Legal Client Satisfaction Surveys: What to Actually Ask

Every managing partner says client feedback matters, and most firms still run their satisfaction survey the same tired way, a generic email that goes out weeks after the matter closes asking someone to rate their experience from one to five with no real explanation of what the number is even supposed to measure. The response rate sits around ten percent, the answers cluster around four, and nobody at the firm can point to a single decision that survey data has ever actually influenced. That is not feedback, it is a compliance ritual dressed up as client care.

A survey that actually earns its place in a firm's operations does something narrower and more useful. It asks specific questions at the right moment, tied to something the firm can genuinely change, and it routes the answers to a person who is accountable for acting on them. Firms that get this right are not necessarily running fancier software, they are asking better questions at better times and treating the results as operational data rather than a marketing checkbox.

This piece walks through what to actually ask, when to ask it, how to segment the questions by practice area, and what a firm should do with the answers once they land, whether that answer is a five or a two. None of this requires an expensive survey platform. It requires knowing what you are trying to learn and building the process around that, not around a template someone found online.

Why the standard post-matter survey doesn't work

The core problem with most legal satisfaction surveys is that they ask about the wrong thing at the wrong altitude. "How satisfied were you with our services, one to ten" tells a firm almost nothing, because it collapses six or seven genuinely different experiences, communication, outcome, billing clarity, responsiveness, staff friendliness, into a single undifferentiated number. A client can be thrilled with their attorney and furious about a confusing invoice, and a single blended score hides that entirely.

The second problem is timing. Surveys that go out a month after the matter closes catch a client who has already moved on emotionally, whose memory of the actual day to day experience has flattened into a vague general impression instead of specific recollections. A client asked about their experience three days after a matter closes remembers the paralegal who returned every call within the hour. Ask them a month later and you get a shrug and a default four out of five, because the specifics have already faded and all that is left is a rough overall vibe.

Timing the survey to the matter, not the calendar

The right moment to survey a client is tied to what actually happened in the matter, not to a fixed number of days on a calendar. A matter that closes cleanly after a straightforward transaction should trigger a survey within a few days of closing, while the experience is still fresh and specific. A long litigation matter benefits from a mid-matter check-in as well as a close-out survey, since waiting until the very end means a firm only learns about a communication problem that has been building for eighteen months after it is too late to fix it for that particular client.

This is where tying the survey trigger to the actual stage of the matter rather than a generic date makes a real difference. In Casely, the matter stage tracker sits at the top of every case file as a configurable stepper, and a firm can set its own stages per practice area, so the natural trigger point for a satisfaction survey becomes the moment a matter actually moves into a closed or resolved stage rather than an arbitrary date pulled from a spreadsheet. A transactional practice group might rename that final stage "Closing Complete," a litigation group might use "Judgment Entered," and either way the survey fires off the real event, not a guess at when that event probably happened.

The core question set: what actually predicts a referral

If a firm can only ask five questions, these are the five worth asking. The first is whether the client felt informed about what was happening in their matter at each stage, since research on legal client experience consistently finds that perceived communication frequency, not case outcome, is the strongest predictor of whether a client refers a firm to someone else. The second is whether the client understood what they were being billed for and why, a question that surfaces billing confusion long before it turns into a dispute or a slow-paying account.

The third question worth asking directly is whether the client felt their questions were answered promptly, phrased around a specific behavior rather than a vague feeling, since "did we respond quickly enough" gets a more honest and more actionable answer than "were you satisfied with our responsiveness." The fourth is whether the client would use the firm again for a different legal matter, which measures trust in the relationship independent of this specific outcome. The fifth, and the one firms skip most often because it feels presumptuous to ask, is whether the client would be willing to refer someone they know, worded plainly rather than dressed up in corporate survey language.

NPS, CSAT, or building your own scale

Net Promoter Score, the "how likely are you to recommend us, zero to ten" question, works reasonably well for legal because referrals are genuinely how most firms grow, and the question directly measures the behavior that matters most. The downside is that NPS alone tells you nothing about why a score is a six instead of a nine, so it needs a follow-up question attached, not left as a standalone metric floating on a dashboard somewhere nobody looks at it.

Customer Satisfaction Score, a straightforward one-to-five or one-to-ten rating on a specific interaction, works better for measuring a single touchpoint, a particular intake call, a specific hearing, a document turnaround. The mistake firms make is picking one scale and forcing it onto every kind of question, when NPS suits a relationship-level survey at matter close and CSAT suits a narrower, in-the-moment check after a specific interaction. A firm that blends the two, a relationship-level NPS survey at close paired with short one-question CSAT pulses after key milestones, gets a fuller picture than either one alone.

Communication and responsiveness: the two questions that matter most

Ask about communication frequency separately from communication clarity, because they measure different things and firms often score well on one and poorly on the other. A firm can update a client every single week and still leave them confused if the updates are full of jargon and short on plain explanation of what actually changed and what it means for them. A firm can also explain things clearly but too rarely, leaving long silent stretches where the client assumes nothing is happening even when substantial work is underway behind the scenes.

Billing clarity deserves its own dedicated question because it is one of the most common sources of quiet dissatisfaction that never shows up as a complaint, it just shows up as a client who does not come back. Ask specifically whether the invoice matched what the client expected to see, not just whether they were satisfied with the amount. In Casely, turning a matter's billed time into an invoice is a one-click action that pulls every unbilled hour into a single itemized draft, which means the invoice a client actually receives reflects real, specific entries rather than a lump sum, and a clear itemized bill tends to generate fewer confused calls and a more honest answer to this particular survey question in the first place.

