Nationwide: Cost-Per-Case Tracking for Legal Marketing

• The signed case is the unit of truth: cost per signed case by channel — not cost per lead — is the number every marketing decision should resolve to, and the whole system exists to produce it reliably.
• Definitions precede data: a written stage glossary — contact, qualified inquiry, consultation, signed — with timestamps at every gate is what makes any number trustworthy; a metric nobody defined is a metric nobody can defend.
• The plumbing is unglamorous and decisive: CRM source fields required at creation, dynamic call tracking, UTM discipline, LSA dispute integration, and the offline-conversion loop feeding signed-case signals back to the platforms.
• Attribution policy is written down and honest: last-touch as the operating default, the intake question as ground truth, monthly reconciliation between the two, brand search as the mass-media proxy, and no precision claimed that the system doesn't have.
• Intake is a channel: speed-to-contact, contact-to-consult, and consult-to-signed clocks — with the missed-call autopsy and QA-to-conversion ties — routinely move cost-per-signed more than any bid change.
• Contingency practices run cohort accounting: signed is not revenue until resolution, so signed cohorts are tracked to outcome across years, and the ledger's patience is part of its honesty.

This manual is the plumbing, written for firms in any U.S. market and any practice area — the operational companion to this library's measurement philosophy and the system our auto-auction manual fights its whole war on. It covers the stage architecture and the glossary discipline that makes numbers trustworthy, the CRM-and-call-tracking spine, attribution policy with its honesty intact, the per-channel ledgers with their channel-specific traps, the intake clocks, the math from cost-per-signed through contingency cohort accounting, the operating rituals, the failure catalog, and the privacy layer none of it may ignore. Marketing operations guidance only; nothing here is legal advice; recording, consent, and advertising rules vary by state and belong to your counsel; and Bar advertising rules govern every asset the system measures.

Why leads lie: a lead count rewards volume that screening will discard, treats the report-lookup call and the catastrophic-injury call as equals, and lets every vendor in the stack claim success while the firm's caseload stays flat. The signed case resists all of that — it is scarce, definable, timestamped by a fee agreement, and connected to revenue — which is why this manual's every system points at it. The operating rule that follows: any metric that cannot be traced forward to signed cases or backward to spend is a diagnostic, not a scoreboard — useful for tuning, never for judging. Clicks diagnose ads; calls diagnose media and hours; consult rates diagnose intake; only cost per signed case judges the machine.

Before any tool is bought, the firm writes its glossary — one page, agreed by partners, intake, and marketing, changed rarely and never mid-year: Contact (any inbound human reaching the firm — call, form, chat, walk-in — logged with source and timestamp); Qualified inquiry (screened-in per written criteria: practice area fit, jurisdiction, conflict-clear, viability threshold); Consultation (the substantive evaluation conversation, scheduled and held as separate timestamps); Signed (fee agreement executed — the e-signature timestamp as the source of truth, not "verbally agreed"); and for contingency work, Resolved (the matter's outcome, years later, closing the loop the cohort section below depends on). Every stage records its timestamp and its owner; stage-aging reports surface where matters stall; and the glossary's discipline is the whole system's foundation — because a number nobody defined is a number nobody can trust, and the first casualty of fuzzy definitions is always the marketing budget's credibility.

• The CRM is the spine. Every contact becomes a record at first touch with source required at creation — not optional, not backfilled Fridays — carrying channel, campaign, and query family where known; stage timestamps write automatically as matters move; and the "unknown source" rate is tracked as a system-health metric with a target near zero, because every unknown is a dollar the ledger can't judge.
• The call layer is instrumented. Dynamic number insertion assigns numbers by channel and campaign so every call carries its provenance; recordings — where properly noticed and lawful — feed intake QA; missed calls log with hour and source for the autopsy ritual; and after-hours behavior is measured, not assumed.
• Forms and chat carry their history. UTM discipline on every destination, hidden fields passing source into the CRM, and chat transcripts attached to records — the unglamorous hygiene that separates ledgers from guesses.
• LSA integrates as a first-class channel. Lead exports reconciled to CRM records, the dispute log feeding effective-CPL math, and badge-unit leads tracked to signed exactly like auction clicks.
• The loop closes back to the platforms. Signed-case events upload to ad platforms as offline conversions so automated bidding optimizes toward retention rather than calls — the single highest-leverage integration in modern legal paid media, because the machine optimizes to whatever you feed it, and most firms feed it noise.

