In today's highly competitive market environment, advertising attribution and advertising performance monitoring have become crucial links in advertising delivery. With an in-depth understanding of advertising attribution and advertising performance monitoring methods, companies can better evaluate the effectiveness of advertising and make precise decisions to improve advertising effectiveness and maximize return on investment.
This article will take you to explore the principles and methods of advertising attribution and advertising effect monitoring, and learn to evaluate the effect of advertising in a more precise manner.
Ad Attribution
Advertising attribution is the basis of advertising effect evaluation. It refers to a set of logic and rules that identify which advertising or channel brings about the key behavior of advertising in the advertising behavior link.
For example, a new game has placed advertisements on Douyin, WeChat, and Weibo. User A saw the advertisements for this new game on these three platforms. Finally, user A downloaded the game and activated the installation. .
Question: Which platform is "attributable" to user A's transformation?
Advertising attribution actually needs to solve the problem of who belongs to the "credit" of the advertising effect. This is an integral part of performance advertising.
The significance and value of advertising attribution?
For advertisers: You can clearly understand the advertising effects of different platforms, and provide full reference for subsequent budget allocation, bid adjustment, material optimization, etc.
For the advertising platform: The attribution of "credit" directly determines the income of the platform; at the same time, it also helps the advertising operation to understand the delivery effect of different materials or copywriting, so as to adjust the delivery strategy.
In addition, attribution data can also provide an important reference for the iteration of the attribution model, and promote the "fair" and "reasonable" distribution of "credit" for advertising effects.
The method of advertising attribution?
To understand the method of advertising attribution, you must first know several elements of advertising attribution:
The "cause" of advertising attribution: that is, the user's touch behavior, generally speaking, click and play for X seconds (mostly for video ads).
The "fruit" of advertising attribution: cause and effect cause and effect, we need to clarify what the "credit" that needs to be attributed is. Generally, it is "transformation", such as downloading, installing, and paying.
Attribution window period: It can be simply understood as "the validity period of attribution". For example, if the attribution window period of an advertising platform is set to 7 days, the conversions generated on the 8th day after the user touches the ad will no longer be attributed to the ad.
All "attribution models (attribution rules)" are actually set based on the above three elements. The following are the more commonly used attribution models in the industry——
1. Single touch attribution model
①Last touch attribution
If a user touches the advertisement on multiple platforms (channels) A, B, C, D, and E and finally converts, according to the last touch attribution model, the conversion of this user will be completely attributed to the last touch platform (channel).
The last-touch attribution model is widely used. Graphical advertisements generally use "last-click attribution", and the attribution window is generally set to 3 days or 7 days;
Video ads (such as Douyin) generally use "last effective play attribution". Since "play" has a large reach and advertisers generally believe that play has a short effective impact on users, the attribution window period Generally shorter, such as 24 hours. Otherwise, it is easy to attribute the conversion of organic traffic to advertising traffic.
②First touch attribution
As opposed to last-touch attribution, the first-touch attribution model attributes conversions to the ad (channel) that the user first touched.
The first-touch attribution model is generally used for the first launch of a new product, which requires a large number of new customers to promote advertising. The disadvantages of first-touch attribution are also more obvious, because it completely ignores the impact of subsequent ad touch, so it is less used now.
③Channel Package Attribution
That is to say, an APP is divided into different channel packages for identification, and then distributed to different media (platforms). However, this method is generally only applicable to the Android ecosystem, and iOS does not support it for the time being. Channel package attribution is still relatively common in domestic game promotion.
Many people will also put "channel package attribution" in the category of "first contact attribution", because once a user downloads with this channel package, all subsequent behaviors, such as payment, will always be attributed to this channel.
So strictly speaking, channel pack attribution is not an attribution model, but a way of statistical data.
The channel package attribution also has its disadvantages. Although the user downloaded the game with channel package A, it does not mean that the user was not "influenced" by the advertisement of channel BCD before downloading.
The three attribution models listed above all have one thing in common, that is, the "credit" is 100% attributed to a certain channel (advertising/platform). We usually call this type of attribution model "single-touch attribution" Model".
