Seller Performance Ratings
BoxCycle now includes key seller performance metrics in listings shown to buyers. This helps buyers determine (1) how quickly the seller is likely to reply and (2) how reliable the seller is likely to be – key concerns in many situations. The reduced uncertainty for buyers should reward high performing sellers with more orders.
Each listing now includes some, or all, of the following details:
- Total number of orders received by the seller
- Number and percentage of orders completed successfully
- Average response time in hours
- Average response time in daytime and open hours
- Time since last activity
All metrics are generated from actual performance. We do not currently use buyer feedback for scoring.
Order Success Rate
Order success rate is shown like this: 83.3% (10/12). The number in parenthesis indicates the number of successful and total transactions that is used for calculating the percentage. Actual numbers are hidden for sellers with more than 15 successful transactions; the metric will then appear like this: 90.4% (15+).
It isn’t possible to cover all the complexities in this post, but two points are worth mentioning:
1) At this time, we consider all orders completed successfully as successful even if there were seller caused issues. The data we’ve seen so far indicates that this rarely misrepresents real-world performance, but it simplifies scoring and creates an incentive for sellers to resolve issues to buyer satisfaction.
2) Orders that are unsuccessful because of the buyer count as successful if the seller replied and isn’t at fault. When the buyer cancels the order prior to seller’s reply, the scoring system will try to properly assign blame based on order specifics.
Average Order Response Time
Response time is counted in hours from order approval until action by the seller (e.g. confirmation, adjustment request, or cancellation.) To give buyers a better idea of likely response time in their specific situation, and to properly represent sellers who reply only during specific hours, we show 3 response time averages like this: 11.1/8.2/1.3 hrs.
The first number is the actual number of hours. For example, an order placed at 1am on Saturday and confirmed at 8am on Monday will count as 55 hours. The second number is the daytime hours; it excludes hours between 11pm and 6am. The example above will count as 36 hours. The final number only counts seller open hours. The example above will count as 0 hours if the seller is open M-F 8am-5pm.
Time Since Last Activity
This metric shows how long it’s been since last activity by the seller. Almost any action by the seller counts as activity (logging in, responding to an order, updating inventory, etc.)
Although we’ve seen very responsive and accurate listings after years of inactivity and unresponsive/inaccurate ones after just a few days, listing freshness is an important quality indicator. Fresh listings are more likely to be accurate and to reply quickly. And buyers are more likely to be uncomfortable with listings that haven’t been updated in a while.
Ultimately, this metric is most useful when the seller has little or no order history. Sellers in this situation can improve the appeal of their listing by logging in occasionally and updating their inventory. However, a large number of orders, or a few recent successful orders, are usually a much better indicator of likely seller performance than time since last activity.
To avoid the need for quality sellers to spend their time worrying about appearance of listing freshness, this metric is automatically hidden in situations when order history makes it unimportant.
We help boxes live longer: 