# GA4 Pricing: When Analytics 360 Is Worth Evaluating Canonical URL: https://www.martechdevs.com/blog/ga4-pricing By [Jafrin Akter Alo](https://www.martechdevs.com/authors/jafrin-akter-alo) Data Engineer and Co-founder, martechdevs Published: 2026-06-24 Last updated: 2026-09-21 Standard Google Analytics 4 has no licence fee. Analytics 360 requires an enterprise contract, with pricing obtained through Google or a sales partner. The useful comparison includes event volume, property structure, export requirements, support and the work needed to maintain reliable measurement. Consider 360 when a documented constraint blocks a valuable workflow and the proposed contract resolves that constraint. A team can spend heavily on GA4 without paying Google for the standard product. Developers maintain event collection, analysts reconcile reports, and the warehouse creates separate usage costs. Another team can purchase 360 and retain the same broken purchase events. Keep the product licence and the measurement system’s operating cost visible as separate decisions. Google presents its standard Analytics product as free of charge. Its enterprise contact route includes Analytics 360 and directs buyers to sales or a partner. The official pages reviewed for this guide do not provide a universal public 360 price that can be applied to your account. [Google Analytics](https://marketingplatform.google.com/about/analytics/) and [enterprise contact](https://marketingplatform.google.com/about/contact-us/) are the primary commercial references. ## What problem would paying for 360 solve? Name the blocked workflow before comparing editions. “We need more accurate analytics” is too broad to establish a purchasing requirement. “Our required daily event export exceeds the standard export ceiling on campaign days” is specific enough to investigate. Useful candidates include a collection or reporting limit, a need for subproperties or roll-up properties, an eligible export freshness requirement, and a contractual service or support requirement. Each needs evidence from the actual account and a business owner who will use the resulting capability. If the issue is duplicate transactions, inconsistent events or missing consent context, begin with an [analytics tracking review](https://www.martechdevs.com/analytics-tracking). A licence upgrade does not repair the application’s event definitions or the organisation’s measurement decisions. If the issue is a reporting workflow, reproduce the query and record the limitation before assuming the entire platform needs replacement. A concise requirement might read: “Finance needs yesterday’s approved acquisition dataset before its morning refresh, at the forecast event volume, with a defined escalation route when delivery fails.” That statement makes freshness, completeness, scale and support independently assessable. ## Which limits should you compare? Google’s current edition comparison gives 360 higher allowances across collection, reporting and configuration. Examples include 25 versus 100 event parameters, 50 versus 125 event-scoped custom dimensions, and exploration sampling thresholds of 10 million versus 1 billion events per query. The complete list includes other limits and qualifications. [Google’s 360 feature comparison](https://support.google.com/analytics/answer/11202874?hl=en) is the source to use when documenting a specific constraint. Higher limits matter only when they support an intentional design. Filling the standard custom-dimension allowance with obsolete properties is a governance problem. Keeping a growing set of legitimately required dimensions for multiple product teams may be an edition requirement. Audit names, definitions, usage and owners before buying more capacity. Distinguish several issues that can look similar in a dashboard: sampling, aggregation into an “other” row, privacy thresholds and retention. They arise for different reasons. A paid edition should not be sold internally as a universal way to remove all reporting restrictions. The [GA4 data-threshold guide](https://www.martechdevs.com/blog/ga4-data-thresholds) helps identify one of those separate mechanisms. For retention, standard properties offer up to 14 months for the relevant event and user-level settings; eligible 360 properties offer longer options, up to 50 months. Large-property qualifications apply. These retention controls affect explorations and certain user-level data rather than establishing a blanket expiry for every standard aggregate report. Check the precise setting and property classification in [Google’s retention documentation](https://support.google.com/analytics/answer/7667196?hl=en-GB). Buying longer retention also cannot recreate detail that was already deleted. A historical-analysis requirement needs an export and retention plan early enough to preserve the data it depends on. ## Does BigQuery require Analytics 360? No. Standard GA4 supports BigQuery export. Google’s daily batch export limit for a standard property is one million events per day. If the property repeatedly exceeds it, daily exports can be paused; the documentation also warns that significant exceedance can trigger an immediate pause and that missed days will not be reprocessed. Export filtering can reduce the included streams or event types. [Google’s export setup guide](https://support.google.com/analytics/answer/9823238?hl=en) explains the limits and consequences. Measure peaks as well as averages. A monthly volume divided by 30 can hide launch days that exceed the daily allowance. Use the property timezone consistently, include automatically