Your answered call rate is the number to fix before anything on the site
Firms redesign pages and buy keywords while a measurable share of their inbound calls ring out during business hours.
Pull the call log for the last full month and count three things: calls received, calls answered by a person, and calls where the person could actually book something. The gap between the first and the second is usually larger than anyone at the firm expects.
Look at when the misses cluster. Lunch, the hour after court, Friday afternoons and the first thirty minutes of the day are the usual patterns. Those are staffing problems with staffing solutions, not marketing problems.
Then check what happens on the calls that are answered. If the person who picks up can take a message but cannot check a conflict, look at a calendar or say what the first meeting costs, they are a switchboard rather than intake, and the caller can tell.
Fix the answered rate before touching anything else. It is the cheapest improvement available and it multiplies the value of every other channel you are paying for.
Your automatic reply is your first written work product, and it reads like a newsletter
The message that arrives thirty seconds after a form submission sets the tone for everything after it, and most firms have never read their own.
Go find yours. Submit the form and read what comes back on a phone. If it thanks the recipient for their interest, includes a logo banner and promises somebody will be in touch soon, it is doing damage at the worst possible moment.
Replace it with something a person would write. Confirm what was received, say when someone will call and from what number, say what to have ready, and give an alternative if it is urgent. Sign it from a named human being.
Include one useful thing. A short explanation of what the first meeting covers and what it costs does more to keep somebody warm than any nurture sequence, because it answers the question they were about to ask three other firms.
Make sure it actually sends and does not land in spam. Test it from a personal address on the common mail providers, not from the office network where everything is trusted.
Ask permission to text, then text, because voicemail goes unheard
A returned call from an unknown number is the least reliable way to reach the person who contacted you, and a message they can read quietly is often the only safe one.
Add a field asking how it is safe to reply and whether text is acceptable. On family matters that is not a marketing nicety, it is a practical safety question inside a household where a phone may be shared or monitored.
When permission is given, use it. A short text saying who is calling and from what number, sent immediately before the call, turns a large share of ignored calls into answered ones.
Set an attempt pattern and write it down. Something like a call and a text on the day of the inquiry, a second attempt the next morning at a different hour, a third a few days later, then a final message that closes the loop politely. Whatever the pattern, it should be the same every time so you can see whether it works.
Keep the records inside your own system. Texting from personal phones scatters the history and creates a discovery and confidentiality problem that is worth raising with your own counsel.
The caller who moved here in the fall becomes a matter in the spring
A metro fed by relocation produces a steady stream of inquiries that are real but not yet actionable, and most firms discard them on the first call.
Residency and filing requirements decide whether a case can be brought here yet, and they are worth confirming with your own counsel for each situation. What matters for intake is that a recently arrived household is often a genuine matter on a delay, not a bad lead.
Handle it deliberately. Explain plainly what generally has to be true before filing, tell them what they can usefully do in the meantime, and ask permission to follow up at a specific point rather than telling them to call back.
Then actually follow up. Put a dated task on it, in the same system as everything else, owned by a named person. Firms lose more work to forgotten future matters than to lost arguments about price.
The same discipline applies to the person who is not ready emotionally. Six months of a short, human check-in tends to be worth more than another campaign, and it costs almost nothing.
A confirmation that says what to bring does more than another reminder
No-shows are usually a preparation failure rather than a change of heart, and the message you send between booking and meeting is where it gets fixed.
Between the booking and the appointment, send one message that does real work. Where to go or how the video link works, how long it will take, what to bring, what it costs, and what will be decided by the end.
Offer the meeting in the format that matches the drive. A caller in Buda, Kyle or San Marcos looking at a late-afternoon trip north will often quietly not come, and would have kept a video appointment or an early-morning slot. Ask rather than assume.
Reduce the wait. The single most reliable predictor of attendance is how many days sit between the call and the meeting, so hold a small number of near-term slots rather than booking everything a week out.
Track attended rather than booked. A calendar full of appointments nobody kept is a reporting illusion, and it hides the exact problem you are trying to find.
You are not running a test, you are keeping a change log
One metro does not produce enough family law inquiries to detect small differences, so stop pretending and start reasoning carefully in public.
Split testing needs volume this practice area rarely has in a single market, even counting Travis, Williamson and Hays together. Anything advertised as a statistically significant win on a few dozen inquiries is noise wearing a suit.
Work sequentially instead. Change one thing at a time, write down the date, write down what you expected to happen, and give it long enough that the number of inquiries is worth looking at. Keep the log where the whole firm can see it.
Reserve real testing for changes big enough to be visible without a statistician. Paid versus free consultation. Answering the phone in the evening. A retainer structure. Those move enough to see.
For everything smaller, use evidence you already have. Listen to a handful of calls each month, read the inquiries that never converted, and ask the clients you did sign what nearly stopped them. Qualitative evidence is not a consolation prize at this volume, it is the better instrument.
Questions we actually get
- Should our consultations be free or paid?
- It is the single biggest lever in family law intake and it changes everything downstream, including who shows up and how much intake time is spent on people who will never hire you. Both models work. Pick one deliberately, run it long enough to see attendance and signing rates, and treat the switch as a firm decision rather than a marketing test.
- What conversion rate should we expect from web inquiries?
- We will not publish a benchmark, because any number we quoted would depend on the mix of matter types, the fee structure and the channel, and it would be a guess dressed as data. Measure your own for one quarter and use that as the baseline. Your own trend is more useful than someone else's average.
- Do we need a CRM, or is our practice management system enough?
- In most cases the practice management system is enough, provided every inquiry is entered whether or not it becomes a matter. The failure is usually not the software, it is that unconverted inquiries never get recorded anywhere, so nobody can see what is being lost.
- Is an after-hours answering service worth it?
- Only if it can do more than take a message. A service that can answer basic questions, book into your real calendar and pass a clean summary is a genuine improvement. One that reads a script and emails a name adds a delay without adding a conversation.
- How many times should we follow up on an inquiry that goes quiet?
- More than once and fewer than you fear. A same-day attempt, a next-morning attempt at a different hour, and one further attempt a few days later covers most of what is recoverable. Then send a polite closing message that leaves the door open, and keep the record so a call six months later is not starting from zero.
