TikTok Shop US IM Dissatisfaction Rate Calculation and Ways to Reduce It

Your team may answer every message and still receive low service ratings. The buyer might have waited for a specific answer, received information for the wrong variant or been passed between agents without a clear next step.

TikTok Shop's IM Dissatisfaction Rate helps sellers investigate those experiences. Improving it starts with reading the conversations behind the result.



I. What IM Dissatisfaction Rate measures

TikTok Shop's US guide defines the metric using 1- and 2-star post-chat ratings over the past 60 days. The basic calculation divides low-rated chats by rated chats and multiplies by 100. Certain negative ratings from platform Chatbot and FAQ channels are excluded. The metric refreshes daily and contributes to SPS. The guide describes below 15% as the performance of top sellers, rather than a universal guarantee of campaign eligibility. [1]

Use the platform report as the starting point. Do not assume every third-party automated message receives the same treatment as TikTok Shop's own Chatbot or FAQ channels.

II. Understand the denominator before judging the result

Rated chats are not the same as all buyer messages. In a simplified example without exclusions, three low ratings among 20 rated chats produce 15%. The same three among 100 rated chats produce 3%.

That difference matters for small shops. Review both the percentage and the number of ratings so a small sample does not lead to an oversized operational change. Use matching date ranges when comparing internal records with Seller Center.

III. Read the conversation from the buyer's point of view

Take a sample of recent low-rated chats and identify the buyer's original request. Then follow the exchange to its last useful action.

Ask whether the answer addressed the actual question. If a buyer asks whether a device supports a particular connection, a generic product description may be accurate but insufficient. If the buyer reports a missing item, another tracking link may miss the problem entirely.

Record the cause in plain language: wrong variant, unclear offer, missing order check, repeated handoff or unresolved promise. A cause that the team can understand is more useful than a broad label such as “poor service.”

IV. Turn repeated causes into specific fixes


	

		

			

			
Pattern found in the chat
			
			

			
Practical correction
			
		
		

			

			
Buyer repeats the same question
			
			

			
Rewrite the answer around the missing detail
			
		
		

			

			
Agent answers for a different variant
			
			

			
Require variant confirmation before giving specifications
			
		
		

			

			
Promotion information conflicts
			
			

			
Give one owner responsibility for approved offer updates
			
		
		

			

			
Buyer is passed between agents
			
			

			
Assign the next action and keep the conversation context
			
		
		

			

			
Agent promises to investigate then disappears
			
			

			
Record who is checking and when the buyer will hear back
			
		
	


Keep the sample conversations beside each proposed fix. They help agents understand how the new answer differs from the old one.

V. Make replies useful without making them longer

For a buyer asking whether a cable is included, start with the verified answer. Then add the detail needed to choose correctly.

Example: “This variant includes the adapter only. The cable is sold separately. Which device model are you connecting it to?”

For an unresolved delivery issue, state what you know and what you will do next: “The latest tracking event shows [verified status]. I am checking [specific issue] with [team]. I will update you by [time we can meet].”

Replace every bracketed field before sending. The point is to make the next step clear, not to create another vague promise.

VI. Review automated answers and human handoffs together

Test the questions that caused recent complaints against your approved information. Include incomplete wording, similar product names and expired offers. If the information does not support a confident answer, the workflow should ask for clarification or pass the case to a person.

Duoke's AI knowledge and unified conversation workspace can support consistent answers across supported stores. Keep a person responsible for ambiguous requests and follow-up commitments. [2]

Once a week, check whether the same complaint category is recurring. Review corrected knowledge alongside real chats rather than counting only the number of templates added.

VII. Keep satisfaction and response speed in the same review

Use your 24-hour response-rate guide to investigate timeliness, then use this conversation review to investigate helpfulness. For a slow refund decision, continue with the After-Sales Handling Time guide. Each question points to a different part of the workflow.

VIII. Frequently asked questions

Q1:Does a fast reply always prevent a low rating

No. A fast but inaccurate or incomplete answer can still leave the buyer's problem unresolved.

Q2:Should we review only low-rated chats

Compare a few successful conversations too. They show the wording, product checks and follow-through the team can reuse.

Q3:Should we pressure buyers to change ratings

Focus on resolving the issue and follow the platform's current communication rules. Do not attach help, refunds or compensation to a requested rating.

Q4:What should a small team improve first

Choose the most common fixable cause in the recent sample. Update the relevant answer or ownership rule, then inspect the next set of conversations.

Explore Duoke to organize buyer conversations and approved product knowledge before your next sales campaign.

Official sources

[1] TikTok Shop US — Instant Messaging Dissatisfaction Rate Requirements. https://seller-us.tiktok.com/university/essay?knowledge_id=5513514540189453

[2] Duoke official website. https://www.duoke.com/en/index.html