Attribution is a policy you write, not a truth you discover — so write it: last-touch as the operating default for channel ledgers, chosen for consistency rather than metaphysics; the intake question as ground truth — "how did you hear about us?" asked on every contact, logged verbatim, never force-mapped into a dropdown that flatters a channel; the monthly reconciliation ritual — digital attribution beside intake answers, with divergence investigated rather than suppressed, because when the click says paid search and the client says "my sister," both are true: one is the path, the other is the cause, and a firm that sees both is smarter than its dashboard; brand search as the mass-media proxy — TV, radio, and outdoor surface as branded-query volume and direct traffic, read as consistent-with rather than proven-by; long windows for long verticals — estate, family, and med-mal decisions gestate for months, and attribution windows respect it; and precision honesty throughout — stated confidence, acknowledged assists, and no decimal-point theater on numbers built from policy choices.

Two ledgers deserve emphasis. Referrals are a first-class channel — every professional-referral playbook in this library ends in a ledger, and here is where it lives: source, matter, value, and reciprocity tracked monthly, because the channel firms call "free" is the one they starve by not measuring. And content is the patience channel — judged on the multi-quarter curves this library documents, with AI-citation share tracked alongside rankings as the entity work compounds.

Two dimensions cut across every ledger. Language: bilingual firms report every clock and every CPS by language, because the Spanish-native funnel frequently carries different economics end to end, and a blended number hides whichever side is winning. Location: multi-office firms run per-office ledgers with geo-source discipline — office-level tracking numbers, location-tagged forms, per-market LSA units — plus a written policy for brand-level demand that any office could claim, so growth debates between offices argue from the same book.

Intake is a channel — often the highest-ROI one in the stack — and the clocks prove it: speed-to-contact on forms and missed calls, measured in minutes against a written standard; contact-to-consult conversion by intake operator, source, and language; consult-to-signed by attorney and practice area; stage aging flagging matters stalled between gates; the missed-call autopsy every week — what rang out, when, from which spend, at what implied cost; and QA-to-conversion ties — scored calls correlated with outcomes, so coaching aims at what actually signs cases. The standing insight the ledger keeps proving: a five-point improvement in intake conversion outperforms most bid-strategy projects at a fraction of the cost, which is why this manual treats the phone room as media.

The core formula, fully loaded: channel spend plus its share of fees and tooling, divided by signed cases attributed under the written policy — because a ledger that ignores the agency retainer and the software stack is flattering itself. The declined-at-cost line: what the firm paid, in media, for inquiries it declined — faced squarely as the number that funds screening and negatives work. Value tiers: signed cases are not equal — practice-area and matter-value tiers weight the ledger so a channel producing few-but-major matters reads fairly beside a volume channel, with blended and per-tier CPS both reported. Cohort accounting for contingency work: signed is not revenue until resolution, so each quarter's signed cohort is tracked to outcome across the years the multi-stage verticals require — fees realized, costs carried, duration — and marketing's true return is read at cohort maturity, not at signing. The flywheel adjacency: transactional practices track the second-matter rate — the planning-to-probate and repeat-client patterns — because acquisition cost amortizes across the relationship, not the first file.

The system's value is interpretive, and three worked patterns teach the skill. Cheap leads, empty consults: a channel with excellent CPL and terrible contact-to-consult conversion is usually buying the wrong humans — a lead-quality problem the negatives file or targeting fixes — unless QA shows those calls mishandled, in which case it is an intake problem wearing a media costume; the clocks and recordings adjudicate. Rising CPS, flat CPC: when cost per signed climbs while click prices haven't, the leak is downstream — speed-to-contact drifting, a new intake hire, consult scheduling friction — and the fix is coaching, not bidding; firms that answer this pattern with budget cuts starve a channel to pay for a phone-room problem. Free cases, flat volume: an organic or referral line with near-zero CPS but no growth is a capacity question — more content, more cultivation, more coverage — and the marginal dollar often belongs there rather than in the auction, which is exactly the reallocation argument the quarterly ritual exists to have. The ledger's job is not to display numbers; it is to make these arguments decidable.

The cadence that keeps the system alive: weekly, the operator view — spend, signed, CPS by channel; the intake clocks; disputes filed; negatives-session output — one page, numbers an operator acts on; monthly, the reconciliation — intake ground truth against digital attribution, unknown-source rate, referral ledger review; quarterly, the reallocation — budgets moved on evidence, kill criteria executed without sentiment, value-tier mix reviewed; annually, the cohort read — resolution outcomes closing old loops, channel curves compared across years, and next year's plan built on the only history that counts. Two dashboards, deliberately: the operator view above, and a partner view that summarizes honestly without inviting micro-management of week-to-week noise — same numbers, different altitude, zero vanity metrics on either.

The ledger is also the agency's report card: any vendor spending the firm's money reports at this manual's level — CPS by channel under the written policy, intake clocks where they touch them, disputes and negatives logs, kill-criteria recommendations against their own campaigns — per the accountability standard this library tells firms to demand. An agency that resists ledger-level reporting, ours included, has answered the evaluation; an agency that builds the ledger for you has understood the assignment.

