When working in an operation position, one of the daily tasks is to compile the latest data of the store, including sales, traffic, conversion rate, advertising expenses, etc.
The compilation of these daily data may take up 5% to 10% of working time every day, but it should be noted that data is not used to obtain inspiration, but to verify inspiration!
p>
Why is it so emphasized? This is because many operators engaged in Amazon cross-border e-commerce like to try to “mine” some valuable information from the data, and then use this information to “overtake” others. Competitors, this idea is obviously wrong for the following reasons:
1. Amazon’s entire online sales platform is similar to a virtual world, and the data of personal stores, whether in terms of traffic or conversion rate, is different from that in the real world. The physical store data are almost the same, so the data itself contains great uncertainty.
For example, you find that the conversion rate suddenly plummeted yesterday, and then you guess that the interface of the new product launched yesterday must not be good enough, or the keywords were written wrong, etc.
In fact, the real reason may be that customers on Amazon participated in some activities for some reasons during that period, or some objective reasons led to a decline in Amazon’s overall traffic. It’s like if you open a physical store and encounter a rainy day, the number of customers will be very low, but you can never guess whether it will rain in the next second.
2. All data is a short-term reflection of the Amazon store you operate, and whether a store can successfully operate in the long term depends on the “real” things such as products, services and logistics. Trying to blindly improve keywords, pictures or titles through changes in conversion rates and traffic will only be counterproductive.
3. Data has a lag. Anyone who has studied statistics or econometrics knows that statistical data includes time series data and non-time series data, and all data available on Amazon are basically Time series data, that is, these data have a strong correlation with time. It can only explain the past and cannot predict the future.
To sum up, as operators we need to change our previous operating philosophy:
From the past “data changes – analyze data – get inspiration – improve operations”. ( x )
To the current “getting inspiration-operation practice-data changes-analyzing data-verifying inspiration”. (√)
Of course, this does not mean that data analysis is not needed before inspiration practice, but that data analysis is to practice new “innovation” and “inspiration” again and again until Until your “innovation” and “inspiration” are proven correct.