Conversion Goals in Google Analytics: A Complete GA4 Guide

Introduction: setting up goals in Google Analytics to steer SEO and GEO

To manage an effective SEO and GEO strategy, goals in Google Analytics serve primarily to connect high-value actions (a lead, a purchase, qualified engagement) to the pages and traffic sources that drive them. If you are starting from a KPI-led approach, first define your tier-one indicators (acquisition, engagement, conversion) as outlined in the article on Google Analytics KPIs, then translate them into GA4 configuration (events, then key events) to build tracking you can genuinely act on, without unnecessary noise.

From Universal Analytics goals to GA4 key events: understanding what changed

"Goals" belong historically to Universal Analytics. With UA now retired, GA4 shifted to an event-based model: you no longer think in terms of goals configured within a view, but in terms of instrumented events, some of which are promoted to key events — GA4's replacement for UA's goals and conversions logic.

Why this shift matters for performance, conversion and ROI

This change is not cosmetic; it alters how you assign value to actions. In UA, you had destination goals, duration goals, and so on. In GA4, you first ensure the event exists (standard, enhanced measurement, or custom), then decide whether it represents a conversion (a key event). This approach supports:

 
• more granular measurement of micro-conversions (clicks, scroll depth, downloads) and macro-conversions (leads, purchases);
 
• better segmentation (landing page, device, source/medium, geography), useful for SEO and essential for GEO;
 
• more robust steering as zero-click behaviour grows: the focus shifts to post-click quality (engagement, conversion, value) rather than raw session volume.

Aligning measurement goals with KPIs without bloating your reporting

A common mistake is marking too many events as conversions "just in case". A more effective approach is deliberately minimal: 1 to 3 macro-conversions (for example, demo requests, purchases, booked meetings), complemented by a small set of micro-conversions that accelerate optimisation (for example, CTA click, form start, pricing page view). The objective is not to measure everything — it is to measure what enables an SEO or GEO decision (optimise a page, refine intent matching, remove friction, prioritise content).

Goals, events, key events and conversions: clearing up the terminology

These terms sound similar, but they carry different meanings in UA versus GA4. Getting the vocabulary right prevents confusion between what you want to achieve (a business goal) and what you actually measure (an event or key event).

What Universal Analytics goals measured: destination, duration, pages/screens and event

In Universal Analytics, a goal measured a specific action and reported it as a conversion (for example, a completed form, a newsletter sign-up, a purchase). The main goal types were:

 
• Destination: a specific page load (for example, a thank-you page). A classic setup uses a URL such as /merci.htm as the form confirmation page (source: Yumens).
 
• Duration: time spent within a session (useful, but often prone to false positives).
 
• Pages/Screens per session: the number of pages viewed during a session.
 
• Event: a tracked action (for example, a click on a "Call us" button).

UA also imposed structural limits: goals were configured at the view level, with a maximum of 20 goals per view (source: Yumens).

What GA4 measures today: events, parameters and marking an event as a key event

GA4 collects events (automatic, enhanced measurement, recommended, or custom). An event becomes a conversion when it is marked as a key event. The practical workflow is:

 
• define the action (business goal);
 
• instrument the event (with the parameters you need);
 
• mark it as a key event if you want it treated as a conversion.

For SEO and GEO, this approach makes it easier to connect "query → landing page → interaction → key event", producing more actionable insights than volume metrics alone.

Acquisition analysis: separating direct, referral and campaign traffic when reviewing conversions

Conversion analysis depends heavily on attribution. Two channels are particularly easy to misread:

 
• Direct traffic does not always mean a typed URL; it can reflect missing UTMs, certain in-app environments, or poorly configured redirects.
 
• Referral traffic can be "polluted" by self-referrals (misconfigured cross-domain tracking, payment gateways, authentication flows), which distorts journey and conversion data.

Before optimising a page, confirm that your sources, mediums and redirects are not undermining your diagnosis.

Migrating goals to GA4 key events: a step-by-step method, mapping and validation

Migration is not about recreating goals. You need to translate each UA goal into one or more GA4 events, then decide which should become key events. To minimise discrepancies, take a structured approach.

Auditing your goals, funnels, values and priorities (macro and micro-conversions)

Start with a structured inventory:

 
• a list of UA goals (and their types);
 
• any funnel definition (steps, pages, rules);
 
• any value assigned to the goal (if you used it for ROI estimation);
 
• business priority: macro-conversion (for example, a lead) versus micro-conversion (for example, a CTA click).

To keep the framework usable, you can also apply a SMART structure (specific, measurable, achievable, realistic, time-bound), which is widely recommended for designing coherent measurement goals (source: Digital Cover).

Translating a goal into a GA4 event: naming, parameters, conditions and de-duplication

For each UA goal, define:

 
• the GA4 event (for example, generate_lead, form_submit, purchase);
 
• the parameters needed for SEO/GEO analysis (page, CTA location, label, content type, and potentially geography);
 
• the conditions (when the event should fire);
 
• de-duplication (preventing double sends via double tagging, duplicate triggers, or page reloads).

