“Performance Marketing 101” with adbrix – Part 2

By |2018-10-30T17:07:17+00:00September 19th, 2018|Working WITH adbrix|

Performance marketing, the second story.

‘Performance marketing’, which defines all marketing activities in quantifiable metrics and plan, execute, analyze, and optimize based upon the metrics is in the air. Let’s take a look at the courses of how to proceed performance marketing, by the Plan-Do-Check-Action (PDCA) cycle, which is the basic concept of the marketing cycle.

>> “Performance Marketing 101” with adbrix – Part 1 (Current Page)
>> “Performance Marketing 101” with adbrix – Part 2 (Go to Page)

Check – Find causal relationships in your data

Trend of membership conversion rate

Trend of membership conversion rate

Trend of new user and membership

Trend of new user and membership

The login conversion rate plummeted three days ago. Marketers looked back on the changes of the app and the marketing back then. They have pushed media for install volume to make up the decrease of new user acquisition trend, but that was a week ago. In addition, the decrease in conversions happened all across media and it was difficult to pinpoint a specific media for the low quality of the acquired users.

Looking at the app update history, they were able to confirm that a new version was released three days ago with bug fixes. They ran a test to check if an event tag for login conversion was missing when updating the app but the result had no such issue. Then what is the next step? They decided to look closely at the causal relationship between the app update and the conversion rate decrease, by checking out the data against the login conversion rate versus date, app version, and OS version.

Trend of conversion rate by app version

Trend of conversion rate by app version

Trend of conversion rate by OS version

Trend of conversion rate by OS version

It was spotted that the app update happened three days ago and the drop of conversion rate started on that day. By looking at the data, the marketing team could find out the conversion rate remained the same in mobile OS versions 4.xx and lower, while there was no single conversion in higher versions. They could conclude that there is a strong relationship between mobile OS version and membership conversion rate after the app update, and the finding was immediately delivered to the development team. The development team found that the updated HotelFirst v1.0.1 is going through a forced shutdown when logged in from mobile OS version 5.xx and above. The team quickly deployed a fixed version of the app and the conversion rate got back to normal.

An easy-breezy way of data analysis

Now you have an seen example of data analysis, but flexible handling of raw data as you’d like to is still difficult. Though you are versatile at using Excel’s pivot tables, a fundamental capability of a performance marketer, extracting large amount of data every time you need and pivoting and visualizing them is not an easy task at all.

With adbrix, you can access marketing performance data for cross-platforms including web as well as Android and iOS at a raw-level, and easily combine them with desired metrics, groups, and conditions.

The combined data can be visualized in any formats you need to see.

adbrix also supports a dashboard that enables you to save reports with frequently used combinations and to edit the arrangements and visual formats. If you are an advanced user, you can create an SQL query to directly extract data of your own combinations.

Action – Once you’ve got the result, make action right away

Marketing by audience group

Now that you have well gone through the “Check” stage, you are ready to divide users into two groups: a group with users who have converted to the preset marketing goal and the other users who have not. For non-converted user group, you may divide the users again into different cohorts depending on which of the stage toward the marketing goal they have paused. Marketing strategies need to be applied differently according to the characteristic of each cohort; while this type of cohort for marketing is usually called ‘audience group.’ Then, compose audience groups based on marketing KPI and implement appropriate marketing actions for each audience group to improve marketing performance.

Example for marketing by audience groups

Let’s start with the above example of HotelFirst app. Users who have already converted should be excluded from direct targeting campaign as purchase frequency of OTA services is usually not high. Since the users are valuable customer with history of purchase conversion, it’s recommended to keep in touch with them with newsletters and coupons and run a retargeting campaign for them when the next travel season comes or when average repurchase cycle comes. Of course the purchase frequency should vary by user level, so it is highly recommended to manage a audience group with higher purchase frequency with seperate strategy.

For the audience group with users who checked the price of hotel multiple times but have not checked out, you might want to send a push message with special offers or special events to stop them from leaving for your competitors.For the audience group with users who do not respond to any marketing or advertising campaigns and activities, it’s a good idea to exclude them from marketing targets for a while and investigate if they have something in common.

As seen above, marketing activities should start by setting up marketing audience groups based on the user behavior data, and the following activities should be planned, executed, and analyzed based upon the properties of the audience groups. This steps should set up as a cycle to improve the performance.

Painless way to create audience groups

Yes, you may not be a data expert. You may well be worried whether you can create proper audience groups without trouble. Or, you may be quite friendly with data. But it’s still not an easy task to keep up with rapidly changing situations. With adbrix, you can not only easily measure data by combining various metrics, groups, and conditions, but you can easily create an audience group by clicking on the chart when spotting noticeable results both good and bad.

For example, let’s say a user group who search hotel prices but don’t purchase shows a sharp decline in retention on the third day. You may assume that users keep searching for hotel prices and end up not satisfied by the result then leave your app on the third day. Upon the assumption, you would send out a coupon with limited, first-purchase discount offering on the second day by a push message or run retargeting ads with the same offering with big volume in order to prevent such exits. With adbrix, you can create and download the audience group simply and quickly by clicking the chart of the corresponding group and select extraction menu. Downloaded audience groups are available for most ad platforms including Google Adwords and Facebook ads.


Social scientists do their utmost to quantify people’s mind, behavior, or trend and conduct experiments to see measurable results. Even now, not everything in society can be evaluated with quantifiable standards, but it is reasonable to say that social scientists’ achievements enabled us to understand people and society better and contributed them to prosper.
The same goes for performance marketing. Marketing is a series of tasks that change a person’s mind and behavior. Still, human mind and action are hard to be quantified with indicators and the factors that generate user reactions are more often emotional elements such as images and copies. So we cannot say that performance marketing is the only and almighty key for all your business challenges. This being said, you can still try to define customer behavior with indicators to the utmost and analyze the data to better understand your customers. If your understanding of the customers gets better, you are ready to change more people’s minds and actions under the same cost.

>> “Performance Marketing 101” with adbrix – Part 1 (Current Page)
>> “Performance Marketing 101” with adbrix – Part 2 (Go to Page)

This post is also available in: ko한국어

About the Author:

애드브릭스의 마케팅을 맡고 있습니다. 마케터의 언어로 말하려 애씁니다.

Get latest news and updates by email