Client Intake

Intake Data Fields Every Firm Should Track

The minimum set of intake data fields a law firm needs for conflicts, follow-up and reporting, with formats, required stages and what not to collect.

AssistCall research teamPublished Reviewed by Faruque Amin, VP of Sales and Marketing6 min read

On this page
  1. A minimum dataset, not a maximum one
  2. The seven field groups
  3. Formats and when each field is required
  4. What to capture at each stage
  5. Reason codes worth having
  6. Rolling out a new dataset
  7. Why preferred language belongs in the dataset
  8. Data quality rules
  9. What not to collect at intake
  10. Reports the dataset makes possible
  11. Frequently asked questions
  12. References
Client Intake: Intake Data Fields Every Firm Should TrackArticle coverCLIENT INTAKEIntake Data FieldsEvery Firm Should TrackAva for LawyersInsightsClient Intake: Intake Data Fields Every Firm Should TrackArticle coverCLIENT INTAKEIntake Data FieldsEvery Firm ShouldTrackAva for Lawyers

Key takeaways

  • Every intake field should serve conflicts, follow-up or reporting; if it serves none, drop it.
  • Seven groups cover most firms: contact, conflicts, matter, source, timestamps, outcome and consent.
  • Timestamps are the most neglected and most useful fields: first contact, first response, consult booked, consult held, engagement signed.
  • Collect less sensitive data, not more: breach damages under the CCPA can reach $799 per consumer per incident[1].

A minimum dataset, not a maximum one

Most intake systems collect either too little, so partners cannot see what is happening, or too much, so staff spend calls filling in boxes nobody reads. The useful middle is a minimum dataset: the smallest set of fields that lets the firm run conflict checks, follow up reliably and report on how intake performs.

It is also a measurement problem. Firms that cannot measure intake cannot improve it, and measurement across the profession is thin. Thomson Reuters found that only 20% of legal professionals are measuring the return on investment of generative AI[2], one example of a wider habit of adopting tools before deciding what to track.

Legal professionals measuring the return on investment of generative AI, 2025
Legal professionals measuring the return on investment of generative AI, 2025Donut chart. Measuring ROI: 20%.Measuring ROI: 20%20%
  • Measuring ROI 20%

Source: Thomson Reuters Institute, 2025[2]

Show the data
ItemValue
Measuring ROI20%

Use the test below for every field you have or want: does it help check conflicts, follow up with this person, or report on intake? If none, cut it.

The dataset below works whether intake runs on a dedicated CRM, a practice management system or a well-structured spreadsheet. The tool matters less than agreeing the fields, their formats and who fills them in, and then keeping to that agreement every day.

The seven field groups

  • Contact: full name, preferred name, phone (normalized), email, preferred contact method, safe-to-message flag, preferred language.
  • Conflicts: other parties' full names, related entities, spouses or family members involved, and the conflict result with who cleared it and when.
  • Matter: practice area (from a fixed list), matter type, jurisdiction (state and county), short description, urgency flag and the reason for it, priority tier.
  • Source: first-touch source from a fixed list, referring person or firm if any, campaign or tracking number, self-reported answer to "how did you hear about us".
  • Timestamps: first contact, first human response, consultation booked, consultation date, consultation held or no-show, engagement sent, engagement signed, record closed.
  • Outcome: engaged, declined by firm, declined by prospect, referred out, no response; plus a reason code for every non-engagement.
  • Consent and preferences: consent to text if the firm texts, call recording notice given, marketing consent kept separate from intake.

Formats and when each field is required

Core fields with formats, required stage and purpose. Pick lists beat free text for anything you will report on.
FieldFormatRequired byUsed for
PhoneDigits only, with country codeFirst contactFollow-up, duplicate matching
Other partiesOne name per entryBefore narrativeConflict check
Practice areaPick listFirst contactRouting, reporting
Urgency flag + reasonYes/no + pick listFirst contactTriage, escalation audit
First-touch sourcePick listFirst contactMarketing reporting
First human responseDate-time, automaticFirst replyResponse-time reporting
Consult held / no-showPick listAfter consult timeNo-show rate
Outcome + reasonPick listRecord closeConversion and loss analysis
Preferred languagePick listFirst contactStaffing, routing

What to capture at each stage

Asking for everything at first contact slows the call and annoys the caller. Spread the fields across the stages where they naturally arise.

  1. First contact

    Name, phone, email, preferred contact method, safe-to-message flag, preferred language, practice area, jurisdiction, urgency, other parties' names and first-touch source. The system stamps the first-contact time automatically.

  2. Conflict check

    Result, who ran it, who cleared it and when. Any related names discovered during the search are added to the record.

  3. Booking

    Consultation type, attorney, date and time, format, and the time the booking was made. Note whether the booking happened on the first call or after a callback.

  4. Consultation

    Held or no-show, length, and the attorney's decision: engage, decline or needs more information. A short attorney note is fine here.

  5. Engagement or close

    Engagement sent and signed times, fee arrangement type, or the outcome and reason code if the matter does not proceed. The record closes with a final status.

