A prospective client calls at 4:40pm on a Friday, gets voicemail, and calls the next firm on the list. Somebody else fills out your website form at 11pm and hears from you on Tuesday. Meanwhile your paralegal is spending two hours a day fielding calls from people who will never be clients, and your conflicts check happens somewhere between the initial call and the engagement letter, depending on who remembers.

That is the actual problem most firms are trying to solve when they start searching for legal intake services. It is rarely a lead generation problem. It is a response time problem, a consistency problem, and a data capture problem, and those three things get tangled together until the firm concludes it needs to buy something. Understanding which of the three is actually broken determines whether the money you spend does anything at all.

What you are actually buying

The phrase covers at least four different products that solve different problems.

There are answering services and call centres staffed by humans who pick up the phone using a script you provide. There is intake software, meaning form builders, CRM pipelines, and automated follow up sequences, which is where Lawmatics, Clio Grow, and MyCase live. There are conversational AI tools that answer calls or web chats and try to qualify before handing off. And there is custom automation, meaning the plumbing that connects your phone system, your website, your CRM, and your case management platform so that a single inquiry moves through all of them without anyone retyping anything.

Firms often buy the first one when they needed the fourth. An answering service that takes a message and emails it to a general inbox has solved the after hours coverage gap and nothing else. The message still lands somewhere a human has to notice it, still gets manually entered into Filevine or Clio, and still depends on someone deciding it is worth following up. You have moved the bottleneck rather than removed it.

Speed to first contact is the clearest win. If your current state is that inquiries sit until a human reads them, almost any structured system will beat that. Automated acknowledgement within a minute, a calendar link for a consultation, and an escalation rule for high value or time sensitive matters are all straightforward to build and hard to argue with. For personal injury and employment work in California, where statutes of limitations and administrative filing windows can be short and government claim deadlines shorter still, being slow to reach someone is not just a conversion issue.

Consistency of information capture is the second win. Every practice area has a set of facts you need before an attorney can sensibly evaluate a matter. Date of incident, insurance carrier, whether the person has spoken to anyone else, prior representation, opposing party names. A structured intake form or a scripted call collects the same fields every time, in the same format, so that when an attorney does look at the file, the evaluation takes eight minutes rather than a callback and a second conversation.

Third, routing and record keeping. A well built intake flow can run a preliminary conflicts search against your existing client and adverse party list, timestamp the inquiry, log the source, and create the matter record automatically. That last part matters more than it sounds. When intake data lives in the same system as the matter, you can actually answer questions like which referral sources produce clients who stay, and which practice areas are consuming intake time without producing work.

Here is a concrete version. A wage and hour inquiry comes through the website at 9:15pm. The form captures employer name, job title, pay period details, approximate dates of employment, and whether the person has filed anything with the Labor Commissioner. An automated response goes out confirming receipt and stating plainly that no attorney client relationship exists yet. The employer name is checked against the firm's adverse party list, and because it matches nothing, the inquiry is created as a lead in Lawmatics with the practice area tagged. At 8:30am the intake coordinator sees it at the top of a queue sorted by claim urgency, calls, and books a consultation for that afternoon. Total human time before the consultation, roughly six minutes. If the employer name had matched, the record would have been flagged and held for attorney review rather than progressing.

What it does not fix

It does not evaluate a case. A qualification script can capture facts and apply rules you wrote, such as flagging anything where the incident date is more than eighteen months old. It cannot tell you whether a matter is worth taking. That judgment involves damages, credibility, venue, the opposing party's insurer, your own capacity, and a dozen things a form cannot see. Any vendor implying otherwise is selling something you should not buy, and under the California Rules of Professional Conduct the responsibility for that assessment and for supervising the non lawyers and systems involved stays with the attorney.

It does not fix a firm that does not follow up. If your existing problem is that leads go cold because nobody works the pipeline, adding a tool that generates more organised leads produces a more organised backlog. Automation makes an existing process faster. It does not create a process.

It does not solve conflicts. Automated matching against a name list catches exact matches and obvious variants. It misses corporate parents and subsidiaries, married names, DBAs, and positional conflicts entirely. Treat automated conflicts checking as a first pass that reduces attorney workload, never as the check itself.

AI voice and chat tools deserve specific caution. They work reasonably well for structured intake on straightforward matters where the caller is calm and the questions are factual. They handle a distressed caller poorly, they mishear names and dates, and they will occasionally state something with confidence that is wrong. If you deploy one, transcribe and review calls, disclose that the caller is speaking to an automated system, and give an obvious route to a human. Do not let it discuss the merits of a matter.

The data question California firms should not skip

Intake data is often the most sensitive information a firm holds about people who are not yet clients. Medical details, immigration status, employment disputes, family matters. Under the CCPA, as amended by the CPRA, a firm meeting the applicability thresholds has obligations around disclosure, deletion, and the treatment of sensitive personal information, and those obligations extend to how your vendors handle the data.

Before signing anything, ask where the data is stored, whether it is used to train models, who at the vendor can access it, how deletion requests are handled, and whether the contract includes the service provider terms the CPRA requires. Ask for the answers in writing. A vendor that cannot produce a data processing addendum is telling you something.

How to judge the options honestly

Ignore conversion rate claims. You cannot verify them and they depend entirely on the firm's practice area and market. Judge instead on three things.

  • Integration depth. Does it write into your case management system with real field mapping, or does it email a PDF? A one way email is not an integration.
  • Configurability. Can you change the qualification questions and routing rules yourself, or does every change require a support ticket and a week?
  • Escalation behaviour. What happens when the system encounters something it was not designed for? Silent failure is the expensive failure mode.

Run a two week parallel test before committing. Route half your inquiries through the new process and keep the other half on your current one. Compare time to first human contact, completeness of the captured record, and how many inquiries required rework. Two weeks of your own data beats any case study.

Where to start

Before you buy anything, spend an hour mapping what happens to an inquiry today. Write down every step from first contact to signed engagement letter, note who touches it and how long each step waits, and mark the points where information gets retyped. Most firms find one or two specific delays that account for most of the damage, and those are usually fixable without a new platform.

Then decide whether you need coverage, structure, or plumbing. Coverage means humans answering phones. Structure means better forms and follow up sequences. Plumbing means connecting systems you already pay for. Buying the wrong one of the three is the most common way firms waste money on legal intake services. If the answer turns out to be plumbing, that is the kind of work Alphovia builds, but the mapping exercise comes first either way.