Key takeaways
- Attribute signed matters and fees, not clicks or calls; most of the useful information sits in your intake records.
- Capture the source twice: a tracked number or web address, and the caller's own answer to how they found you.
- Legal services revenue swings by season, falling 18.9% from Q4 2025 to Q1 2026, so compare each month with the same month a year earlier[1].
- A simple monthly report with six columns beats a complex model nobody updates.
What attribution should answer
Marketing attribution for a law firm has one job: tell you which sources produce signed matters and fees, at what cost, so you can move money and time toward them. Agencies tend to report impressions, clicks and calls, because those are easy to count. But a channel that produces many calls from people outside your practice areas can be your most expensive source. The unit that matters is the signed engagement, and the data for it lives in your intake log and case management system, not in an ad dashboard.
Attribution is also more forgiving than it looks. You do not need perfect tracking of every touch. You need a consistent way to record where each new inquiry came from, a way to connect that inquiry to whether it signed, and a habit of reviewing the numbers monthly. Firms that do those three things make better decisions than firms with sophisticated software and messy intake data.
The chain from first touch to fee
Source
The channel that created the contact: referral, search, a directory profile, targeted mail, a seminar, a sponsorship, a lead vendor.
Inquiry
The call, form, chat or email that reached the firm, with date, time and the tracked number or page it came through.
Qualified inquiry
An inquiry that meets your written acceptance criteria for practice area, geography and conflicts.
Consultation
A consultation booked and attended, recorded against the same inquiry record.
Signed matter
An engagement letter or fee agreement signed, linked to the original inquiry.
Fee
Fees billed or expected on the matter, so you can compare channels by value as well as volume.
Every step needs the same record identifier, usually the inquiry ID or the contact record in your case management system. If the consultation is booked in one tool and the signed matter in another with no shared ID, attribution breaks at that handoff.
Capture the source at intake, twice
Use two independent signals. The first is technical: a separate phone number for each offline channel, unique web addresses on print pieces, and campaign tags on online ads. The second is self-reported: the caller's answer to a question your intake staff ask every time. The two often disagree, and both are useful. A caller may have found you through search but chosen you because a friend mentioned your name; the tracked number will say search, and the caller will say referral.
- Ask: Before we finish, how did you first hear about us? Then ask: And what made you call today?
- Record the answers as picklist values, not free text: Referral from client, Referral from lawyer, Search, Map listing, Directory, Social, Mail, Seminar or event, Sponsorship, Lead vendor, Returning client, Other.
- Add a free-text note only for the specific referrer or event name, so referral sources can be thanked.
- Make the source field required before an inquiry can be saved or converted to a matter.
Call tracking without losing callers
Call tracking numbers make offline channels measurable, but they create one practical problem: callbacks. Pew Research Center found that 80% of US adults generally do not answer calls from unknown numbers, and 67% let them go to voicemail and then check the message[2]. If staff return calls from a number the caller does not recognize, many calls will go unanswered. Return calls from your main line, leave a voicemail that names the firm, and keep tracking numbers for inbound use.
Where calls arrive matters too. Many inquiries come in the evening or on weekends, when nobody is logging sources. An SI Employee such as Ava can answer those calls live, ask your source question along with the rest of your approved intake, and hand off a clean summary so the record is complete by morning.
Choosing an attribution model
| Model | Credits | Good for | Blind spot |
|---|---|---|---|
| First touch | The channel that created the first contact | Judging awareness channels such as sponsorships and seminars | Ignores what finally prompted the call |
| Last touch | The channel used just before the inquiry | Judging search ads and tracked numbers | Over-credits search, which people use to look up a firm they already heard about |
| Self-reported | What the client says | Referrals and word of mouth that tracking cannot see | Memory is imperfect; people simplify |
| Blended | First touch and self-reported side by side | Small firms that want a usable view without software | Takes discipline to keep both fields filled in |
For most small and mid-sized firms, the blended approach works best. Report signed matters by tracked source and by self-reported source in two columns, and look for channels that score well on both. Online reviews are a good example of why: Pew found that 82% of adults at least sometimes read reviews before buying something for the first time[3]. A prospective client may hear your name at a seminar, read your reviews, then search for your firm by name. Last-touch reporting would credit search for a client the seminar created. With 97% of US adults using the internet[4], almost every client, referrals included, will look you up online before calling, so online touches appear in most journeys whether or not they started them.
