For post-ticket surveys, a 3- or 5-point scale beats binary thumbs up/down — but only if you pair it with a one-tap way to say why. Binary scales get you the highest click-through rate, but they throw away the information you actually need: the difference between "fine" and "great," and the reason behind a low score. If you're choosing a scale for the first time, start with a 5-point scale (icons or numbers, doesn't matter much) and add an optional one-click reason tag for anything below the top rating.
Here's the reasoning, because the "which scale" question keeps coming up in MSP communities and most of the advice floating around is written for SaaS product teams, not for ticket-close surveys.
Thumbs up/thumbs down is tempting because it's the lowest-friction option — one tap, done, no scale to interpret. It genuinely does produce higher raw click rates, which is why you see it used for things like "was this article helpful?" widgets.
The problem for an MSP is that a support ticket isn't a yes/no experience. A client can get their issue resolved (thumbs up) while still being irritated that it took three days and two reopens. Binary surveys collapse that nuance into a single bit, so you lose the ability to tell "the tech nailed it" apart from "it finally got fixed, barely." Over time your CSAT percentage looks great — because almost everything that gets fixed earns a thumbs-up — while your actual service quality trend is invisible.
Binary scales work well for things with a genuinely binary outcome: did this KB article solve your problem, did this appointment reminder arrive on time. A closed ticket after a multi-touch support interaction isn't that.
A 5-point scale gives you a neutral middle, room for "satisfied but not thrilled," and enough spread to actually segment technicians and ticket types over time. It's also what most PSA-native survey tools (including the one-click links in Autotask completed-ticket emails) are built around, so you're not fighting your own tooling.
The tradeoff is a slightly lower response rate than binary — in our experience the dropoff is modest, maybe a few percentage points, not the difference between a survey that works and one that doesn't. The 5-point scale costs you a little bit of friction and buys you a lot more signal. A 3-point scale (frown/neutral/smile) splits the difference: faster to answer than 5, less lossy than binary, and it's a reasonable compromise if your response rates have historically been weak.
A 1–10 scale is overkill for a single ticket interaction. That granularity makes sense for relationship-level NPS, not "how did we do on this one issue." Ten options on a one-click email survey mostly just add decision fatigue without adding real insight — nobody can reliably distinguish a 7 from an 8 on a support ticket.
Whatever scale you pick, the single highest-leverage change you can make is adding a short, optional second step: a one-tap reason code (slow response, didn't fully fix it, technician communication, other) that only appears after someone picks a middling or low score. This is what turns a CSAT number into something actionable. A 3-out-of-5 tells you a client was unhappy; it doesn't tell you if that was about speed, communication, or a fix that didn't stick. Keep this step optional and genuinely one-click — a required open-text box will tank your response rate faster than any scale choice will.
The actual mistake to avoid isn't picking "the wrong" scale — it's switching scales partway through the year, or running different scales for different ticket types. CSAT is only useful as a trend line, and a trend line breaks the moment you change how you're measuring it. If you've been running 1-5 for two years, don't switch to binary because a vendor's default template uses it; you'll lose your ability to compare this quarter to last year.
If you're setting this up for the first time on InsideLoop, the default we recommend is a 5-point one-click scale embedded directly in the Autotask ticket-completed email, with an optional reason tag triggered on anything below a 4. That combination gets you response rates in a reasonable range for ticket surveys, keeps the data segmentable by technician and ticket type, and gives you an actual reason to act on when a score comes in low — which is the entire point of running the survey in the first place.
InsideLoop embeds a one-click satisfaction survey in your ticket-close emails, writes ratings back to the ticket, and alerts you the moment a client is unhappy.