Cross-border e-commerce evaluation analysis and emotional persuasion strategies

Evaluation word frequency analysis and sentiment score assessment

When processing consumer feedback, cross-border e-commerce companies often need to use data analysis tools to conduct evaluation word frequency analysis and sentiment score evaluation. This process not only helps understand the real needs of consumers, but also provides direction for store improvement.

Evaluation word frequency analysis

First, collect online text records (including but not limited to chat records and comments) of consumers in the target store, and then use the PQ editor to connect to dictionary resources to split and group statistics on the collected text information. The specific steps are as follows:

• Open data source: Select “Project 6, Task 4, Consumer Public Opinion Analysis” as the data source.
• Import into PQ Editor: Select “From Table/Range” under the “Data” tab in the menu bar, and specify the data range in the pop-up dialog box.
• Process data: By adding custom columns and applying specific formulas (such as =Csv. Document (Web. Contents("http://api.pullword.com/get. php?source="&Uri.EscapeDataString([Evaluation])&"&.param1=0.8& param2="))) to implement word segmentation processing of text. Subsequently, the word segmentation results are filtered, grouped, counted, and sorted.
• Export data: Once processing is complete, select the “Close and upload” option to save the data.
• Create a pivot table: Use the pivot table function of Excel to build a pivot table based on the “Participle.Column1” field row and the “count” field value.
• Insert Chart: Select “Clustered Column Chart” in “Recommended Charts” to visually display the word frequency distribution.
• Chart optimization: Optimize settings for charts, such as displaying the top 20 high-frequency words, removing unnecessary grid lines and coordinate axes, etc.

Through the analysis of the top 20 high-frequency words, it is found that excluding some common English words (such as “the”, “will”, etc.), the number of consumers’ positive words is significantly more than the negative words, reflecting the store’s logistics Good performance in aspects such as (shipping), product quality (quality) and service (service).

Sentiment Score Assessment

In addition to word frequency analysis, it is also necessary to evaluate consumer evaluations with sentiment scores. This process also relies on the functionality of the PQ editor:

• Import data: Similar to the above steps, select the data from “Project 6 Task 4 Consumer Public Opinion Analysis” and import it into the PQ editor.
• Emotional Computing: By adding a custom column “emotion” and its related formula (such as =Json. Document(Web. Contents("http://106.13.102.102/query ? key="&[Evaluation]))), get the positive (positive_probs) and negative (negative_probs) scores of each comment.
• Data filtering: Retain some sample rows for subsequent analysis.
• Export data: Export the processed data for further analysis.

The results show that among the 5 randomly selected reviews, 4 reviews have positive scores higher than negative scores, indicating that consumers generally have a positive attitude towards the store; for negative reviews, further analysis is required. and make suggestions for improvements.

Persuading the other party based on emotion

In the process of cross-border e-commerce negotiations, how to effectively impress the other party emotionally is particularly important. Here are a few practical tips:

• Resolve doubts: First of all, you must win the trust of the other party and avoid arousing their wariness. A connection can be established gradually by complying with the other person’s ideas.
• Empathy: Think about the problem from the other person’s perspective and show a sincere attitude. Even in business negotiations, keep a human touch and strive to be a friend rather than a mere adversary.
• Provide reasons: Provide a reasonable explanation for the other person’s decision, so that they feel that they are not only considering themselves, but also taking into account the interests of their family, friends or company.

For example, when selling gold jewelry to a hesitant female customer, the salesperson can emphasize that buying jewelry can both enhance personal charm and please your partner; at the same time, gold, as a value-preserving asset, will not cause any damage even if it is resold in the future. economic loss. Similarly, when recommending exquisite tableware to a male customer, you can subtly imply that his wife will not let him wash the dishes because she has new tableware, thereby increasing the likelihood of a successful transaction.