Referrals and offline channels need names, not categories
Referrals are the largest source for many small firms and the hardest to track, because no number or link carries them. The fix is to record the person, not just the category. When a caller says a friend sent them, ask who, and store the referrer as a contact linked to the new inquiry. Over a year, that list shows which clients, lawyers and community contacts send matters, and how many of those matters sign.
- Client referrers: record the referring client's name so you can thank them, within your state's rules on giving anything of value for referrals.
- Professional referrers: record the referring lawyer, accountant, financial adviser or social worker, and review the list each quarter.
- Events and sponsorships: use a dedicated web address or sign-up code at each event, plus the self-reported answer, because most attendees will call weeks later from memory.
- Print and outdoor: give each piece its own tracked number and short web address; reuse across campaigns erases the signal.
Review the referrer list alongside the monthly report. A referral source that sends three signed matters a year deserves a lunch, a thank-you note and an update on how you can help their contacts; it is usually the best-value channel you have.
The monthly report
Keep the report to one table that a managing partner can read in two minutes. Columns: source, inquiries, qualified inquiries, consultations, signed matters, fees expected, and marketing spend. Add cost per signed matter and fees per dollar spent as the last two columns.
| Source (Example) | Inquiries | Consults | Signed | Spend | Cost per signed |
|---|---|---|---|---|---|
| Client referral | 18 | 11 | 7 | $0 | $0 |
| Search ads | 64 | 14 | 5 | $4,000 | $800 |
| Directory profile | 22 | 6 | 2 | $900 | $450 |
| Seminar | 9 | 5 | 3 | $1,200 | $400 |
Example: in the fictional month above, search ads produced the most inquiries but the highest cost per signed matter, and the seminar produced the fewest inquiries but a strong signing rate. One month proves little; three months of the same pattern justifies moving budget.
Account for seasonality
Legal demand is seasonal, which can make a channel look broken when the whole market has slowed. Census Bureau estimates show unadjusted legal services revenue peaked at $132.1 billion in the fourth quarter of 2025 and fell 18.9% to $107.2 billion in the first quarter of 2026[1]. Compare each month with the same month last year, and judge channels on their share of signed matters, not raw counts, during slow quarters.
Source: U.S. Census Bureau, 2026[1]
Show the data
| Item | Value |
|---|---|
| Q4 2025 | 132.1 billion dollars |
| Q1 2026 | 107.2 billion dollars |
Common attribution mistakes
- Counting calls instead of qualified inquiries, so channels that attract wrong-fit callers look strong.
- Letting staff skip the source field when they are busy, which removes the busiest days from the data.
- Crediting branded search for clients who already knew your name from another channel.
- Changing several channels at once, so no change can be measured.
- Ignoring time lag: estate planning or business clients may sign months after the first touch, so judge those channels on a longer window.
A 30-day setup plan
- Week 1: agree the source picklist and make the field required in your intake form and case management system.
- Week 2: assign tracked numbers to each offline channel and campaign tags to each online ad; route all of them to the main intake line.
- Week 3: train staff on the two source questions and on calling back from the main line.
- Week 4: build the monthly report from the first full month of data and set a standing review date.
Frequently asked questions
What is the simplest way to track where law firm clients come from?
Ask every new caller how they first heard about you, record the answer as a picklist value, and link it to whether the matter signed. Add tracked numbers for offline channels when you can.
Should we return calls from a tracking number?
Usually not. Most people do not answer unknown numbers, so return calls from your main line and leave a voicemail with the firm name[2].
How long should we wait before judging a channel?
At least three months for high-volume consumer practices, longer for practices where clients take months to decide. Compare with the same period a year earlier because demand is seasonal[1].
References
U.S. Census Bureau (2026). Quarterly Selected Services Estimates, Second Quarter 2026 (CB26-146) (opens in a new tab) โ
Pew Research Center (2020). Most Americans don't answer cellphone calls from unknown numbers (opens in a new tab) โ
Pew Research Center (2016). Online reviews and ratings (opens in a new tab) โ
Pew Research Center (2026). Americans' Use of Internet, Broadband in 2026 (Internet, Broadband Fact Sheet) (opens in a new tab) โ