaigns and identify areas for improvement. The post-campaign analysis is an evaluation of the campaign’s effectiveness after the campaign has ended.
The post-campaign analysis definition involves analyzing the performance of the campaign in terms of the objectives set before the campaign. The analysis should include an evaluation of the campaign’s success in achieving its goals, as well as an assessment of the marketing channels used, target audience, and messaging.
The post-campaign analysis marketing report should provide a comprehensive overview of the campaign’s performance, including reach, engagement, conversions, and ROI. The report should also include insights into the effectiveness of individual marketing channels and messaging, as well as recommendations for future campaigns.
To conduct a successful post-campaign evaluation, it’s essential to have a clear post-campaign analysis format in place. This format should include a detailed description of the campaign, its goals, and its target audience, as well as an analysis of the metrics and data collected during the campaign. The format should also include a section for insights and recommendations, outlining areas for improvement and opportunities for future campaigns.
In conclusion, post-campaign analysis is an essential process in marketing, providing valuable insights into the effectiveness of campaigns and opportunities for future improvements. By following a clear post-campaign analysis format, businesses can ensure that they are effectively evaluating the success of their campaigns and optimizing their marketing strategies for future success.
Conclusions
- Most importantly: consider post-view data in your campaigns, Facebook has 2 metrics to measure conversions – post-view (view attribution) and post-click (click attribution). Each of the metrics can be viewed in the context of 1, 7, and 28 days. And get numerous variations for the analysis of advertising campaigns.
- As a rule, view-through conversions make up a large share of advertising campaigns and grow from one attribution model to another by 50-200%.
- Don’t do campaign analysis right away. Wait for the conversion count. If the campaign ended on the 20th, and we are doing a performance report on the 21st, this is not relevant!
- Always analyze click and impression attribution data depending on your targeting. Keep in mind that cold audiences have few conversions in the 1-28 window, but have a larger increase in view-through conversions. Hot is the opposite.
- As a rule, everyone analyzes campaigns by the standard window, forgetting about impression data. Today we figured out the theory and showed in practice why it is important to take into account the effectiveness of this data.