The seven ways these systems die, named so yours doesn't: counting leads, not cases — the original sin, decorating everything downstream; the unknown-source plague — optional source fields quietly reaching a third of records until the ledger is astrology; double-crediting — every channel claiming the same signed case with no policy to adjudicate; optimizing platforms to calls — teaching the machine to buy report-lookups at scale; ignoring declined-at-cost — celebrating cheap leads the firm pays to reject; definition drift — "qualified" quietly redefined mid-year, breaking every trend line the firm thought it had; and vanity dashboards — impressions and engagement performing for partners while the caseload answers to nothing. Each has the same cure: the glossary, the required source field, the written attribution policy, and the weekly ritual — which is to say, this manual, actually run.

The system records humans in distress, and it behaves accordingly: call-recording consent laws vary by state — Florida, notably, requires all-party consent — and your counsel governs notice language and practice everywhere the firm answers; intake data touching medical matters is handled with the heightened care those matters demand, retention is a written policy rather than an accident, access is role-limited, and vendor agreements cover the data they carry; platform rules restrict remarketing and personalization for sensitive categories, so follow-up lives in consented first-party channels; and every measurement asset — numbers included — stays inside Bar advertising rules, because a dashboard screenshot in a pitch deck is advertising too. Instrumentation never outranks dignity: the recording that improves QA and the autopsy that improves speed exist to serve callers better, and the firm that forgets it eventually measures its way into trouble.

• Days 1–30 — Glossary and spine. The stage glossary written and signed by partners, intake, and marketing; CRM configured with required source fields and stage timestamps; the attribution policy drafted; recording-consent practice confirmed with counsel; baseline reconstructed from whatever history exists, gaps named honestly.
• Days 31–60 — Plumbing live. Dynamic call tracking deployed with QA and the missed-call log; UTM and form hygiene shipped; LSA exports and disputes integrated; the offline-conversion loop connected and feeding signed events; the weekly operator dashboard running on real data.
• Days 61–90 — Rituals and first reads. The monthly reconciliation run twice; intake clocks published with the QA-to-conversion tie live; declined-at-cost and value tiers reporting; the first quarterly reallocation executed on evidence — and for contingency practices, the first cohort opened with its multi-year read scheduled like the appointment it is.

Astra Results Marketing builds legal measurement as this manual describes: the glossary first, the plumbing complete, attribution honest, intake treated as media, the math fully loaded through cohorts, and the rituals actually kept — because every playbook in this library ends at the same place, and this is the system that makes that place real. Engagements begin with a tracking and ledger audit through our business consulting team.

The one your team will actually keep clean — the principles outrank the product: required source fields at creation, stage timestamps, call-tracking and LSA integration, offline-conversion export, and reporting your operator view can live in. Legal-specific intake platforms and general CRMs can all run this manual; the failure mode is never the software, it is optional fields and definition drift, which no vendor can sell you out of.

One full quarter of disciplined data for the operating picture — enough contacts, consults, and signings under the written glossary to read CPS by major channel — and a year before trend lines deserve real weight; contingency cohorts mature on their own multi-year clocks. Trust arrives with the reconciliation ritual, not the dashboard install: when intake ground truth and digital attribution have been argued into agreement a few months running, the ledger has earned its authority.

With two honest books: the operating ledger reads cost per signed case now — the decision metric for media and intake — while cohort accounting tracks each quarter's signed matters to resolution across years, reading marketing's true return at maturity. The discipline is refusing to collapse the two: signing velocity funds decisions today, cohort outcomes judge strategy later, and neither number is asked to do the other's job.

Yes — the system is habits, not headcount: the glossary is a page, source fields are a CRM setting, call tracking installs in a day, and the weekly operator view is fifteen minutes once the plumbing exists. What a small firm cannot skip is the discipline — required fields, the intake question, the weekly look — and what it gains is outsized: in small caseloads, every misattributed case distorts the picture more, which makes the ledger more valuable, not less.

Investigate, don't override: the click path and the client's story are answering different questions — how they arrived versus why they chose — and the divergence is information. Log both, reconcile monthly, look for patterns (a "referral" surge inside paid-search clicks often means the brand layer is working), and let the written policy decide which book each report uses. Firms that force one truth lose the insight the disagreement was offering.

The number is yours or it is nobody's: it varies by practice area, market, matter value, and model so widely that any benchmark would mislead someone — which is why this library refuses them. Good is defined by your ledger: CPS that your case values support, improving over quarters, with declined-at-cost shrinking and intake clocks tightening. The firm chasing someone else's benchmark optimizes a stranger's business; the ledger exists so you can optimize your own.

READY TO KNOW WHAT A SIGNED CASE ACTUALLY COSTS? Astra Results Marketing builds the signed-case ledger end to end — glossary, plumbing, honest attribution, intake clocks, cohort math, and the rituals that keep it true. Start with a tracking and ledger audit for your firm. ▸ CALL (786) 321-2866 · ▸ REQUEST YOUR CONSULTATION