Intercom Surveys Integration

Responsly syncs survey responses to Intercom user profiles as custom attributes. Agents see satisfaction data in the conversation sidebar. Automation rules route and tag conversations based on feedback. Outbound messages target segments defined by survey answers.

Intercom captures what users ask about and what they do in your product. Surveys capture what they think — the structured signal that conversation transcripts and event data cannot provide. When both live on the same user profile, every support interaction, every automated message, and every routing decision is informed by direct feedback.

Why user profiles need structured feedback

Intercom event tracking tells you a user logged in 47 times last month and used the reporting feature 12 times. Conversation history tells you they asked about CSV exports twice. Neither tells you they’re frustrated with report load times and considering a competitor.

An NPS score of 4, synced as a custom attribute, changes how your team interacts with this user. The support agent handles the next ticket with extra care. The automation engine routes their conversations to a senior queue. The product team knows this active user — someone who logs in 47 times a month — is unhappy about something specific.

Structured survey data provides the sentiment context that behavioral and conversational data lack. When it lives on the Intercom profile, it’s available everywhere: in the inbox, in custom bot logic, in outbound message targeting, and in reporting.

Connecting Responsly to Intercom

The integration syncs through Intercom’s REST API.

1. Create an API token — in Intercom, go to Settings → Developers → Developer Hub. Create an app with read/write permissions for users and contacts.
2. Connect in Responsly — go to your survey’s Integrations tab, select Intercom, and paste the access token. Responsly validates the connection and loads available attributes.
3. Map survey fields — assign each question to an Intercom custom attribute. Scores map to number attributes, categories to string attributes, text to string attributes.
4. User matching — Responsly matches by email address or Intercom user ID passed in the survey URL.
5. Test — send a message with a survey link, complete the survey, and verify attributes update on the user profile.

Post-resolution CSAT that fixes support quality issues

A SaaS platform with 40 support agents struggled with quality consistency. Overall CSAT was 3.8/5, but the team suspected wide variation across agents and issue types. Monthly QA reviews caught problems too late — weeks after the conversation happened.

They added a post-resolution CSAT survey triggered when an agent closes a conversation. The survey link went out via Intercom Messenger 30 minutes after resolution. Two questions: satisfaction rating (1-5 stars) and “What could we have done better?” (optional text).

Responses synced to the user profile and were tagged to the conversation via a custom attribute last_csat_conversation_id.

What the data revealed:

• Agent-level CSAT ranged from 2.9 to 4.8. The bottom three agents weren’t bad at technical support — they were slow to respond. Average first-response time correlated with CSAT at r=0.72.
• Billing-related conversations scored 2.1/5 on average, regardless of agent. The problem wasn’t agent quality — it was a confusing billing page that created preventable tickets.
• Users who received a response within 2 hours rated satisfaction 4.3/5 on average. After 4 hours, satisfaction dropped to 3.1/5.

Actions taken:

• SLA targets were tightened from 8 hours to 3 hours for first response, justified by the CSAT correlation data.
• The billing page was redesigned, reducing billing-related tickets by 44%.
• An automation rule was created: CSAT ≤ 2 auto-tags the conversation “low-csat-review” and routes it to the team lead’s inbox.

After one quarter: overall CSAT improved from 3.8 to 4.3. Agent variance narrowed. The team could tie specific process changes to measurable satisfaction improvements. See our guide on customer service metrics for more measurement approaches.

In-app NPS that segments product communication

An NPS survey triggered after 30 days of product usage captures sentiment at the critical adoption window. The score syncs to the Intercom user profile and powers three distinct communication tracks.

Promoters (9-10): Users who love the product receive an Intercom outbound message inviting them to a beta program for an upcoming feature. The logic: engaged, satisfied users provide the best beta feedback. A second outbound message two weeks later invites a case study interview. Promoter contacts are tagged for the product team’s “Power User” segment.

Passives (7-8): These users see a targeted Series message highlighting features they haven’t used yet. The message is personalized using Intercom’s product data — if a passive user hasn’t tried the integrations module, the message showcases integration use cases. The goal: deepen adoption to convert passives into promoters.