2. Multi-Touch Attribution Model
The relative "multi-touch attribution model" adds the rules of "attribution weight", for example——
① Linear attribution
Under the linear attribution model, all channels of access have "credit", and the rain and dew are equal.
The principle of linear attribution is relatively equal to the distribution of credit, which is prone to cheating. After all, the advertising effects of different channels are still quite different.
② Attribution of time decay
That is, the closer the channel to the user conversion node, the greater the weight.
The time decay attribution model not only pays attention to the final conversion channel, but also pays attention to the impact value of other channels, which is a relatively fair attribution model.
③ U-shaped attribution (basic position attribution)
That is, the "first touch channel" and "last touch channel" are assigned larger weights, and the rest of the touch channels are assigned smaller weights.
The U-shaped attribution model pays more attention to "leads" and "final conversions", so the weight distribution also shows a U-shaped distribution.
Summary: Due to the complexity of the multi-touch attribution model, the single-touch attribution model is commonly used in China. No matter which advertising attribution model is used, it needs to be adjusted according to different advertising scenarios. It is not static. Different attribution models will also produce completely different effect analysis data.
Advertising Effect Monitoring
Advertising effect monitoring is the process of collecting, analyzing and evaluating advertising data to understand the effectiveness and effectiveness of advertising campaigns.
It is based on the following principles and methods:
Data collection : Advertising effect monitoring first needs to collect relevant data. This data can include ad impressions, clicks, conversion rates, sales, etc. Data can be obtained from various sources, including website analytics tools, advertising platforms, social media platforms, etc. Ad monitoring can also use specialized tracking codes or tags to collect data to ensure that every aspect of an ad campaign is accurately recorded.
Data Analysis : The collected data requires in-depth analysis. This includes counting and comparing various metrics and data points for ad serving. Data analytics can reveal campaign performance and trends, such as which ad channels or ad creatives are generating the best returns. In addition, data analysis can also discover potential problems or opportunities to help optimize advertising strategies.
Attribution Models : Ad performance measurement typically uses attribution models to determine the contribution of an ad campaign to a conversion or sale. The specific introduction is described in detail above.
A/B test : A/B test is a commonly used advertising effect monitoring method, which is used to compare two or more different versions of advertising ideas or delivery strategies. By randomly assigning different ad versions to your target audience and collecting data on them, you can determine which version of your ad performs better. A/B testing can help optimize advertising creative and delivery strategies, and improve the effectiveness and return rate of advertising.
Data visualization : Advertising effect monitoring results are usually presented through data visualization. Data visualization turns complex data into charts, graphs or reports that are easy to understand and analyze. Such visualization tools can help decision makers better understand the results of advertising and guide decisions based on data.
Through advertising effect monitoring, we can obtain real-time and accurate data on the effect of advertising, so as to make data-based decisions, optimize advertising strategies, and improve advertising effects and return rates.
The relationship between advertising effect monitoring and advertising attribution?
Simply put, advertising effect monitoring is the guarantee of the authenticity of effective advertising.
We can see the relationship between advertising attribution and monitoring through the figure below: advertising effect monitoring provides a means to collect and analyze advertising data, and provides a data basis for advertising attribution.
The importance of comprehensive advertising effect monitoring and advertising attribution can be said to jointly build a complete advertising effect evaluation and optimization system. Through advertising performance monitoring, companies can track advertising data and indicators in real time to understand the performance of advertising campaigns. Advertising attribution helps companies understand which factors in advertising have the greatest impact on conversions and sales, as well as the interaction between various advertising touchpoints.
With the continuous development of technology and data analysis, the methods and tools of advertising attribution and advertising effectiveness monitoring are also constantly improving. We should also actively adopt advanced attribution and detection technologies, constantly optimize advertising strategies, and keep pace with market trends.
References:
"[Computational Advertising] Talking about Advertising Attribution", Lazy Programming-22
"Two Attribution Models of Advertising Attribution and Explanation of Four Common Attribution Methods (Part 1)", PMCoder