This article provides an overview of the OI Predicted Satisfaction score - including how it is calculated, and tips to improve this metric for agents.
What is the Predicted Satisfaction score?
Unlike traditional customer surveys, predicted satisfaction is determined through analysis of the words a customer uses in a transcript. Meaning every interaction is assessed, not just those where we get a survey response. The scale for measurement is as follows:
- 1 – Very Unsatisfied
- 2 – Unsatisfied
- 3 – Neutral
- 4 – Satisfied
- 5 – Very Satisfied
As an agent, you can directly influence Predicted Satisfaction by managing the customer’s experience and using positive language. Since the score is based purely on the positive and negative words a customer says, your goal should be to guide the interaction in a way that encourages positive customer expressions while reducing frustration or negativity.
How is Predicted Satisfaction calculated?
- Satisfaction analysis: The model scans customer language for positive and negative words and phrases (e.g. frustrated and useless vs happy and appreciated).
- Weighted scoring: Each word is assigned a satisfaction weight based on trained data.
- Scoring scale: The satisfaction of the entire conversation is analyzed and compared against all other interactions from your company.
What does this mean for me, as an agent?
Better insights: Only ~10% of customers respond to surveys, but this method ensures every customer’s sentiment is considered.
Removes survey bias: No need to rely on who received or responded to a survey, so all interactions are treated equally.
Consistent and objective: Satisfaction is measured the same way across all interactions, allowing for meaningful analysis.
What can I do to increase my predicted satisfaction score?
Create a positive experience from the start
- Use a warm, friendly tone to set the right mood for the call.
“Thanks for reaching out today! I’m happy to help you with this.”
- Acknowledge the customer’s emotions early on to prevent frustration from escalating.
“I understand how frustrating this must be, let’s get this sorted for you.”
Reduce customer frustration
- Avoid vague or unclear answers that might cause confusion.
“Let me check that for you now, so we can get you a clear answer.”
Instead of
“I’m not sure what’s going on with your account.”
Take ownership of issues instead of pushing responsibility back onto the customer.
“I’ll monitor this for you and follow up if needed.”
Instead of
“You’ll need to call back if this doesn’t go through.”
Use positive language and reframe negative situations
There may be situations where you can’t give the customer exactly what they want, but how you say things matters.
“Here’s what I can do to help.”
“Let me check all possible options for you.”
Use solution focused alternatives.
“You’ll receive an update within 3 days.”
“I will refresh my system to make sure I'm giving you the most accurate information.”
Always show empathy and reassurance
Customers who feel heard are more likely to use positive language
“I completely understand why this is frustrating, and I’m here to help.”
“I’ll personally make sure this is handled for you.”
End on a high note
Summarise what’s been done and reassure the customer before ending the call.
“I’ve updated your details, and your issue is now fully resolved. You should receive confirmation shortly. Is there anything else I can do for you today?”
Thank the customer and invite a positive response.
“I appreciate your patience, and I’m glad we could sort this out for you today!”
What to avoid (triggers for low satisfaction scores)
- Allowing the customer’s frustration to escalate without expressing a willingness to help.
- Sounding robotic or indifferent (lack of warmth, empathy, or engagement).
- Forcing the customer to repeat themselves (causing unnecessary frustration).
- Using dismissive or negative phrases that discourage a positive sentiment.
- Leaving uncertainty (e.g. “I’m not sure,” “Call back later,” “Nothing I can do”).
Final Takeaway
By focusing on positive language, active problem-solving, and a warm approach, you can increase the likelihood of customers expressing satisfaction, which will boost your Predicted Satisfaction score.
Comments
0 comments
Please sign in to leave a comment.