Freshdesk CSAT & Satisfaction Surveys Explained

What a Freshdesk CSAT survey is measuring

CSAT — customer satisfaction — is a transactional metric. Unlike NPS, which asks how someone feels about your brand overall, a CSAT survey fires at the end of a specific support interaction and asks the customer to rate that experience. Freshdesk's survey engine is built around a single primary question ("How satisfied are you with your support experience today?") that the customer answers on a rating scale, with the option to attach follow-up questions and a free-text box for comments.

The score itself is a percentage. As Freshworks documents in its Customer Satisfaction Survey report, Freshdesk buckets every response into positive, neutral, or negative, and the headline CSAT rating is the percentage of responses that landed in the positive bucket. On a 7-point scale that split works out to three positive gradients, one neutral, and three negative — so the scale you pick directly shapes how forgiving or harsh your score reads.

Rating types and point scales

Freshdesk gives you two independent choices: what the rating looks like, and how many steps it has.

For the visual rating type, you can present the question as emoji/smiley faces, stars, a numbered scale, or a text-based choice. For the point scale, Freshdesk supports 2, 3, 5, or 7 points, with 3 as the default. A 2-point scale is a blunt thumbs-up/thumbs-down; a 5- or 7-point scale gives you more gradient to detect "fine but not great" from "genuinely happy."

• 2-point — happy / unhappy. Highest completion rate, least nuance.
• 3-point (default) — negative / neutral / positive. A reasonable balance.
• 5-point — the classic "very dissatisfied → very satisfied" Likert range.
• 7-point — finest granularity; best when you have the volume to make small gradients meaningful.

One constraint worth knowing up front: if you add extra questions beyond the primary one, per Freshworks' survey setup documentation they must all use the same point scale and the same choices as the main question. You can't mix a 5-point primary with a yes/no follow-up on the same survey.

How to build a survey, step by step

In the upgraded CSAT module — documented in Setting up surveys in the upgraded Freshdesk CSAT module — everything lives under one admin path.

1. Go to Admin → Workflows → Customer Satisfaction and click New Survey.
2. On the Build survey tab, name the survey and enter your primary question.
3. Pick a rating type (emoji, stars, numbered) and a point scale (2 / 3 / 5 / 7).
4. Optionally add follow-up questions — multi-choice, rating, yes/no, short text, or long text — remembering they inherit the primary point scale.
5. Switch to the Collect Responses tab to choose channels, triggers, and frequency (see below).
6. Save — then flip the ON/OFF toggle next to the survey to make it live. Saving alone does not activate a survey.

Which channels fire a survey, and when

On the Collect Responses tab you choose the delivery channels. Freshdesk can send CSAT surveys over Email, Web Chat, WhatsApp, Instagram, and Facebook Messenger — so the same survey can meet a customer wherever the conversation actually happened.

Timing is the part people trip on. The most common trigger is ticket resolved or closed, but the survey only actually goes out if the matching requester email notification is switched on. For an email survey, you need to enable, under Admin → Workflows → Email Notifications → Requester Notifications, either the "Agent solves the ticket" or "Agent closes the ticket" notification — the survey rides along inside that email. If that notification is off, the trigger fires into the void and nobody gets a survey. You can also attach a survey manually from the reply editor, or drive sends with automation rules for more specific conditions.

Plan gating and account limits

Two limits are worth committing to memory. First, an account can create up to 10 satisfaction surveys, but only one can be active at any given time — so CSAT is effectively one survey per account, with nine on the bench. Second, survey frequency capping — the "send at most one survey every N requests per day/week/month to the same contact" control that stops your best customers being surveyed to death — is available only on the Enterprise plan.

Because plan names and gating shift over time — and Freshdesk Omni packages this differently again — confirm the exact tier against your own account before you promise a stakeholder a feature.

Where CSAT scores surface

Once responses come in, the numbers live in Analytics → Curated Reports → Customer Satisfaction Survey Results. The report is more than a single headline figure: it shows the positive / neutral / negative distribution, breaks satisfaction down by agent and by group, and correlates scores against response time, resolution time, ticket reopens, and the number of back-and-forth interactions. A "Show tabular data" drilldown lets you read the individual tickets and the free-text comments behind any rating — which is where the actual, actionable feedback usually hides. Individual responses also show up on the ticket itself, so an agent can see exactly what a customer said about a conversation they handled.

