Receive information about platform updates, training materials, and other news.
Receive information about platform updates, training materials, and other news.
Pricing impacts how fast a task is taken up, but it does not directly influence the quality of results. The only way to actually influence quality with money is by using a smart motivation scheme, where the task price changes depending on the quality of individual performers.
Performers try to maximize earnings while minimizing effort and time spent. However, it is highly undesirable for requesters who want their data to be labeled with the best quality possible.
This can be solved with performance-based payment. Instead of having a fixed price for each approved task, payment can be connected to the quality of the work done by the performer. There are various ways to calculate individual quality, but most of them rely on checks that are hidden from the performer. This means that to maximize their earnings they need to complete every assigned task carefully, thereby increasing the overall quality of data.
Here's an example of how a bonus of $0.50 can be awarded to performers with different quality. Note that in all three cases quality is reasonable, meaning that the performers have successfully passed through basic quality filters.
Fixed payment | Accuracy bonus | Total | |
One complex task | $1 | 0.7 x $0.5 | $1.35 |
Several smaller tasks | $1 | 0.5 x $0.5 | $1.25 |
Correct on 10 out of 10 | $1 | 1.0 x $0.5 | $1.50 |
Important points to keep in mind when using performance-based pricing:
In Toloka, you can configure different prices for different skill values by using dynamic pricing pool setting.