Based on recent concerns, we have decided to answer some typical questions. This issue focuses on e-commerce and conversions. This issue includes three aspects: cross-platform data integration, referral analysis and conversion analysis.

Cross-platform data integration

Q:Self-built e-commerce platforms, such as self-built websites, and third-party e-commerce platforms, such as amazon and eBay, are generally used by large brands simultaneously. Can we transfer the third-party platform traffics to our own self-build website through a direct link?

A:If you want to build your own brand, you must have your own independent station. I'm a big fan of self-building, not only because it's a symbol of a strong corporate brand, but also an effective way to balance the increasingly worrisome platforming monopolization.

It is feasible to direct the flow of the platform to independent stations. Many of the cross-border e-commerce merchants currently on amazon's platform do. The main reason is that the amazon platform is too unreliable for a large number of small and medium-sized sellers. Once the store is closed it incurs heavy losses for such retailers.

However, it is unlikely to attract traffic directly to the main website. Some e-commerce platforms do not allow external links. So, you have to direct people, through direct messages or images, to desired location. . So, You need a boot instead of a direct link.

Referral analysis

Q:For the cross-border e-commerce platform like Ali Express, besides P4P, do we have any other channels to direct the traffic flow? Currently, it hopes to find another way of attracting traffic outside the platform. Previously, it tried to use Google Adwords to direct the traffic flow, but it failed to track the final data like the website. Industry BBS and social media platforms are also available, but with little effect. Did we ignore any other way to generate traffic? Are there other we can use to avoid relying solely on the platform's own traffic?

A:Currently, the main way of driving cross-border flow is using major platforms, manual search and social media as mentioned. It seems that there are two problems in this scenario. The first is whether there are better channels. The second is how to track the effects of these channels.

When it comes to tracking, there are alternative solutions even if you can't see the actual ROI at the end.

Build your own diversion station and use it for monitoring. Traffic mainly comes partly from search, part social traffic, and some vertical platforms.

Conversion analysis

Q:How does a retailer site improve its sales and quality of orders?

A:Retail website is not different from e-commerce in essence, so the improvement of sales volume and orders’ quality must start from the product itself. If the product is good, things work out.

Let’s see, how to increase sales, regardless of the product itself:

Traffic can be increased to retailing website by offering various packages based on price discrimination. Prices of the commodity can be used to do several external promotions. In addition, it is also important to increase traffic cooperation through business development.

 

Q:Does the quality of the order refers to the increase of customer unit price?

A:The control on the order quality is essentially dependent on the product quality and crowd quality. I will not mention the goods; the quality of the crowd, however, mainly depends on the control of front-end traffic, as well as site conversion guide combined.

 

Q:The conversion rate of subsequent purchases by registered users who are directed from paid traffics has declined. How should this decline be? From the perspective of traffic quality, it can be seen that the traffic quality changed, because the conversion rate of initial registered users remains unchanged.  However, the conversion rate from the users subsequently registered has decreased. Can this be analyzed from the user behavior, and if yes, what indicators should we use to analyze this?

A:Yes, it is possible to analyze user behavior.

For indicators, we must analyze the performance on the channel before and after the reduction, the performance of the heat map on the landing page, and in particular, the conversion performance of what you may call the “sign-up to purchase” step.

You can also look at other information, such as the time of stay, whether there is a difference between the user’s device, and the location before and after the period.

 

Q:Why do we use the conversion data and sales revenue promoted by Facebook and the difference between the conversion and sales revenue recorded by GA is particular, with a gap of more than five times.

A:The biggest reason is attribution.

Facebook monitors conversion using the code that Facebook adds to advertisers' websites. This code is based on cookies. Once a user visits the advertiser's site via Facebook ad, Facebook can always identify whether the person is back on that site, even if the user comes from another site. Then, if such a user makes a purchase, Facebook is credited with the conversion and makes money off it.

Consider a rather interesting situation given below, which is actually quite common.

If a user who enters a page through Google AdWords (GA) ads, but doesn't buy anything, then he or she gets into the same page through Facebook ads, and still doesn't buy anything. However, on the third occasion, if the same person enters that very page through an Email ad (EDM) and buys something, Who will get the credit?

Clearly, the Facebook code on the page would take credit for the Facebook ads, but GA, by default, also has a right to claim for directing the email advertisement to that user.

This is the core reason for the transformation difference. Comment below if you can think of any other reason. In this case, I think the attribution analysis using Google Analytics seems to be the most accurate.