Over time, a stable naming convention (for example, snake_case with action verbs) and thorough internal documentation materially reduce reporting errors.

Promoting an event to a key event, surfacing it in reports and aligning it with conversions

Once the event is collected and validated, mark it as a key event so it feeds your conversion reporting. This is also the moment to align language across teams: if a "conversion" in the business does not correspond to a GA4 key event, teams will end up reading inconsistent dashboards.

To go further on reading and making use of GA4 conversions, see the Incremys article on conversions.

Checking data collection: testing, consistency and tracking governance

Before any SEO or GEO analysis, confirm data quality:

 
• test in Real-time reports (triggering, parameters, no duplicates);
 
• exclude internal traffic (teams, agencies, suppliers);
 
• check cross-domain settings to prevent self-referrals;
 
• log every tracking change so you can interpret breaks in trends.

Bear in mind that some data loss is inevitable (private browsing, ad blockers, consent), which is why trend analysis matters most when collection is stable (source: Digital Cover).

Key-event configurations that matter for SEO and GEO

In SEO and GEO, the objective is not simply to count conversions, but to understand which content and which sources drive high-value actions. Your key events should reflect real user journeys (read, compare, validate, act).

GEO perspective: measuring visibility and conversions driven by generative AI answers

GEO changes how you evaluate performance: a significant share of searches now end without a click. Studies cite 60% of searches as "zero-click" (Squid Impact, 2025). In that context, GA4 still only measures post-click behaviour — but post-click becomes even more strategic: you need to demonstrate that the visits you earn (including those originating from AI environments when attribution allows) deliver genuine engagement and conversions.

According to Squid Impact (2025), visitors arriving from AI-generated answers are reportedly 4.4 times more qualified than those from traditional search. Segmenting key events by source/medium and comparing quality (engagement rate, conversion rate) therefore becomes a practical lever for GEO steering.

Selecting micro-conversions that differentiate organic traffic quality from AI-driven traffic

To distinguish quality, prioritise intention-led micro-conversions rather than overly generic signals:

 
• a click on a "request a demo" or "contact us" CTA (a click event, contextualised by position and label);
 
• a form start (useful for assessing offer attractiveness);
 
• views of key pages (pricing, customer stories, integrations), ideally paired with an interaction event (scroll, internal click);
 
• a PDF download (whitepaper) when it genuinely signals qualification.

Also segment by device: mobile accounts for 60% of global web traffic (Webnyxt, 2026), and Google reports a mobile abandonment rate of 53% when load time exceeds three seconds (Google, 2025). These gaps often explain conversion differences across segments.

Building a "content → action → lead" view for your SEO pages (guides, comparisons and landing pages)

An effective SEO view connects:

 
• the landing page (for example, a guide, a comparison page, a landing page);
 
• proof interactions (scroll, internal clicks to pricing or customer stories);
 
• contact actions (CTA, form, meeting booking);
 
• the final conversion (the primary key event).

This prevents a classic mistake: concluding that "content doesn't convert" when it is actually driving upstream micro-conversions that are critical in B2B.

Comparing GEO impact: visibility, engagement and conversions generated by generative AI answers

GA4 cannot capture visibility without a click. To compare GEO impact, adopt a dual reading:

 
• Before the click: impressions and queries (via Google Search Console) to spot patterns such as "impressions rising, clicks flat or falling" — common as AI summaries become more prevalent.
 
• After the click: GA4 engagement and key events (session quality, micro-conversions, macro-conversions), segmented by source where available.

Some analyses report organic traffic declines of -15% to -35% linked to the rise of AI overviews (SEO.com, 2026; Squid Impact, 2025). In that context, the objective becomes maximising value per visit and measuring robust conversions, rather than chasing volume alone.

Practical examples of Analytics goals and their GA4 equivalents

Below are three common scenarios — lead generation, engagement, and e-commerce — and how to translate them into GA4, moving from a "goal" mindset to an "event, then key event" approach.

Form submission: event design, de-duplication, B2B qualification (MQL/SQL) and attribution

UA legacy: a destination goal based on a thank-you page (for example, /merci.htm) is a straightforward approach (source: Yumens).

GA4 equivalent:

 
• a submission event (for example, generate_lead or form_submit) triggered on confirmed success (server OK response, or a verified success message displayed);
 
• de-duplication (preventing a second send on page refresh);
 
• useful parameters: form_id, form_name, page_location, cta_text, cta_position;
 
• marking the event as a key event for the "lead" macro-conversion.

In B2B, if you qualify MQL/SQL within your CRM, treat GA4 as a measure of intent and volume, then compare trends rather than expecting exact parity across systems — consent, de-duplication and counting rules all introduce differences.

