How do you read GA4 conversion paths for Leads?

Select one lead-related key event, confirm the date range and lookback window, choose a useful path dimension, and read early, mid, and late touchpoints under a named attribution model. Treat the result as an aggregated marketing path to a GA4 key event, not as a verified person-level Lead history or proof of a successful Delivery.

Answer first: a GA4 conversion path explains which eligible touchpoints appeared before a selected key event and how the current model assigns credit. It does not prove that the inquiry was contactable, qualified, routed, or accepted by a Destination.

Google now calls this the Key events attribution paths report. The older phrase Conversion paths remains common in searches and older guides. Google's current help pages also use both Attribution paths and Key event attribution paths for navigation. If your property shows one label rather than the other, use the Advertising workspace and look for the paths report under attribution.

Interface note, checked August 4, 2026: Google documentation is not uniform about the navigation label. This guide uses Google's current report title and calls out controls by purpose, so minor interface wording changes do not change the method.

What must be configured before the report is useful?

GA4 must collect the event you care about, and the event must be marked as a key event. Google says marking an event is prospective and can take up to 24 hours to appear in standard reports. For a lead funnel, choose an event that represents a meaningful browser action, such as a correctly implemented generate_lead event. Do not assume that every button click or page view is a Lead.

All key events are selected and aggregated by default in the paths report. Filter to one lead-related key event before interpreting a lead journey. Mixing a newsletter signup, a demo request, and a purchase can produce a path summary that answers no specific business question. The qualified Lead tracking guide explains how to keep the browser event distinct from a first-party Validation Outcome.

What does each report control mean?

The report has two jobs: show how touchpoints are positioned within paths and show how attribution credit is distributed. Read the configuration before the visualization. A polished chart with the wrong event, scope, model, or date range is still the wrong answer.

Report element What it controls or shows Lead reporting check
Key event selector Chooses which completed key event journeys are included. Select one meaningful lead event before reading totals.
Attribution model Changes how credit is assigned across eligible touchpoints. Name the model in every exported or shared report.
Path dimension Groups touchpoints by Primary channel group by default, with Source, Medium, and Campaign documented as alternatives. Choose the dimension that matches the decision, not the prettiest chart.
Touchpoint filter Includes or excludes paths based on touchpoint rules. Document any filter so readers know which paths disappeared.
Path length Filters by the number of touchpoints. Google documents paths up to 20. Check whether a narrow filter hides shorter or longer journeys.
Date range Limits the key events and report data being analyzed. Avoid treating recent, still-processing data as final.

The table below the visualization includes Key events, Purchase revenue, Days to key event, and Touchpoints to key event in Google's current documentation. Purchase revenue is usually irrelevant for a lead-only key event. Keep it out of a lead report unless the property has a documented revenue relationship that the reader actually needs.

How should you interpret early, mid, and late touchpoints?

The stage labels describe positions within the displayed path. They are not funnel stages, sales stages, or a claim about buyer intent. Google divides each path into an early 25 percent, middle 50 percent, and late 25 percent, with rounding for whole touchpoints.

Displayed stage Google-documented position Important limit
Early touchpoints First 25 percent of touchpoints, rounded to the nearest whole number. A one-touch path has no early stage.
Mid touchpoints Middle 50 percent of touchpoints. Paths with fewer than three touchpoints have no mid stage.
Late touchpoints Last 25 percent of touchpoints, rounded to the nearest whole number. A one-touch path assigns the path to the late stage.

Say “first eligible touchpoint in the displayed path,” not “the person's first-ever interaction.” The key event lookback window can exclude older interactions, and GA4's reporting identity affects how activity is associated. A displayed path is bounded by collection, eligibility, configuration, consent, and reporting rules.

Does changing the attribution model change the path?

The model selector changes the credit shown across eligible touchpoints. It does not rewrite the order of the displayed interactions. Google's current attribution overview lists data-driven attribution, paid and organic last click, and Google paid channels last click. Direct visits receive no credit unless the path consists entirely of direct visits.

Data-driven attribution can assign fractional key event credit across contributing interactions. Last-click models assign all credit according to their documented rule. Compare models to understand how the credit rule changes, but do not call one model a reconstruction of objective causality. The GA4 attribution models guide provides the detailed model comparison and current lookback choices.

Why do paths disagree with acquisition reports?

The reports answer different scoped questions. Attribution paths use event-scoped traffic dimensions around a key event. User acquisition uses first-user dimensions. Traffic acquisition uses session dimensions. Google's traffic-source scope documentation describes these user, session, and event boundaries.

View Typical scope Question answered Why totals can differ
User acquisition User, such as First user source Where was the user first acquired? It assigns an acquisition value at the user level.
Traffic acquisition Session, such as Session source Which source initiated the session? A user can have several sessions with different sources.
Attribution paths Event-scoped attribution around a key event Which eligible touchpoints appeared, and how is credit assigned? Model, lookback, path eligibility, and event selection all matter.
Lucidity Lead timeline First-party operational record What happened from Intake through Validation and Delivery? It counts operational facts, not GA4 users, sessions, or model credit.

