We all know that the ultimate goal of information flow advertising is to obtain more high-quality traffic at the lowest cost, so as to achieve accurate conversion , which is the legendary "making every penny invested count".
Therefore, when it comes to information flow optimization, we will feel obliged to rack our brains, constantly deduce, and try and error, in order to make the optimization and conversion path more and more perfect. However, however... the road to perfect optimization is not always so smooth, and there will always be bumps along the way... What I am going to tell you below is what you desperately want! This is a story about how we found out the conversion path of customer accounts and the problems with landing pages , and then solved various difficult problems one by one in a targeted manner, thereby improving conversion and reducing costs. I hope to use this experience to exchange experiences with all of you Aidoers, and learn and improve together.
First, let’s start with the origin of the story:
Protagonist's optimized account: A certain information technology company's account background: In early 2013, the Renren Weipai art purchase website was officially launched. Currently, the website has nearly 1.8 million registered users and more than 4,500 cooperative merchants. As Renren Weipai has further developed and grown in recent years, it has invested heavily in advertising in order to find clues to better quality and more valuable art products.
Account promotion records:
The advertiser cooperated with Baidu in August 2017. In the early stage of cooperation, the client focused its budget on the search side, and the investment in information flow was relatively weak. The client's daily budget was 200, but the retention conversion was less than 5, and the cost was much higher than the search conversion. As a result, within less than two weeks of testing, the client felt that the information flow was poor, and began to reduce the budget and suspend the delivery.Account distress rescue: First, we gained a comprehensive understanding of the industry and population in which the products were being marketed, and conducted a comprehensive analysis and troubleshooting of the account data. We identified the problem and developed an effective solution, allowing the account to be marketed a second time. Now that the origin story has come to an end, what kind of rescue plan can rekindle hope for the information flow? Next, we will enter the core part of the whole story – account rescue and optimization routines! After checking the account structure, target audience, and creative materials, we found that the existing creatives had low eye-catching ability. This was mainly due to the fact that the creatives did not match the content of the landing pages, the landing pages were too simple, and the product advantages were not prominent, resulting in a bounce rate as high as 98%.The first rescue step: Account structure optimization Optimization action: Analyze customer industry and accurately target the target group Through Baidu Index, we can understand the industry terms of advertisers, extract and expand high-quality keywords , and refine keywords in multiple dimensions. Understand the distribution of hot regions of customer industries through big data and carry out key monitoring. By comparing the data of customers’ industry search popularity, gender and age ratio, we conduct classified delivery tests.Second rescue method: creative optimization Optimization action: use keywords + interests to comprehensively target and A/B test the creative quality Before optimization: CTR : 0.34% Visit duration 6s After optimization: CTR: 4.2% Visit duration 23s
1. The copywriting and pictures do not have highlights to attract users, and the target user group is not clear
2. The copywriting informs customers that the auction is risk-free in the form of a notice, dispelling the concerns of the target group
The third rescue method: Landing page optimization Optimization action: Continuously optimize the landing page and select the best one For landing page testing, you can create 2-3 sets of landing pages based on the optimization data, with the same theme and creative materials, and divide the landing pages into ABC segments to compare the delivery data. Landing Page A Landing Page B Landing Page C The above picture shows A, B, and C from top to bottom. The above landing pages are tested continuously until the effect is stable.Optimized results:Overall style: The visual effect is clean and clear, the module framework is refined, the colors are divided into primary and secondary tones, and the font selection and size are appropriate; the material information is modularized, and each module information is a theme, which effectively highlights the key information and weakens the secondary information. Data improvement:
In order to better optimize the effect and monitor the data, we made two mobile marketing pages, and compared the monitoring data according to the planned launch. From the overall background data, we can see that the conversion effect of the optimized landing page is very obvious, the retention amount is increasing, the user bounce rate is gradually decreasing, and the average page stay time is getting longer and longer.
From September 21st when the app went online to October 16th, there were 933 clues in total.
From 0 conversions in early August to about 145 appointments per day now, and about 70 form submissions per day; CTR increased by 12%; conversion volume increased 20 times, and conversion cost dropped from 100 to 10, successfully meeting customer conversion expectations. Let me summarize my information flow optimization experience again, please see the picture below! In summary, when setting the delivery direction for information flow ads, you must control every detail to ensure a seamless connection! If the 6 points in the above picture can be perfectly matched, the information flow effect will always be 666, and then there will be no need to worry about the boss being dissatisfied!
This article was compiled and published by @Baidu Marketing Observation (Qinggua Media). Please indicate the author information and source when reprinting!