Detractors (0-6): The response triggers an immediate notification to the product lead. A personal message from the product team asks the specific question: “What would need to change for us to earn a higher score?” The reply is routed to a dedicated inbox. Detractor contacts are excluded from marketing messages for 30 days — no one wants a promotional email the day after expressing frustration.

One product team tracked this system over eight months and found:

• 28% of detractors who received a personal follow-up improved their score in the next quarterly survey.
• Beta invitations sent to promoters had a 61% acceptance rate vs. 23% when sent to the general user base.
• Passive-to-promoter conversion from feature-highlight messages ran at 14% per quarter.

For broader NPS implementation strategy, see our guide on how to calculate and use NPS.

Onboarding feedback that adapts bot flows

Onboarding is the highest-leverage moment for a survey because early dissatisfaction predicts churn more strongly than any other signal. An Intercom custom bot can trigger a survey at the end of the onboarding wizard and adapt the next step based on the response.

Bot flow with survey integration:

1. User completes onboarding → bot sends a two-question survey: “Was the setup process easy?” (1-5) and “What do you want to accomplish first?” (multiple choice).
2. Responses sync to the user profile as onboarding_ease and primary_goal attributes.
3. The bot branches based on the response:
   • Easy setup + specific goal → bot sends a targeted resource (help article, video) for that goal and schedules a follow-up check-in for 7 days.
   • Difficult setup → bot connects the user to a live agent immediately. The agent sees the onboarding_ease: 2 attribute and knows the context before the user types anything.
   • Primary goal data → stored for the product team’s quarterly analysis of what new users are trying to achieve. This informs onboarding flow redesigns.

A project management tool using this approach found that users who rated onboarding ≥ 4 had 73% Day-30 retention vs. 31% for users who rated ≤ 2. The immediate live-agent connection for low-onboarding-ease users improved their Day-30 retention to 52% — a 68% relative improvement. For onboarding cost considerations, see our onboarding costs analysis.

Feature feedback linked to real user profiles

Product teams running feature validation surveys through Intercom get feedback enriched with usage context. A survey asking “How useful would [Feature X] be for your workflow?” (1-5 scale + optional comment) syncs the response to the user profile.

The product manager can now:

• Segment demand by plan tier. If 78% of enterprise users want Feature X vs. 34% of free-tier users, the prioritization calculus changes.
• Cross-reference with usage data. Users who actively use the current workaround for Feature X and rate it 5 are the most credible demand signal.
• Follow up with interested users. When Feature X ships, an outbound message targets users who rated it 4-5 — they become early adopters and provide launch-day feedback.
• Read comments with full context. Each comment is attached to a user with visible plan tier, usage patterns, conversation history, and support interactions. Feedback without context is noise; feedback with context is signal.

Conversation routing based on survey sentiment

Intercom’s routing rules can reference survey attributes to send conversations to the right team before the first reply.

Routing logic examples:

• survey_nps_score < 7 → route to Senior Support queue. Detractors get experienced agents.
• survey_nps_score >= 9 AND plan = "enterprise" → route to Growth/Expansion team. Promoters on large plans are expansion candidates.
• onboarding_ease <= 2 AND days_since_signup < 14 → route to Onboarding Specialist queue. New users struggling with setup need specialized help, not general support.

This routing happens before the agent sees the conversation. The right expertise is applied immediately, reducing escalations and improving resolution quality. For more on customer service excellence, see our customer care guide.

Practices for survey-Intercom integration

Time surveys to interaction moments. Post-resolution, post-onboarding, post-feature-adoption — these are the moments when users have context and opinions. Random NPS emails without a trigger event produce lower response rates and more ambiguous data.

Show survey data in the agent sidebar. Configure Intercom’s user profile layout to display nps_score, last_csat_rating, and onboarding_ease prominently. Agents should see sentiment before they see the message.

Use clear attribute naming. Prefix survey attributes with survey_ to distinguish them from product event data or sales CRM data that may also appear on the profile.

Avoid survey fatigue. Intercom users often interact daily. Don’t survey after every conversation. Survey after significant interactions: resolution of a complex ticket, completion of a milestone, or at fixed intervals (quarterly NPS). Quality of responses drops when frequency is too high.