Personal injury lead follow-up
Most personal injury firms have a first-contact process and no second-contact process. Someone calls or fills in a form, intake tries once, nobody picks up, and the lead quietly stops existing. This page is how we design the sequence that runs after that first attempt — the channel mix, the timing, what each message actually says, and the part almost nobody writes down: the rules for stopping.
On this page
- Why one attempt is where most firms stop
- Call and SMS do different jobs
- Timing, and why the first hour is not like day three
- A worked sequence
- What each message should actually say
- Stop rules
- Marking the lead, and what a "no" is worth
- The compliance question that is yours, not ours
- How we build this
- Questions firms ask
- Go deeper
Why one attempt is where most firms stop
Nobody decides to give up after one callback. It happens structurally, and the mechanism is worth naming because you can only fix what you can describe.
- The first attempt is the only one anybody is accountable for. An intake specialist working a queue is measured on whether the new lead got called. Once it has been called, it has been handled — the record shows an attempt, the task closes, and the lead moves out of the active pile without anyone deciding it is dead.
- Not answering an unknown number is normal behaviour, not disinterest. A prospective client who submitted a form an hour ago is not screening you specifically. They are doing what everyone does with an unfamiliar caller ID. Treating a missed callback as a signal of intent is reading a decision into something that was never a decision.
- Nothing schedules the second attempt. If attempt two lives in someone's memory rather than in a system, it competes with every live call, every walk-in, and every partner request that arrives between now and then. It loses that competition reliably.
- The lead never gets formally lost. Because it was never marked dead, it never appears in a report. A firm can lose a meaningful share of its inbound this way and see nothing in its numbers except a conversion rate that is quietly lower than it should be.
- The person on the other end is unusually distracted. Someone dealing with the aftermath of an accident is handling treatment, transport, work, and an insurer, often while in pain. A missed call from your firm is one of the least urgent things in their day, and that has no bearing on whether they want representation.
The fix is not more effort. It is a defined sequence that runs the same way every time, with an end. The full lead response system treats follow-up as one of four layers precisely because firms that fix response speed alone still lose the leads that did not resolve on contact one.
Call and SMS do different jobs
The two channels are not substitutes and a sequence that treats them as interchangeable wastes both. What each is actually good at:
- A call is the only channel that can complete an intake. It is the only one where a person can tell their account of what happened, be asked follow-up questions, and be booked in a single interaction. Everything else in the sequence exists to make a call happen.
- A call is also the easiest touch to ignore. It is synchronous, it demands the person be free right now, and from an unknown number it is trivially declined. High call frequency is the fastest way to look like a collections agency.
- SMS is asynchronous and identifies you. A text puts the firm's name in front of the person on their own schedule and gives them a way back in when they are ready. It also converts an unknown number into a known one, which materially changes whether the next call gets answered.
- SMS is a terrible qualification channel. Trying to run screening questions over text produces partial answers, long gaps, and a record nobody can rely on. Use it to establish contact and get to a call, not to conduct the intake.
- Email is the archive, not the sequence. It is where confirmations, directions, and anything the person needs to keep should live. Treating it as a primary follow-up channel for accident enquiries is optimistic.
The practical pattern that falls out of this: text first so the number is recognised, then call, and let every subsequent call be preceded or followed by a text that makes the number identifiable.
Timing, and why the first hour is not like day three
Follow-up sequences are usually designed with even spacing, which is the wrong shape. The value of a touch is not constant over time, and the sequence should be front-loaded to match.
- The first hour is a different activity entirely. In the first hour the person is still in the moment that made them reach out, and they are frequently still working through a list of firms. Contact here is competitive. Contact on day three is not — by then they have either retained someone or gone quiet.
- An immediate reply is a response, not an interruption. Someone who submitted a form at 11pm has just contacted you. Replying within seconds is answering them. An unprompted message at 11pm on day three is a different thing, and the sequence should treat those cases differently even though both are "an SMS at 11pm".
- Spacing should widen, not stay flat. Minutes, then hours, then a day, then two. Compressed touches early are responsive; compressed touches late are pestering, and the person experiences the difference clearly.
- Respect quiet hours for anything unprompted. Every touch after the initial acknowledgement should sit inside sending windows your firm sets. This is partly regulatory and partly obvious — nobody has ever retained a firm because it texted them at 6am.
- The whole sequence should be short. If the useful window for an accident enquiry is measured in days, a sequence that runs for weeks is not persistence. It is a system nobody switched off.
