How law firms should investigate unknown lead sources

Treat unknown as a status, not a channel.

A personal injury firm can know that an inquiry became a wanted case and still be unable to say what created it. The call record may exist without a campaign identifier. A form may reach intake after tracking parameters disappear. A prospect may name a friend, a billboard, and Google in the same conversation.

The tempting response is to assign the inquiry to the most likely source so the report balances. That produces a cleaner chart and weaker evidence. Unknown should remain a visible status until the firm finds support for a source or accepts that the record cannot be resolved.

This distinction matters most when leadership is comparing investment, case quality, and signed matters. If unknown cases are distributed by assumption, a channel can receive credit it did not earn and another can appear less valuable than it is. The owner may then make a budget decision from precision the firm does not actually have.

Define the population and the decision first.

Do not begin with every missing field in every system. Choose a defined inquiry cohort and the decision it needs to support. For a PI firm, that might be valid inquiries received during one acquisition month, or wanted cases signed during a period that has had enough time to mature.

State which source question you are answering. Original source, the source of the session that produced the inquiry, and the prospect’s self-reported influence are not interchangeable. Google Analytics documents first-user, session, and event-scoped traffic-source information separately. The firm should preserve those distinctions instead of compressing them into one field.

Start with the unknown records that could change a decision. A signed case with no source deserves more attention than a duplicate form submission. A cluster of after-hours calls with missing attribution may point toward a system or routing problem that can affect future records.

Source: Google Analytics: Scopes of traffic-source dimensions

• Name the acquisition cohort and observation period.
• Define the exact source field leadership needs.
• Prioritize wanted, signed, and decision-relevant records.
• Retain unresolved records instead of deleting or redistributing them.

Clarify the source definitions and evidence through law-firm marketing attribution.

Use an evidence hierarchy for each record.

Build a repeatable sequence for the investigation. Check the systems closest to the original interaction first: call-tracking detail, form payload, landing-page record, chat transcript metadata, referral field, and the intake system’s creation history. Then compare the available website and advertising records using a stable identifier where the systems support one.

Campaign tags can help identify manually tagged traffic. Google advises using the relevant UTM parameters together and notes that missing parameters can create “(not set)” values in reporting. This makes the tagging standard and redirect behavior part of the investigation, not merely a reporting preference.

Use the prospect’s answer as useful evidence, but label it accurately. Someone may remember a search, a television ad, a referral, or several exposures. That answer can inform influence and message continuity without automatically proving the technical source of the recorded inquiry.

Record the evidence that supports the conclusion. A source recovered from a preserved campaign identifier has a different confidence level from a staff inference based on the caller’s geography. If the evidence remains ambiguous, the correct result is still unknown.

Source: Google Analytics: Collect campaign data with custom URLs

Separate record recovery from prevention.

Recovering one record answers a historical question. Preventing the next failure requires a different action. Classify the reason the source went missing so the responsible team can fix the point of capture.

A missing parameter may belong to campaign governance or a redirect. An unrecorded referral answer may require a clearer intake prompt and field definition. A call that cannot be matched may require a review of number pools, routing, or the handoff into the intake platform. A source overwritten by a later interaction points toward data ownership and integration rules.

The marketing leader should coordinate the diagnosis, but ownership belongs with the team that controls the failure point. Agencies own agreed campaign tagging. Technical partners own implementation they are assigned. Intake leadership owns staff adoption of the approved disposition process. Firm leadership approves definitions and resolves conflicts between systems.

The responsible role should be confirmed against the firm’s actual systems, contracts, and operating structure.

Report recovery and confidence together.

Give leadership a simple reconciliation. Show the cohort total, records with a known source at the start, records recovered through documented evidence, and records that remain unknown. Keep counts beside percentages and identify whether the unresolved group is concentrated by channel, time, location, or workflow.

Consider a clearly labeled hypothetical cohort of 100 valid PI inquiries. Eighty-six begin with a supported source and 14 are unknown. The team recovers eight from preserved call or form evidence, while six remain unresolved. The final report is 94 known and six unknown, not 100 attributed. This is an illustrative method, not a client result or industry benchmark.

Carry wanted-case and signed-case status into the same review. Six unresolved low-fit inquiries may have little effect on a budget decision. Six unresolved signed cases could materially change the apparent channel mix. The firm should understand both the count and the business relevance without converting settlement value into assumed revenue.

Compare known and unknown records in the sample attribution report.

Use the unknown rate to make a better operating decision.

Set a target for data completeness only after the firm understands its current baseline and sources of failure. The target should create action, not encourage staff or vendors to guess. Pair it with acceptable evidence, assigned owners, a due date, and a review of whether the fixes are being used.

If a material share of decision-relevant cases remains unknown, leadership may need to delay a fine-grained reallocation while fixing capture and reviewing other evidence. That does not require freezing every marketing decision. The firm can still act on clear waste, capacity constraints, or strong channel evidence while stating what remains uncertain.

The goal is not perfect attribution. It is enough reliable evidence to choose the next investment without hiding the limitations. Preserve the unknowns, investigate them in a controlled queue, correct the recurring failure, and let the remaining uncertainty stay visible in the decision record.

Bring the remaining uncertainty into a structured marketing performance audit.