The client intake process is everything between a prospect first contacting your firm and a signed engagement letter: capture the lead, screen it, run a conflict check, evaluate fit and scope, agree on fees, and get the agreement signed. Intake ends the moment the engagement letter is executed. Onboarding starts there. Firms that blur those two lines end up doing real work for clients they never formally accepted.
Last updated: July 2026.
Intake is where a firm decides who it works for. That is a strange thing to run informally, and yet most firms do, because intake feels like admin and gets pushed to whoever answers the phone. The cost surfaces later, as the client who was never a fit, the matter that conflicts with an existing one, the scope nobody wrote down, and the invoice that gets disputed in month three.
A written intake process is not bureaucracy. It is the point where you can still say no cheaply.
Intake vs onboarding vs sales
These three overlap in time and are distinct in purpose, and naming them properly is most of the fix.
| Stage | Question it answers | Starts | Ends | Owner |
|---|---|---|---|---|
| Sales or business development | Should this person consider us | First touch, marketing, referral | Prospect asks to work with you | Partner, principal, BD |
| Intake | Should we take this client, on what terms | Inbound inquiry | Engagement letter signed | Intake coordinator, with partner sign-off |
| Onboarding | How do we start the work well | Engagement letter signed | Client is set up and the work is underway | Delivery team |
The boundary that matters is the engagement letter. Before it, you are deciding. After it, you are delivering. Work performed in between is work you may not be able to bill for and, in a law firm, may have created duties you never intended to take on. Everything downstream of the signature, the welcome, the systems access, the kickoff, belongs to your customer onboarding process and should not be running before you have accepted the engagement.
The client intake process, step by step
Eight steps. The owners and turnaround times below are a starting point to adapt, not a standard. What matters is that every step has one name against it and one clock.
| # | Step | Owner | Target turnaround | Output |
|---|---|---|---|---|
| 1 | Lead capture | Whoever receives it, into one system | Immediate | A record. Name, matter, source, timestamp |
| 2 | First response | Intake coordinator | Same business day | Contact made, intake form sent |
| 3 | Pre-screen | Intake coordinator | 1 day | In or out of your practice area and capacity |
| 4 | Conflict check | Conflicts staff or designated partner | 1 to 2 days | Clear, waivable, or declined |
| 5 | Fit and scope evaluation | Partner or principal | 2 to 3 days | Scope, exclusions, staffing, risk flags |
| 6 | Fee agreement | Partner | Same meeting where possible | Fee structure, retainer, billing terms |
| 7 | Engagement letter | Partner drafts, client signs | 2 days to send, then track | Executed agreement |
| 8 | Handoff to onboarding | Intake to delivery | 1 day | Client record, scope, budget, deadlines transferred |
Step 8 is the one firms skip. Intake accumulates context, the client's real problem, the deadline they mentioned in passing, the fact that their controller is the actual decision maker, and then that context evaporates because it lived in one person's inbox. Write the handoff down. It takes ten minutes and it is the difference between a client repeating themselves in the kickoff call and a client who feels heard.
The four decision points in a client intake flow chart
Draw the flow and you find it is not eight boxes in a line. It is eight boxes and four diamonds, and each diamond has an exit that ends the process. A flow chart without exits is a wish, not a process.
- Is it in scope? Practice area, jurisdiction, capacity, minimum engagement size. Exit: decline politely, refer out. A good referral out is a marketing asset.
- Does it conflict? Exit: decline, or seek informed written consent where the conflict is waivable.
- Do we want it? Fit, risk, ability to pay, how the prospect has behaved so far. Exit: decline. This is the diamond firms pretend does not exist.
- Do they accept the terms? Exit: negotiate or part ways. A client who negotiates the engagement letter aggressively will negotiate every invoice.
Time the decline. Declining in step 3 costs you an email. Declining in step 7, after a partner has spent four hours scoping, costs real money, and by then sunk cost makes the decline unlikely to happen at all. Push every reason for a no as early in the flow as it can honestly be evaluated.
Conflict checks: what the rules actually say
For US law firms this is not discretionary. ABA Model Rule 1.7 bars representation directly adverse to a current client, or materially limited by another responsibility, absent informed consent confirmed in writing. Model Rule 1.9 bars representation against a former client in the same or a substantially related matter. Model Rule 1.10 imputes one lawyer's conflict to everyone else in the firm, which is why the check has to run against the entire firm's history, not the individual lawyer's memory. Your state's rules govern, and states vary from the model rules, so check yours rather than these.
Practically, a conflict check is only as good as its database. It has to run against current clients, former clients, adverse parties, related entities, and the individuals behind them. Nickname and entity-name variations defeat exact-match searches routinely: the same company appears as Acme Corp, Acme Corporation, and ACME Corp Holdings across three matters, and a search for one finds none of the others. Search names, aliases, and subsidiaries, and record the result, the date, and who ran it, for every prospect, including the ones you decline. The declined ones are exactly the ones that come back through a different door.
Accounting firms are not under the ABA rules, but the AICPA Code of Professional Conduct addresses conflicts of interest and independence, and independence rules bite hardest on attest engagements. A bookkeeping client who is a competitor of an audit client is a conversation to have before, not after.