collected events in the inventory, and forecast changes from new products or increased capture. Streaming export has no event-volume limit, but it is a best-effort feed with completeness and attribution limitations. It is not an equivalent replacement for a completed daily dataset merely because the volume ceiling differs. Google also documents additional streaming ingestion charges and separate BigQuery storage and processing costs. [The export comparison](https://support.google.com/analytics/answer/9358801?hl=en) describes these trade-offs. The [GA4 BigQuery export guide](https://www.martechdevs.com/blog/ga4-bigquery-export) covers the implementation choices. Price a warehouse workflow around the actual data needed, query frequency, retention and transformation work. Upgrading Analytics does not make those downstream costs disappear. ## How can property structure change the bill? A source property, a subproperty and a roll-up property do different jobs. A subproperty supports a filtered portion of source data. A roll-up combines data from multiple eligible sources. These can support organisational access and reporting requirements, but routing data through them also adds processing. Google documents that events copied into a subproperty or roll-up incur an additional charge at half the initial event rate, subject to the 360 contract. An event included in multiple destinations contributes at each destination. The account-structure guide also describes a partner bill-preview capability. [Google’s account-structure guidance](https://support.google.com/analytics/answer/9679158?hl=en-GB) is the relevant reference. Here is an illustrative monthly routing example, expressed as billing-weighted event units rather than a dollar quote: - Two source properties each receive 20 million events, for 40 million in total. - A regional subproperty receives 12 million events from the first source. - A product subproperty receives 8 million events from that same first source, including some also present in the regional subproperty. - A roll-up receives all 40 million events from the two source properties. The weighted total is 40 million plus half of 12 million, half of 8 million and half of 40 million. That equals 70 million source-rate-equivalent event units. The additional routes increase the weighted volume by 75% relative to the two sources alone. The overlapping regional and product events are intentionally counted in both additional routes for this calculation. Deduplicating them across destinations would understate the documented processing model. This is a planning calculation, not a claim that the vendor invoice is simply 70 million multiplied by a public flat rate. Contract tiers, commitments and adjustments still need the partner’s calculation. Draw the proposed property structure before requesting a quote. Ask whether each additional property exists because of a real access or reporting requirement. Avoid creating copies merely to make the organisation chart look tidy. ## How should you prepare an event forecast? Create three forecasts: the current measured workload, the approved near-term workload and a plausible peak case. Preserve the evidence behind each assumption. For the current workload, inventory daily events by property and event name. Flag bursts, duplicate senders and test traffic. Do not remove legitimate events simply to make the quote look smaller; classify what is unnecessary and what supports an actual analysis. For the planned workload, attach volume assumptions to specific releases. A new interaction event that fires 20 times per session has a different impact from one subscription-success event per paying account. Ask the product team to estimate frequency, then verify it during the rollout. For the peak case, include the likely campaign or seasonal multiplier and any new property routes. A retailer’s largest sales day and a SaaS company’s broad telemetry release create different risk patterns. The vendor should quote the version that reflects your operating reality. Use units explicitly. Source events, copied events, exported events, API tokens and BigQuery bytes are different quantities. A spreadsheet with a single column called “usage” invites errors because each service applies its own unit and limit. ![Identify a documented edition constraint, forecast routed events, price operations and approve only if the proposal resolves the workflow.](https://www.martechdevs.com/assets/blog/ga4-pricing.svg) ## What should the vendor return with a quote? Request a quote that can be reconciled to your forecast, not only a first-year total. It should identify the account and properties covered, the billing period and currency, the event-volume assumptions, the treatment of subproperties and roll-ups, and what happens when actual usage differs. Ask the partner to show the calculation for your current, planned and peak cases. If a volume band changes, identify where the change occurs and whether the commercial impact applies incrementally or through another contractual mechanism. Do not infer a tier rule from someone else’s historical contract. Separate the licence from partner services. Migration, event cleanup, training, reporting development and support may have different scopes and owners. Record who handles a collection incident, who handles a BigQuery billing problem, and who maintains your own transformations. “Support included” is too imprecise for an operational dependency. Review the agreement’s service definitions and remedies. A promised response time is different from a guaranteed resolution time. A service-level