Here at Lilt, translator performance is measured using a range of Key Performance Indicators (KPIs). KPI data is captured for each translator on a per-Project basis. Lilt analyzes KPIs for long-term trends to help us:
- Efficiently build teams for our customers.
- Provide high quality and consistent translations.
- Support our translator throughout their journey with Lilt.
On-Time Delivery (OTD)
All tasks assigned to translators contain a deadline that indicates when the task needs to be completed by. The OTD rate is the percent of tasks the translator has delivered on time.
When a translator is assigned a task, they can accept or reject the assignment. Acceptance rate is the percent of tasks the translator has accepted.
Whenever possible, our Services Managers provide advance notice to translators about upcoming Projects, but sometimes Projects come up unexpectedly. We invite our translators to fill in their calendars so we can most efficiently assign Projects.
Error Rate can be viewed by project managers and is a useful metric to evaluate the performance of translators. Error rate is the per percent of segments reviewers have marked as having errors out of the total number of segments reviewed.
Depending on the customer, errors and error categories can be marked by our reviewers as well as by the customer’s reviewers. For further details on how content is reviewed by our reviewers, see the Reviewer Role & Responsibilities article.
The following error categories contribute to error rate:
- Language Adequacy: punctuation, typos, and spelling errors
- Text Spec: client or industry approved terminology
- Mistranslation: translation is not accurate, or source text is mistranslated
- Readability: target text has an awkward word or sentence structure
- Technical: incorrect tag placement, broken variables, or missing spaces
- Repeated Error: errors repeated from TM matches
The following error categories do not contribute to error rate:
- Source: indicates an error in the original source segment
- Not an error: stylistic update or omissions/additions that do not affect meaning
Customer Feedback and Quality
Lilt is committed to our customers. We evaluate customer feedback in detail to assess the root cause of any issues so we can apply corrective actions.
Lilt receives feedback from our customers regarding the Projects we deliver. Some of our customers work with their in-house reviewers or with partnered Linguistic Quality Audit (LQA) reviewers to provide detailed feedback on an ongoing basis. This feedback is shared with our translator teams.
With each Project assigned in Lilt, we capture data on the translation productivity and review productivity of our linguists. This data is used to ensure that linguists are utilizing all Lilt’s features and functionalities to work in the most efficient way possible.
The Lilt platform utilizes AI and Neural Machine Translation (NMT) to improve the productivity of translators. We are continuously evaluating and improving the quality and speed of MT suggestions to help translators be more productive.
We work in a fast-paced environment that requires fluid and open communication for us all to be able to do our best work. Maintaining a high level of responsiveness is key to producing the most optimal solutions for our customers. This relates to all communication, whether it’s accepting or rejecting Projects, sharing additional Project instructions, or discussing feedback.