In our communication this time, the final plan is to business email list make statistical reports to monitor the changes in prices.
However, I still maintain a "skeptical" point of view in my heart: the business side is not necessarily just looking at price changes, it is also possible that price changes have an impact on their performance, and they get price change data for statistical performance. If we can further directly calculate the performance of each colleague in the group through the system, the use of the business side will be more convenient.
Demand is invisible, and users often don’t know what they want. Due to the limitations of their own cognition, the solutions proposed by users are often a certain method that he has seen. All real needs come from communication. We must continue to delve into the core problems and provide solutions.
2. Maintain a "skeptical" attitude towards technical solutions
Maintaining a "skeptical" attitude towards technical solutions does not mean distrusting the level of technology, but just reminds us to always be prepared to deal with unknown risks.
Below is a problem with one of the systems I'm responsible for.
Demand: Calculate the trend value of sales in the past 30 days. The sales may fluctuate greatly in the past 30 days, so it is necessary to judge and use different calculation formulas.
The calculation logic of the design scheme: determine the number of values with sales > 0 in the past 30 days :
If the number is greater than 15, the calculation formula uses the median;
If the number is <= 15, the calculation formula uses the average number.
Calculation logic implemented by technology: determine the number of values with sales = 0 in the past 30 days :
If the number <= 15, the calculation formula uses the median;
If the number is > 15, the calculation formula uses the average.
For the convenience of judgment, the technology has changed the way of judgment. In most cases, the calculation results of these two methods are absolutely consistent. The test students simulated data to verify, the results are consistent, and the demand acceptance is online.
After going online, we ran for two days with a large amount of real data. When I checked the data, I found that most of the data were normal, and some data were abnormal. is a bit inconsistent.