How Can Ad Optimizer Make English Novel App’s Daily ROAS Surged to 2!?

5 min readNov 27, 2020


This article comes from the contribution of Ethon, an Ad Optimizer, and the article analyzes the advertising optimization ideas of the novel app on Facebook. Hope this will help other Ad Optimizers.

Product Promotion Overlook

  1. Project Background:

While the novel apps’ markets are steadily rising, we put English novel apps into English-based markets. The novel types are mainly modern romance, ancient romance, werewolves, the god of war, and other genres. We strive to introduce English online novels to the world.

2. Launched Product:Novel App

3. Advertising Channel: Facebook

4. Covered Region:Worldwide

Detailed Promotion Strategies

1. Advertising Promotion Methods

Different from the advertising of other apps, for thousands of novels will be in the app, the advertising data of each novel will be needed for the advertising based on the dimensions of the book, that is, each book corresponds to a campaign; Since our KPI is to increase the user’s paying amount and increase the ROI (roas 0.3–0.5 on the day of the assessment), we adopt AEO (purchase) and VO methods to maximize our revenue.

2. Audience Targeting


At the beginning of the advertising campaign, we did not know in which countries our app would perform well, and different types of novels performed variously, so we chose not to filter countries at the beginning and allow the system to freely choose which country to launch the advertising campaign in all English-using countries. According to the historical data, the national dimension data performance of different books and devices was screened. Then in the later period, it is planned that the female-oriented books were mainly advertised in the United States, the United Kingdom, Canada, and the Philippines; the male-oriented books were mainly advertised in the United States, the Philippines, and India;

2)Advertising Serving Audience:

In terms of choosing the target user, after I took over the account, I operated on creating a custom user in the first place, collecting the number of users whose purchase and paid amount was in the top 25%. After the overlap check, the overlap rate was 30% which is acceptable to me; Then we used these two types of audiences to create LAL audiences. Later, when it was finished, we checked again the overlap ratios, both within the acceptable range.

3) Ad Materials:

After analyzing the materials put in the past, by comparing the data of some key indicators (such as CTR, CPI, and ROAS), we conjectured the common points of high-quality materials, in other words, what may be the reasons for the excellent data. After that, we did a batch of material verification test again, and then verified the previous guess based on the data; we found that materials which highlight the couple and with warm colors and higher saturation have better effects on female-oriented books; the male-oriented books highlight the leaders and gangsters, and materials using cold color and with cool style have performed better.

You can find the ad materials for similar products on major advertising platforms on the App Growing Global material platform. Take a look at the ad materials for different companies’ products, which will also inspire the material creation of your own products. Examples of ad materials with better performance:



Promotion Campaign Data

In these three months of advertising, the advertising agency did not use LAL to target the audience until October. It can be found that our ROAS surpassed advertising agency in August and September. While in October, the advertising agency had a soar in ROAS after targeting purchased audiences. The growth shows that when audiences are targeted, it also has a relatively large positive effect on the final profit. (From September to October, our ROI dropped by 5%; while the advertising agency’s ROI increased by 7%)

Note: The date of statistics is at 10.28, the overall data in October has a slight error.

When summarizing the data, it is also crucial to know each book’s data on each device. We found that the female-oriented category performed better on IOS, while the male-oriented category performed better on Android. Only by digging out hit books and increasing the amount of advertising budget can ultimately increase the overall ROI.

To sum up, shortly after taking over the account, the advertising method and the direction of updating creativity were basically settled. ROAS has also been positively improved after targeting the LAL audience. However, the optimization of the novel app is still in a state of gradual exploration; we still need to explore and accumulate experience step by step; in the future, we will consider using GG assisted advertising, but the difficulty is that UAC cannot distinguish between gender. How do we advertise the corresponding books to the right people? I think we can overcome this difficulty ultimately.

Author’s brief introduction:

Ethon, started to do Chinese market advertising in 2017, and global advertising in 2019. The platforms Ethon mainly deal with are Facebook, Google, and Tiktok. Ethon has also involved in payday loan app, casual games, and novel app advertising and likes to discover users’ behaviors through data analysis.

Thanks to Ethon for sharing in App Growing Global, App Growing Global will continue to share more. Welcome ad optimizers who are interested in this to contribute actively, so as to enhance personal value, help ad optimizers to grow rapidly, and expand the exchange circle of ad optimizers.




Global Marketing, know what, know why, know how.