Asking about privilege and confidentiality without making it awkward

A survey question should never ask a client to restate or evaluate the substance of privileged legal advice, both because it puts the client in an odd position and because a poorly secured survey tool storing that kind of detail is a real liability if it is ever compromised. Keep every question focused on the experience of working with the firm, communication, responsiveness, clarity, and never on the specifics of legal strategy, case facts, or the advice itself. This sounds obvious written out, but it is a genuinely easy line to cross when someone is drafting survey questions quickly and reaches for "were you satisfied with our handling of the settlement negotiation" instead of the safer, more useful "did you feel informed throughout the negotiation process."

Delivering the survey through a channel the client already trusts with sensitive information also matters more than most firms consider. In Casely, the client portal already gives each client a filtered, real-time view of their own matter, and privilege filtering on documents happens automatically at the tag level rather than through manual review, so a firm that surfaces a satisfaction survey inside that same portal is asking for feedback through a channel the client has already been using to review real case information, on a phone or a desktop, without creating a new account or a new login just to answer a few questions.

Segmenting the questions by practice area and matter type

A family law client and a corporate transactional client are not measuring the same firm against the same yardstick, and a single generic survey sent to both will produce answers that are technically complete and practically useless when a partner tries to compare them. A family law client weighs empathy and emotional steadiness heavily, so a question about whether staff treated them with patience and respect during a difficult time carries real signal. A corporate client closing an acquisition cares far more about speed, clarity of next steps, and whether deadlines were hit, so a question about whether the timeline matched expectations matters more there than a question about emotional tone.

Referral sources deserve their own separate, shorter survey entirely, since a referring attorney or a repeat referral partner is not the client and should not be asked client-facing questions about the matter itself. They should be asked whether the referral was handled promptly, whether they were kept in the loop as appropriate, and whether they would refer again. Tracking who actually sends referrals over time, and following up with that specific group differently than with clients, is exactly the kind of segmentation that gets lost when a firm treats "contacts" as one undifferentiated list. In Casely, contact labels let a firm tag a person's specific role on a matter, referral source among them, and see that pattern building across matters over months and years rather than relying on someone's memory of who tends to send business their way.

Closing the loop: what to do with a bad score

A survey program without a defined process for handling a bad score is not actually a feedback system, it is a data collection exercise that occasionally produces uncomfortable numbers nobody follows up on. Every response below a defined threshold, a six or lower on a ten-point scale for instance, should trigger an automatic notification to a specific named person, usually the responsible partner or a designated client experience lead, within 24 to 48 hours, not whenever someone happens to review the spreadsheet next.

The follow-up itself matters as much as the speed. A generic apology email accomplishes little. A short, specific phone call that references the actual concern the client raised, and that ends with a concrete next step rather than a vague promise to "do better," is what actually repairs the relationship and, just as importantly, teaches the firm something specific about what went wrong. Log the resolution against the matter so the next person who touches that client relationship, an intake coordinator handling a new matter for the same client a year later, for instance, has the context rather than starting from zero.

Turning good scores into referrals and marketing proof

A high score is not the end of the process, it is the beginning of a second, smaller ask. A client who rates their experience a nine or a ten and says yes to a follow-up question about willingness to refer is exactly the person worth asking, at that moment, whether they would be open to a short testimonial or an online review, while the positive experience is fresh and specific rather than a vague memory months later. Waiting until quarter-end to compile a list of happy clients and asking then produces a much lower response rate than asking in the moment.

Trend data across a full year of survey responses is where the real strategic value shows up, not in any single survey. A firm that reviews scores by practice area and by individual attorney over time can spot patterns that no single response would reveal on its own, a particular practice group consistently scoring lower on communication regardless of who staffs the matter, for instance, which points to a process problem rather than a person problem. That kind of pattern only becomes visible when a firm actually looks at the aggregate, which is a discipline most firms intend to build and few actually maintain past the first quarter.

Common mistakes firms make with client feedback

The most common mistake is surveying too late, after the specifics have faded into a generic impression, followed closely by asking too many questions and watching completion rates collapse past the third or fourth item. A ten-question survey with a five percent completion rate produces less usable data than a three-question survey with a sixty percent completion rate, even though the longer survey looks more thorough sitting in a folder somewhere.

The second cluster of mistakes centers on ownership and follow-through. A survey with no named owner for reviewing responses becomes a survey nobody reads, and a firm that never closes the loop with clients who gave a low score trains its own clients to stop bothering with honest answers, since nothing ever visibly changes as a result. Treating the survey as a marketing exercise rather than a genuine diagnostic tool is the deepest version of this mistake, since a firm that only wants a good number to put on its website will unconsciously design questions that are more likely to produce one, and the whole exercise stops teaching the firm anything true about how it is actually doing.

Getting client feedback live at your firm

Start smaller than feels comfortable. Pick one practice area, draft three to five specific questions tied to that practice area's actual client experience, and tie the send trigger to a real event in the matter rather than a calendar date. Run it for a full quarter before adding more complexity, and resist the urge to build an elaborate multi-branch survey logic tree before you have even confirmed that clients will answer three good questions in the first place.

The mechanics matter less than most firms assume, a plain email with three sharp questions beats an elaborate survey platform with generic ones every time. What matters is tying the trigger to a real stage in the matter, routing every response to a specific accountable person, and actually closing the loop on the low scores instead of letting them sit in an inbox. A firm running its matters through a system where the stage tracker, the client portal, and the referral history already live in one place has a much easier time building this kind of survey program on top of infrastructure that already exists, rather than bolting a separate tool onto a process that was never designed to talk to it. If you are still deciding how client-facing visibility should work at your firm before you even get to the survey question, our client portal software page walks through what that actually looks like end to end, and if referral tracking is the piece you are missing, our legal intake software page covers how a firm keeps that data connected to the rest of the matter instead of scattered across someone's memory.