A worked sequence
This is an illustrative default — the shape we start from before a firm's own preferences, sending windows, and ethics guidance change it. It is not a template we apply unchanged, and the counts and gaps are decisions your firm makes.
| Step | Timing | Channel | Purpose |
|---|---|---|---|
| 1 | Within seconds | SMS | Acknowledge by name, identify the firm, set the expectation that a call is coming. Converts your number from unknown to known before anyone dials. |
| 2 | Within minutes | Call | The real attempt, made while the person is still engaged with the problem that made them reach out. |
| 3 | Same session, ~1 hour later | Call, then SMS if unanswered | Catches someone who was driving, in treatment, or on another call. The text gives them a way back that does not require them to be free at a particular moment. |
| 4 | Next morning, inside sending hours | Call, then SMS | The first touch of a new day, and the last one that is genuinely competitive on speed. |
| 5 | Day two, afternoon | SMS | One direct question that can be answered with a single line of text — a yes or no, not an invitation to explain. |
| 6 | Day four | Call, then SMS | The final attempt. The message says explicitly that it is the last one and how to reach the firm later. |
| Stop | After step 6 | — | The sequence ends. The lead is marked with a reason, and nothing further is sent unless the person contacts the firm again. |
Six touches over four days, front-loaded, with a stated end. A firm that runs this consistently is doing more follow-up than most of its competitors and is still finished inside a week.
What each message should actually say
The wording below is illustrative — template shapes to adapt to your firm's voice, your intake process, and whatever your ethics counsel tells you about how your firm may communicate with prospective clients. Do not deploy them as written.
Immediate acknowledgement. Name the person, name the firm, say what happens next, and give an opt-out. Something in the shape of: "[Name] — this is [Firm], confirming we received your message about your accident. Someone from our intake team is calling you now from this number. Reply STOP to opt out." The two jobs here are identifying your number and setting the expectation that a call is imminent.
After a missed first call. Acknowledge the miss without implying fault, and offer control over timing: "We just tried to reach you about your accident. If now is not a good time, reply with a time that works and we will call then." Asking for a time converts a decline into a scheduling problem, which is easier for a distracted person to solve.
Next-morning touch. Give one concrete reason to respond that is about them, not about your firm: "Following up on your message from last night. There are a few details we would need from you before we can tell you whether this is something we can help with — it takes about ten minutes on the phone." Note what this does not do: it makes no claim about the case, its merits, or its value.
Day two. One binary question. "Are you still looking for representation? Yes or no is fine either way." Making "no" an easy answer is the point — a clear no is worth more to your intake team than a fifth unanswered call.
Final message. State that it is the last one: "This is our last message — we do not want to keep bothering you. If you want to talk later, call [number] any time and ask for intake. Best of luck either way." Firms find this the hardest one to send and it is frequently the one that gets a reply, because it is the only message in the sequence that is not asking for something.
Three rules that hold across all of them: never assert anything about the case itself, keep the firm's name in every message so nothing arrives as an anonymous nudge, and make every message answerable in one line.
Stop rules
This is the section other vendors leave out, and it is the one that protects your firm. A follow-up system without stop rules is not a follow-up system — it is an outbound campaign that nobody is watching, attached to your firm's name, aimed at people who have just been injured.
Two categories of stop, and they behave differently:
| Trigger | Type | What the system does |
|---|---|---|
| Any opt-out language | Hard stop | Stops immediately on the keyword your firm defines and on plain-English equivalents. Contact is suppressed for that number permanently unless a person at your firm reverses it deliberately. |
| The person says they have counsel | Hard stop | Ends the sequence at once and routes to a person. Continuing to solicit someone who has told you they are represented is a problem for your firm, not for the software. |
| Wrong number or wrong person | Hard stop | Suppresses the number and flags the record, so a data error does not become repeated contact with an uninvolved person. |
| Any reply at all | Hard stop on automation | The sequence halts and a human takes over. Automation continuing to fire alongside a live conversation is the single most damaging failure mode in this category. |
| Consultation booked | Hard stop | Follow-up ends and confirmation and reminder messaging begins instead. These are different sequences and mixing them produces contradictory messages. |
| Delivery failures | Hard stop | Repeated failed sends mean the number is not reachable. Continuing to retry does nothing except degrade your sending reputation. |
| Touch limit reached | Natural end | The defined number of attempts is exhausted. Sequence closes, lead is marked, no further contact. |
| Time window elapsed | Natural end | A cap in days regardless of how many touches landed, so a paused or delayed sequence cannot resurface a week later out of context. |
A sequence with no end is a bet that the marginal touch is free. It is not. Every message after the point of usefulness is attached to your firm's name and arrives with someone who is injured, dealing with insurers, and possibly already represented by a competitor — who will hear about it. Messaging that reads as harassment gets discussed with friends, in reviews, and occasionally with a bar association. Meanwhile the mechanical costs are real too: carriers filter senders whose messages generate complaints and no engagement, which means an unbounded sequence to unresponsive leads degrades delivery for the leads that would have answered. The upside of touch nine is close to nothing. The downside is not. That asymmetry is the whole argument, and it does not require a statistic to be true.