Run the conflict check before you give substantive advice. Advice given at the inquiry stage can be the thing that creates the relationship you were checking for.
Designing the client intake form
The most common intake form mistake is asking for everything. A prospect who has not yet decided to hire you will not fill in a 40-field form, and the data you collect from someone who has not committed is mostly unreliable anyway.
Split it in two. The inquiry form captures only what you need to route and screen: who they are, how to reach them, one paragraph on what they need, how urgent it is, how they found you, and any names involved so you can start the conflict check. That is six or seven fields. Everything else, entity details, financial data, historical documents, belongs in the post-signature intake, where the client is committed and will actually complete it. That deeper set of questions is a design problem in its own right, and worth building carefully as a structured client onboarding questionnaire rather than an ad hoc email.
Three things earn their place on the short form. Ask the names of all parties involved, because that feeds the conflict check and costs the prospect nothing. Ask about deadlines, because a real deadline is the strongest qualification signal there is. And ask how they found you, because six months of that answer tells you where to spend.
Then send the deeper requests through a tracked channel rather than email, so that each requested document has a state and someone can see what is outstanding without opening a mailbox. That is precisely what client portal software exists to do, and intake is where its value shows up first.
The engagement letter: scope, and the absence of scope
An engagement letter sets out who the client is, what you will do, what you will not do, what it costs, how and when you bill, and how the relationship ends. The clause that prevents the most disputes is the one listing what is out of scope, because scope disputes are almost never about what the letter said. They are about what the client assumed was included.
Write the exclusions in the client's language, not yours. "Does not include representation in any appeal" is clear. "Limited to the matter described herein" is not. For an accounting engagement, spelling out that monthly bookkeeping does not include tax planning, audit support, or advice on the client's financial decisions prevents the conversation where a client discovers in April that a service they assumed they were buying was never being provided.
Name the client precisely. Representing a company is not representing its founder, and confusion about which one you act for is a conflict waiting for a bad quarter.
Then send it for signature and track it as an open item with a deadline, because an unsigned engagement letter is the single most common place intake stalls. A partner sends it, the prospect means to sign it, three weeks pass, and work has quietly begun anyway. Signature friction is a solved problem, and the argument for solving it is the same one that applies everywhere else in the firm: every day a document waits for ink is a day of contract friction that costs you more than the software would.
Measuring intake
Four numbers. Track them monthly and the process manages itself.
| Metric | Definition | What a bad number tells you |
|---|---|---|
| Time to first response | Inquiry received to a human reply | Rising means leads are dying in a queue. Measure it in hours |
| Intake conversion rate | Engagement letters signed divided by qualified inquiries | Low means bad lead sources or a slow, leaky process. Split by source before blaming the process |
| Time to engagement letter | Inquiry to executed agreement | Long means bottlenecks, usually one partner and one conflict check |
| Decline reasons, categorized | Why you said no, by reason | The most valuable and least tracked. Mostly conflicts means the database. Mostly out of scope means your marketing is attracting the wrong work |
Do not benchmark these against a number in an article, including this one. The distribution is enormous across firm types and practice areas, and a personal injury firm and a corporate tax practice have nothing useful to say to each other about intake conversion. Benchmark against your own last quarter.
Resist the temptation to optimize conversion rate alone. A firm can drive intake conversion to ninety percent by accepting everyone, and will feel the consequences for years. The metric that keeps the others honest is what fraction of the clients you accepted this year you would accept again.
Where intake breaks
Six failures, in rough order of how often they appear.
No single system of record. Inquiries arrive by phone, email, web form, and referral, and land in four places. Nobody can say how many leads came in last month. Everything else on this list is downstream of this one.
Response time depends on who answers. The prospect who reaches a partner gets a call back the same hour. The one who fills in the web form waits four days. Same firm, two experiences, and the second one has already hired someone else.
Conflict checks run late, or from memory. A partner who has been at the firm eleven years is confident there is no conflict. Model Rule 1.10 does not care about confidence.
Nobody is allowed to say no. If declining requires a partner meeting, the firm accepts bad clients by default, because accepting requires nothing.
The engagement letter sits unsigned while work begins. Fix it with tracked signature requests and a hard internal rule that no billable time is opened against an unsigned matter.
Context dies at the handoff. The delivery team asks questions the client already answered during intake. To the client, this is the first evidence that your firm does not talk to itself, and first impressions in the paperwork stage set expectations for everything after, which is why onboarding paperwork is a first impression, not a formality.
Start here
You do not need software to fix intake, and buying it first is how firms end up with an expensive version of the same mess. Do this instead, in order. Put every inquiry into one system, whatever it is, this week. Write the eight steps down with a name against each. Move the conflict check earlier than feels comfortable. Cut the inquiry form to seven fields. Set a same-day response standard and actually measure it.
Then, once the process is written and stable, automate the parts of it that are repetitive: the form, the reminders, the signature request, the handoff. Automating a process you have never written down just makes the confusion faster. Once the letter is signed, hand off cleanly into a real client onboarding checklist, and pick your systems afterward, because the tooling decision is much easier when you already know exactly what the process is asking the tool to do. The same principle governs every other paperwork workflow in back office customer operations: the process first, the software second, and never the reverse.