Skip to main content

​
Rating and NPS System

​
1. Overview

After every completed job, you rate your expert’s performance and they rate your collaboration - no retaliation risk, because ratings stay hidden until both sides submit or seven days pass. These ratings build reputation scores that improve matchmaking quality and help you identify the best experts to work with again. The system also collects NPS feedback to track platform satisfaction and spot areas where we can improve your experience.

​
2. Step-by-Step Guide

Rating an expert after job completion:
  1. When your job is marked complete, you receive a prompt to rate the expert’s performance
  2. Rate the expert on four 5-star scales: quality, communication, timeliness, and professionalism
  3. Add written feedback if you want to elaborate on specific strengths or areas for improvement (optional)
  4. Submit your ratings - they remain hidden from the expert until they submit their ratings or seven days pass
  5. After submitting performance ratings, answer the NPS question: “How likely are you to recommend 10,000 Experts to a colleague or friend?” on a 0-10 scale
  6. If you select 0-6 (Detractor), you’ll see a follow-up question: “What could we improve?”
  7. Your ratings permanently update the expert’s overall score, which other clients can see as an aggregate
Viewing expert ratings before hiring:
  1. Browse experts in the Jobs Board or Marketplace
  2. Check each expert’s aggregated rating scores (averages across all jobs, not individual client ratings)
  3. Look at the rating count - experts with fewer than five ratings may not have established track records yet
  4. Review category-specific averages (quality, communication, timeliness, professionalism) to identify strengths
  5. Check the trend over the last ten jobs to see if performance is improving or declining
Understanding your own collaboration score:
  1. Navigate to your profile or dashboard
  2. View your aggregated collaboration score based on expert ratings
  3. See category breakdowns: clarity of requirements, responsiveness, professionalism, and payment timeliness
  4. Individual expert identities remain private - you see only aggregated scores

​
3. Common Questions

Why can’t I see who rated me poorly? Ratings are anonymous to prevent retaliation and bias. You see aggregated scores and category averages, but not which specific expert or client provided each rating. This protects honest feedback and keeps the system fair for everyone. What happens if I don’t submit ratings after a job? You’ll receive reminders for seven days. If you don’t respond, the job closes without your ratings. The system tracks non-response rates, and consistently skipping ratings may affect your collaboration score. Your expert still submits their rating of you - it just releases after seven days instead of when you submit yours. Can I change or dispute a rating? No. Ratings are final and cannot be changed once submitted. If you believe a rating is unfair, you can add a public rebuttal visible alongside the rating. Extreme cases (offensive content, clear errors) can be escalated to moderators, but the numerical rating remains. How do ratings affect expert matching? Experts with higher average ratings receive higher match scores when the algorithm pairs experts with jobs. If you have consistently high ratings, you’ll get matched with better opportunities. If a client has low collaboration ratings (below 3.5), experts see a warning about the client’s history before accepting the job. What if an expert is brand new with no ratings? The system displays the rating count prominently. Until an expert has five or more ratings, matchmaking uses the platform average as a baseline. This gives new experts a fair shot while protecting you with transparent information about their track record.

​
4. Troubleshooting

Problem: I submitted ratings but they’re not showing up on the expert’s profile yet Solution: Ratings reveal only after both parties submit or seven days pass. If your expert hasn’t submitted their ratings yet, yours remain hidden to prevent you from seeing their rating and retaliating. Check back after the seven-day window or after they submit. Problem: I received an offensive comment in written feedback Solution: Flag the feedback for moderator review using the flag button next to the comment. The system automatically filters profanity, but moderators will remove inappropriate content while preserving the numerical ratings. Flagged users face behavioral review if patterns emerge. Problem: An expert has a suspiciously perfect 5.0 rating across all categories Solution: The system monitors for suspicious patterns like all 5-star or all 1-star ratings. Check the rating count - a small number of ratings may legitimately result in perfect scores. If you suspect manipulation, report it. The system requires completed paid jobs for ratings to count and flags anomalies for moderator review. Problem: My collaboration score dropped suddenly after one job Solution: Check your category averages to identify which area declined (clarity, responsiveness, professionalism, payment timeliness). One poor rating can impact your average, especially if you have few total ratings. Focus on improving in that specific area for future jobs. The system tracks trends over your last ten jobs, so consistent improvement will repair your score. Expert Dashboard - View all your ratings, track performance trends, and see how your scores compare to platform averages. Your dashboard breaks down category strengths and highlights areas for improvement based on client feedback patterns. Matchmaking System - Your ratings directly influence which experts the algorithm matches you with. Higher collaboration scores get you paired with top-tier experts who prefer working with responsive, professional clients. Dispute Resolution - If a job ends badly and you receive unfair ratings, the dispute system lets you escalate issues to adjudicators. While ratings remain final, successful disputes may result in notes added to your profile explaining the context behind outlier scores.