Reason codes worth having

When an inquiry does not become a matter, the reason is the most valuable thing the firm can learn, and it disappears if nobody records it. Keep the list short enough that staff use it consistently:

  • Outside practice area.
  • Outside jurisdiction.
  • Conflict of interest.
  • Prospect chose another firm.
  • Cost or fee arrangement.
  • Timing: the prospect was not ready, or the firm could not meet the timeline.
  • No response after follow-up.
  • Matter not viable in the attorney's judgment.

Review the counts monthly. A rising "chose another firm" share often points at speed; a rising "cost" share points at how fees are explained.

Rolling out a new dataset

  1. Write the data dictionary first, one line per field, and agree it with the partners who will read the reports.
  2. Configure pick lists and required stages in whatever system you use, even a shared spreadsheet with validation.
  3. Train intake staff with five sample calls, entering the data as they go.
  4. Run the new fields alongside old notes for two weeks, then switch.
  5. Check data quality weekly for the first month: blanks, wrong picks, missing timestamps.

Why preferred language belongs in the dataset

Language is often captured in notes, if at all. As a field, it drives staffing and routing decisions. In US Census data, 22% of people aged 5 and older spoke a language other than English at home, and about 61% of them spoke Spanish[3][4]. A firm that records preferred language on every inquiry can see whether it needs a bilingual intake hire, a translated web form or an interpreter arrangement, instead of guessing.

Data quality rules

  1. Pick lists for anything you report on: practice area, source, outcome, reason codes. Free text cannot be counted.
  2. Automatic timestamps wherever possible. Staff-entered times drift and get rounded.
  3. One record per person per matter, with duplicates merged using written matching rules.
  4. Every record closes. Open records older than 30 days with no activity get reviewed weekly and closed with a reason.
  5. Required fields enforced at the stage they matter, not all at first contact. Do not block a booking for want of a campaign code.
  6. A short data dictionary, one line per field, kept with the intake manual.

What not to collect at intake

More data is more risk. Before the firm is engaged, keep the record to what it needs.

  • No Social Security numbers, financial account numbers or full dates of birth unless a specific matter type requires them at intake, and then only in a secured field.
  • No detailed narrative before conflicts clear. A prospective client's information is protected even if no relationship follows; California's Rule 1.18, for example, bars a lawyer from using or revealing it[5].
  • No health details beyond what routing needs, such as "still in treatment" rather than diagnoses.
  • No biometric data without the right consent. Illinois' Biometric Information Privacy Act treats a voiceprint as a biometric identifier and requires written notice and a written release before collection[6]. Check whether any phone, identity or call tool creates one.
  • No marketing tags mixed into intake consent. Keep marketing permission separate.

The cost of a breach scales with what you hold. Under the CCPA, statutory damages for a data breach rose in 2025 to between $107 and $799 per consumer per incident, or actual damages if greater[1]. Data you never collected cannot be breached.

Reports the dataset makes possible

  • Response time: median minutes from first contact to first human response, by channel and hour.
  • Conversion: share of qualified inquiries that book, attend and sign, by practice area and source.
  • Losses: reason codes for non-engagement, so recurring causes become visible.
  • Source quality: signed matters, not just inquiries, by first-touch source and referrer.
  • Workload: inquiries by hour and day, for staffing.

Example: A fictional three-office family law firm replaced free-text intake notes with 24 structured fields. Within a quarter it could see that weekend web forms took a median of 31 hours to get a reply, that Spanish-speaking callers made up a fifth of inquiries at one office, and that "cost" was the leading reason prospects did not proceed after consultations. Each finding led to a specific change, which free-text notes had never revealed.

For turning these fields into a weekly view, see building an intake dashboard for partners. For the source field, see asking "how did you hear about us?" the right way.

Frequently asked questions

What data should a law firm collect at intake?

Contact details, other parties for conflicts, matter type and jurisdiction, urgency, lead source, timestamps for each stage, the outcome with a reason, and any consents.

Which intake fields matter most for reporting?

Timestamps (first contact, first response, consult booked and held, engagement signed), practice area, source and outcome with a reason code.

Should intake collect a full case narrative?

Not before the conflict check clears. Collect the names and facts needed to route and check conflicts; the detail belongs in the consultation.

Free text or pick lists?

Pick lists for anything you will count or report on, with a short free-text note only where a person needs context.

References

  1. California Privacy Protection Agency (2025). Updated Monetary Thresholds in CCPA (opens in a new tab) โ†‘

  2. Thomson Reuters Institute (2025). 2025 GenAI report: Executive summary for legal professionals (opens in a new tab) โ†‘

  3. U.S. Census Bureau (2025). New Data on Detailed Languages Spoken at Home and the Ability to Speak English (opens in a new tab) โ†‘

  4. U.S. Census Bureau (2025). New Data on Detailed Languages Spoken at Home and the Ability to Speak English (opens in a new tab) โ†‘

  5. The State Bar of California (2026). California Rules of Professional Conduct (2026) (opens in a new tab) โ†‘

  6. Illinois General Assembly (2025). Biometric Information Privacy Act (740 ILCS 14) (opens in a new tab) โ†‘

Book a demo

See Ava handle your calls.

Book a short demo. Weโ€™ll walk through your calls and workflow, and agree what Ava should take care of.

Book a demo