Is a GA4 path a person-level journey?

No. The report groups path patterns using the property's reporting identity and reporting layer. Depending on property settings, GA4 can use User-ID, device ID, and modeling to associate activity. Consent, device changes, missing tags, and identity settings can split or combine reported activity differently from a CRM or Lead platform.

Privacy and identity boundary: do not use an attribution row as proof of one named person's complete history. Google documents thresholding for some low-count data, modeling in reports, and aggregated (Other) rows when cardinality limits apply.

Google also documents differences between standard reports, explorations, and BigQuery exports. Supported dimensions, sampling, retention, thresholding, and reporting-layer modeling can differ. If two surfaces disagree, first compare the event, scope, date range, filters, identity setting, and freshness before assuming an implementation defect.

Why can conversion path numbers change or look incomplete?

Recent data can revise while GA4 processes events and modeled attribution. The property lookback window also determines which older touchpoints are eligible. Google documents a default 90-day window for key events other than first_visit and first_open, with 30-day and 60-day alternatives. Changes to that window apply going forward.

Symptom Documented or likely boundary Check next
Recent values changed Processing or modeled attribution revised recent channel credit. Record the retrieval time and avoid calling recent data final.
Acquisition totals differ User, session, and event scopes answer different questions. Compare dimension scope before comparing totals.
(not set) Required dimension information was not received. Audit tagging, campaign parameters, and linked-product data.
Unassigned The traffic did not match a channel-group rule. Inspect Source and Medium values before changing channel rules.
(Other) High-cardinality rows were aggregated. Use a broader dimension or a supported export for deeper analysis.
GA4 and Lead counts differ A browser key event is not the same record as a received, validated Lead. Reconcile event IDs or safe attribution keys against first-party records.

Google's report and exploration comparison documents several reporting-surface differences. These are reasons to make a report reproducible, not reasons to force every surface to match.

What belongs in the separate Lead timeline?

Put browser and channel context beside the operational record, not in place of it. A first-party Lead timeline should show when an Intake received the inquiry, which Validation Checks ran, the Validation Outcome, the chosen Destination, and every Delivery attempt and outcome. Those records let an operator answer a question that GA4 cannot: did the Lead reach the place it was supposed to go?

System boundary: GA4 paths are traffic and attribution context. Lucidity Leads, Validation Outcomes, Destinations, and Deliveries are first-party facts. A later change in channel credit must not rewrite the operational timeline.

This separation also protects Data Freshness. GA4 context may lag or revise while a Lucidity Delivery record remains the timestamped fact of an attempt. The Lead Source attribution guide explains how to preserve deterministic evidence and use GA4 as additional context.

A six-step GA4 conversion paths reading method

  1. Select one lead key event. Avoid an all-events aggregate when the question is specifically about demo requests, quote requests, or another defined Lead action.
  2. Confirm date range and lookback. Record both. A date range selects report data, while the lookback window controls older touchpoint eligibility.
  3. Choose the path dimension. Start with Primary channel group for a summary, then use Source, Medium, or Campaign when the decision needs more detail.
  4. Read positions, not intent stages. Early, mid, and late describe relative path position. They do not label awareness, consideration, or sales stage.
  5. Name the attribution model. Compare credit only after stating the rule that distributed it. Keep the path sequence separate from model-assigned credit.
  6. Reconcile with first-party facts. Verify the Lead, Validation Outcome, Destination, and Delivery in Lucidity or another operational system.

Publication-ready QA checklist for a lead path report

Record with the report Why it matters
Property and key event Prevents an all-events chart from being mistaken for one lead action.
Date range and timezone Makes repeated extracts and cross-system comparisons understandable.
Attribution model and lookback Defines the credit rule and touchpoint eligibility.
Dimension and filters Shows how paths were grouped and which journeys were excluded.
GA4 retrieval time Labels potentially revisable traffic context.
First-party freshness States through which timestamp Lead and Delivery facts are complete.

How does Lucidity complete the picture?

Lucidity captures first-party inquiries as Leads, evaluates explicit Validation Checks, assigns a Validation Outcome, routes qualified Leads to configured Destinations, and records each Delivery attempt. GA4 can remain connected as traffic context without becoming the source of truth for those operational facts.

Use GA4 attribution paths to explain the marketing journey visible to the property. Use Lucidity to show whether the resulting Lead was safe to route and whether Delivery succeeded.

Put Lead outcomes beneath every GA4 path

See how Lucidity keeps first-party Validation and Delivery evidence beside GA4 traffic context for each Business.

Request a Lucidity Lead timeline demo