The honest limits — and where an AI layer picks up

Freshdesk CSAT is a solid, native feedback loop, and for most teams it's all the survey tooling they need. But be clear about what it is and isn't. It measures satisfaction; it doesn't change it. A survey is a rear-view mirror — by the time a customer rates a two-day, three-reopen ticket a 2 out of 5, the damage is done. The score is a diagnosis, not a treatment.

There are sharper edges, too. Only one survey can be active at a time, so you can't easily run different questions for different segments without swapping surveys in and out. Response rates for CSAT are notoriously low across every help desk, which means small volumes and noisy, easily-skewed averages. Frequency capping is Enterprise-gated, so on lower tiers a prolific customer can get surveyed on every single ticket. And the report tells you that satisfaction dipped for a group last week — it can't tell you why, or read the 300 comments to find the pattern for you.

That "why," and the work of preventing low scores in the first place, is where an AI agent layer fits. The category of AI agents for customer service exists to do the reasoning-heavy work a survey can't — and it's worth understanding the build-versus-buy tradeoff before you reach for one. Macha is one such layer: it runs on top of the Freshdesk you already use as a native connector — it does not replace your help desk or its CSAT module. You connect Macha to Freshdesk with your subdomain and API key, and it works the same tickets your survey is about to rate: drafting grounded first replies (often reusing the same approved snippets behind your canned responses), resolving routine questions faster, and looking up order or account status through a custom tool so the customer gets a real answer instead of a stall. The upstream fix — faster, more accurate resolutions — is what actually moves the CSAT number the survey reports.

There's a read side, too: because Macha can automate work across Freshdesk with AI, the same connector that resolves tickets can help triage and summarise the free-text comments a survey collects, turning a wall of one-line gripes into themes a team lead can act on. (Macha's connector is for Freshdesk specifically — not Freshchat, Freshservice, or Freshcaller. And credits are consumed per AI action, not per resolution or per survey — see the pricing breakdown.)

The clean division of labour: keep Freshdesk CSAT as the source of truth for how satisfied customers are, and layer an agent on top for the part a survey can't do — improving the experience before the rating, and making sense of the comments after it.

FAQ

Where do I set up CSAT surveys in Freshdesk? Go to Admin → Workflows → Customer Satisfaction and click New Survey. Build the question and rating on the Build survey tab, configure channels and triggers on the Collect Responses tab, save, and flip the ON/OFF toggle to make it live. Saving alone does not activate the survey.

What point scales does Freshdesk CSAT support? You can choose a 2-, 3-, 5-, or 7-point scale, with 3 as the default, and present it as emoji/smiley faces, stars, or a numbered scale. Any additional questions you add must use the same point scale as the primary question.

When does a CSAT survey actually get sent? Most commonly when a ticket is resolved or closed — but only if the matching requester notification ("Agent solves the ticket" or "Agent closes the ticket") is enabled under Admin → Workflows → Email Notifications → Requester Notifications, because the email survey rides inside that notification. Surveys can also go out over Web Chat, WhatsApp, Instagram, or Facebook Messenger, or be attached manually.

How many surveys can I run at once? An account can create up to 10 surveys, but only one can be active at a time. Survey-frequency capping (limiting how often the same contact is surveyed) is available only on the Enterprise plan.

Where do CSAT scores show up? In Analytics → Curated Reports → Customer Satisfaction Survey Results, which shows the positive/neutral/negative split, per-agent and per-group breakdowns, correlations with response and resolution time, and a tabular drilldown into individual tickets and comments. The CSAT score itself is the percentage of responses in the positive bucket.

Can I add AI to Freshdesk CSAT without replacing Freshdesk? Yes. An AI agent layer like Macha connects to Freshdesk as a native connector and runs on top of your existing help desk and CSAT module — it doesn't replace them. It helps improve the score by drafting grounded replies and resolving tickets faster, and can help summarise survey comments, while Freshdesk stays the system of record for satisfaction.

Ready to move the CSAT number, not just measure it? Start a free trial of Macha and connect it to your Freshdesk in minutes.