How to Configure GA4 Session Timeout for Ecommerce

GA4 describes a session as a group of interactions within a time frame. A session begins when a user opens an app in the foreground or views a page or screen while no session is active. The default web inactivity timeout is 30 minutes, and Google allows an adjusted timeout up to 7 hours and 55 minutes. A longer timeout can keep interactions in one session after a longer idle gap; a shorter timeout can create another session sooner. Neither setting identifies a person or proves one continuous shopping intention.

The engaged-session timer answers a different question. Google defines an engaged session as one that lasts longer than the configured threshold, has a key event, or has at least two page or screen views. The documented default duration criterion is longer than 10 seconds, and the Google tag settings expose an adjustment. Do not change engagement time merely to make engagement rate move. Write down which decision the altered definition would support and why the ordinary key-event and page-view conditions are insufficient.

Map representative journeys that might include legitimate idle time: product research across tabs, saved carts, account verification, financing, customer-support contact, business approval, subscription configuration, or a checkout that hands off to another domain. Measure event timestamps and inactive gaps from consented, appropriately governed data. Separate humans who return with the same intent from browsers left open in the background. The property-wide timeout affects every measured journey, so one slow workflow is not enough evidence for a broad change.

Review timezones, app streams, cross-domain setup, user identity, unwanted referrals, offline events, and Measurement Protocol events that can make sessions appear fragmented for reasons unrelated to the timer. Repair collection defects first. If a payment provider starts a false referral or a duplicate tag creates page views, extending sessions masks the evidence. Preserve the current setting, baseline session count, engaged sessions, engagement rate, sessions per user, key events, and affected reports before any configuration change.

Create a change record with property and web stream, current timeout, proposed timeout, engaged-session threshold, evidence window, owner, approver, effective time, affected dashboards, expected definitional direction, and rollback value. Confirm the operator has the correct property and permission. Use the Google tag settings path documented by Google to adjust session timeout and engaged-session timing. Keep screenshots or exported configuration evidence without exposing account secrets.

Make the smallest change that tests the hypothesis. Do not simultaneously alter referral exclusions, attribution, key events, data retention, channel groups, or campaign tags. Property-wide session metrics will become definitionally discontinuous at the effective time, so annotate reports and notify consumers before saving. If several brands or storefronts share a property, confirm the selected value makes sense across them; otherwise the architecture may need review rather than a compromise timer that fits none.

Run a test journey with a clearly tagged, non-customer identity and no real purchase or lead communication. Capture event timestamps, session identifiers available in debugging, page sequence, idle duration, key-event state, and the expected boundary. Test just below and just above the new inactivity timeout when practical, plus a journey that becomes engaged by time, by two page views, and by a safe key event. Do not shorten a production timeout merely to make testing convenient.

Compare Realtime or debugging evidence with processed reports after normal latency. Verify the new setting is active on the intended stream and that session_start behavior, engagement, source attribution, and cross-domain navigation match the documented expectation. A single browser test verifies mechanics, not business impact. If the result conflicts with the plan, restore the previous value and inspect tags, consent, referrers, and event timestamps before retrying. Preserve both the failed and successful evidence in the change record.

After the effective date, compare complete periods while labeling the session-definition break. Expect possible changes in session count, average engagement time, engaged sessions, engagement rate, sessions per user, and session-scoped acquisition reports even if customer behavior did not change. Do not credit those shifts to merchandising, media, or site experience without separate evidence. Continue to use orders, revenue, and verified key events from their authoritative records.

Review the setting when checkout, authentication, support, consent, domains, apps, or customer journeys change. Keep a registry of every session and engaged-session timer with dates and reasons. If analysts need a special time window for one investigation, prefer a documented analysis rule over repeatedly changing the property. A good timeout is stable, understandable, and tied to observed journey boundaries. Its success is reporting consistency and clearer interpretation, not a higher engagement percentage.

Add the session-definition change to recurring dashboards, analysis notebooks, scheduled exports, and stakeholder notes for as long as any visible comparison crosses the effective date. Include both timer values and the exact change time. A small footnote is insufficient when a scorecard uses sessions as a denominator.

Remove the annotation only when every displayed period uses one definition. If a long-range report must cross the change, show a break or separate panels instead of smoothing the series. The goal is to stop a configuration effect from being narrated as a customer-behavior trend.