commitment for one export mode does not automatically cover every downstream dashboard. Include the selected export type and property eligibility in the discussion. Finally, ask what would happen at expiry or downgrade. The increased configuration allowances create dependencies that should be documented before they are adopted. Google’s upgrade and downgrade guidance confirms that a downgraded property becomes subject to standard limits. [The 360 documentation](https://support.google.com/analytics/answer/11202874?hl=en) describes those effects. ## How do you build a defensible business case? Compare incremental cost with a specific avoided cost or enabled decision. Keep the estimate separate from claims of causal revenue improvement. Consider a hypothetical team whose current reporting workaround takes 10 analyst hours per week. At an assumed loaded cost of USD70 per hour, 48 working weeks represent USD33,600 of annual effort. A pilot indicates that the proposed workflow would remove six of those weekly hours, leaving four for analysis and validation. The estimated released capacity is therefore USD20,160 per year. That amount is a capacity estimate, not a cash saving unless staffing or purchased hours actually change. It also does not establish the value of 360 by itself. Add the quoted licence increase, incremental warehouse usage, partner services and ongoing administration. Then consider any separately evidenced value from access controls, support or a decision that the current workflow cannot support. Do not count the same benefit twice. If the six saved hours already include manual reconciliation, do not add a second line for eliminating that same reconciliation. If a faster report could improve campaign decisions, state the mechanism and test it rather than assigning an unsupported revenue percentage. The [BigQuery pricing guide](https://www.martechdevs.com/blog/bigquery-pricing) can help estimate the warehouse component. Compare realistic alternatives, such as improving the existing export model or using a separate product-analysis tool, only when they can meet the same requirement. ## What should be settled before signing? - [ ] The blocked workflow is demonstrated, with its current limit and business owner recorded. - [ ] The forecast distinguishes daily export peaks, source volume and additional property routes. - [ ] The partner has priced current, planned and peak usage using the proposed structure. - [ ] Licence, services, cloud usage and internal ownership are separate budget lines. - [ ] The selected capabilities have acceptance criteria that can be checked after the upgrade. - [ ] Renewal, overage, downgrade and support terms have been reviewed against the actual operating plan. A useful decision memo can be short. State the requirement, the evidence, the chosen configuration, the full incremental cost and the condition under which the team would reconsider. The licence becomes accountable to a business need rather than a vague expectation that an enterprise label will improve the data. ## Frequently asked questions ### Is GA4 really free? The standard Google Analytics product has no licence fee. Implementation, maintenance, consent tooling, server hosting and warehouse use can still create costs. Budget those independently so a free product is not confused with a cost-free measurement operation. ### What is the exact price of Analytics 360? Obtain a current quote from Google or a sales partner for your account and workload. The official commercial pages reviewed here do not publish a universal price list. Historical prices from other contracts are not a reliable substitute for your volume, structure and service terms. ### Do I need 360 to export GA4 data? Standard properties can export to BigQuery, subject to the documented export limits and cloud costs. Daily batch and streaming exports have different completeness and attribution characteristics. Choose the feed around the required analysis rather than treating them as interchangeable. ### Will 360 fix inaccurate conversion tracking? An upgrade does not correct a duplicate sender, a missing transaction reference or an event that fires before the business action succeeds. Repair the collection and counting rules first. Evaluate higher limits or contractual features once the remaining problem is actually an edition constraint. ### Can subproperties increase the price even when source traffic is unchanged? Yes, copying eligible events into subproperties or roll-ups adds processing under Google’s documented model. Include every route in the forecast and ask the partner to show its billing treatment. Changes in reporting structure can therefore affect cost without any new website visitors. ## Sources 1. Google, [Google Analytics overview](https://marketingplatform.google.com/about/analytics/), accessed 21 September 2026. 2. Google, [Contact enterprise sales](https://marketingplatform.google.com/about/contact-us/), accessed 21 September 2026. 3. Google, [Analytics 360 feature limits and upgrades](https://support.google.com/analytics/answer/11202874?hl=en), accessed 21 September 2026. 4. Google, [Data retention](https://support.google.com/analytics/answer/7667196?hl=en-GB), accessed 21 September 2026. 5. Google, [Set up BigQuery Export](https://support.google.com/analytics/answer/9823238?hl=en), accessed 21 September 2026. 6. Google, [BigQuery Export](https://support.google.com/analytics/answer/9358801?hl=en), accessed 21 September 2026. 7. Google, [Analytics account structure](https://support.google.com/analytics/answer/9679158?hl=en-GB), accessed 21 September 2026.