Marking the lead, and what a "no" is worth
A sequence that ends without recording why has thrown away the only thing it produced. Every lead that exits should exit with a reason code your firm can report on:
- Booked — went to consultation. The outcome you are paying for.
- Declined by the firm — did not meet your criteria, screened out by a person.
- Already represented — worth tracking separately, because a high rate here is telling you something about your response speed rather than your marketing.
- Explicit no — the person said they are not pursuing it. This is a good outcome. It costs one text and saves your intake team five attempts.
- Opted out — suppressed permanently, and never re-enrolled by any future campaign.
- Unreachable — full sequence ran, nothing came back. This is the pile worth counting, because its size and its source mix tell you whether you have a lead quality problem or a response problem.
One rule that matters more than it sounds: a closed lead is never automatically re-enrolled. If the person contacts your firm again later, that is a new enquiry and it enters at the top. Nothing in the system should be able to wake up an old record and start texting someone months after they went quiet.
The compliance question that is yours, not ours
Outbound SMS and calling to prospective clients sit inside a real regulatory framework, and law firms carry an additional layer on top of it in the form of advertising and solicitation rules that vary by jurisdiction. What is permissible for your firm, in your jurisdiction, to someone who submitted a form versus someone whose number arrived from a referral partner, is yours to determine with your own ethics counsel. We are an operations vendor and we do not give legal advice. What we build is the mechanism, configured to the consent capture, sending windows, opt-out handling, touch limits, and record-keeping your firm specifies. Any vendor telling you their sequence is automatically compliant for your firm is claiming something they cannot know — they do not know your jurisdiction's rules, your lead sources, or your firm's own risk posture.
How we build this
- Audit what actually happens now. We look at your call and messaging records and count second attempts, not first ones. Most firms are surprised by this number, and some find they do not need us — which we would rather establish before you commit.
- Write the sequence with your intake lead. Touch count, spacing, sending windows, channel per step, and the exact wording. Your people approve every message before it exists in a system.
- Define stop rules first, not last. Hard stops and natural ends are configured before the sending logic is switched on, so there is never a version of this that can send without an exit.
- Test against your own numbers. Your team runs through the whole sequence, including opting out mid-way and replying mid-way, to confirm the automation actually stops when it should.
- Go live on one lead source. Usually web forms only. Small blast radius, easy to read the transcripts, reversible immediately.
- Read the replies together, then widen. We look at what people actually said back before adding more sources or more touches.
Follow-up sits on top of response. If your firm is not yet answering the first contact quickly, sequencing the second one is fixing the wrong end — measuring your real response time comes first, and deciding what should and should not be automated in intake comes with it.
Questions firms ask
How many touches is the right number?
There is no universal answer and we will not invent one. What we can say from mechanism: the touches that matter are heavily concentrated in the first day, and the marginal value of touch seven or eight is very low while its downside is not. We start firms at around six over four days and adjust from what their own replies show, not from a number in a vendor's brochure.
Should the follow-up come from the AI or from a person?
Both, in different roles. The system runs the schedule, sends the texts, and makes the attempts nobody has time to make. The moment a person replies, automation stops and a human takes the conversation. Anything else produces the worst outcome in this category — a bot arguing with a live conversation your intake specialist is already having.
Does texting prospective clients make our firm look desperate?
An unbounded sequence does. A short, clearly identified sequence that ends when asked does not — it reads as a firm that is organised. The variable is not whether you text, it is whether you stop. That is why the stop rules are configured before anything is allowed to send.
What happens if someone replies at 2am?
They get an immediate response, because they contacted you. Replying to inbound is not the same as initiating contact, and the system is configured to treat those cases differently — unprompted touches respect your sending windows, responses to a live message do not have to.
Can we run this on leads from before we installed it?
That is a question for your ethics counsel and your own judgement about consent, not for us. Technically it is possible. Whether an old form fill from four months ago is someone your firm may now text is exactly the kind of decision we build to your specification rather than make for you.