Duration goals: GA4 engagement-based alternatives to avoid false positives

A duration goal in UA can overstate quality: an open tab does not indicate active reading. In GA4, prefer engagement-led alternatives:

 
• engaged sessions (GA4's built-in engagement logic);
 
• interaction events (meaningful scroll depth, table-of-contents click, video watch);
 
• engagement time thresholds combined with an action (for example, at least X seconds and a click through to a key page).

Device differences also matter here. Benchmarks indicate an average session duration following a paid search click of 1 minute 38 seconds on mobile and 2 minutes 13 seconds on desktop (Start'Her, 2026). These figures suggest setting realistic, device-segmented engagement thresholds rather than a single universal duration goal.

E-commerce: purchase, basket, checkout and value (when a goal becomes a journey)

In e-commerce, the "goal" is not a single action — it is a journey: add to basket → begin checkout → purchase. UA often handled this via goals and funnels. GA4 supports it well through e-commerce events (for example, add_to_cart, begin_checkout, purchase) and value measurement (revenue) when available.

For context, some analyses cite an average Google Search paid conversion rate of 3.75% (WordStream, 2025) and an e-commerce paid search conversion rate of 4.8% (Odiens, 2025). Use such benchmarks cautiously: product mix, UX and brand recognition drive significant variation.

SEO interactions: tracking CTA clicks, downloads and outbound links

To accelerate SEO optimisation, instrument interactions that demonstrate intent:

 
• CTA clicks (demo request, contact, pricing);
 
• downloads (guides, product sheets);
 
• outbound clicks to partners (when they reflect genuine business value).

The key is to contextualise every action (page, position, label) — otherwise you end up with a raw click total that cannot inform decisions. Analytically, aim for "fewer events, richer parameters".

Using GA4 to analyse your content goals and key events

Once key events are in place, the value comes from analysis: identifying what converts, understanding why, and prioritising editorial and UX improvements.

Measuring SEO contribution: landing pages, journeys, triggered key events and high-performing content

From an SEO perspective, start with organic landing pages and work through the journey:

 
• which pages drive sessions and then engagement?
 
• which pages trigger micro-conversions and then macro-conversions?
 
• which assisting content types appear before conversion (comparisons, guides)?

This approach reduces decisions based on vanity metrics and helps you invest in content that creates value, even when the final conversion happens on a different page.

Comparing by source: organic, direct, referral and campaigns (diagnosing conversion gaps)

Compare key events by source/medium and landing page, then qualify anomalies:

 
• a rise in direct traffic alongside a drop in conversion: suspect missing UTMs or an attribution break;
 
• a rise in referral from a technical domain: suspect self-referrals or cross-domain configuration issues;
 
• stable organic traffic but declining micro-conversions: suspect an intent mismatch (SEO promise versus content delivered) or UX friction.

Building an analysis funnel to pinpoint friction and prioritise improvements

For a lead-generation site, model a simple funnel:

 
1. entry via SEO content (landing page);
 
2. proof interaction (scroll or click to a key section);
 
3. CTA click;
 
4. form start;
 
5. form submission (primary key event).

You can then identify drop-off points and prioritise accordingly: refine a CTA, simplify a form, strengthen a section, or improve load speed. This logic aligns with why funnels are valuable for spotting abandonment (source: Digital Cover).

Common pitfalls and best practice for GA4 tracking configuration

Many GA4 analyses fall short not because of the tool itself, but because goals are poorly defined or data collection is unstable. Below are the most costly pitfalls in SEO and GEO contexts.

Poorly defined goals: "easy" indicators versus actionable indicators for SEO and GEO

Avoid turning weak signals into conversions: a simple scroll or a standalone session duration does not prove business intent. Prefer actionable indicators such as CTA clicks, form starts, key offer-page views, contact requests and purchases.

Attribution issues: self-referrals, redirects, missing parameters and tagging errors

Attribution errors distort SEO and GEO interpretation:

 
• self-referrals (cross-domain issues) that turn legitimate journeys into noisy referral traffic;
 
• redirects that break sessions and create spurious new sources;
 
• missing UTMs that inflate the "direct" channel;
 
• inconsistent event naming across pages, making comparisons impossible.

Data quality: naming conventions, documentation, test environments and ongoing maintenance

Core best practice includes:

 
• stable, documented naming conventions;
 
• a test and validation environment before deployment;
 
• regular (ideally weekly) checks of trends, particularly after new campaigns or site changes (source: Digital Cover).

Finally, remember that tracking changes are not retroactive, and logging the dates of those changes helps you interpret breaks in data (source: Yumens).

Putting results in context: using SEO, SEA and GEO statistics to support your goals

Without context, a rise or fall in conversions can lead to the wrong decision. Sector benchmarks and macro trends (SERPs, AI, paid search) help set realistic expectations and interpret performance more accurately.

When to use benchmarks: interpreting variation and setting realistic targets

Use statistics to frame analysis, not to copy a target:

 
• SEO: understand click concentration (for example, the top three results) and the impact of zero-click behaviour on your editorial strategy (see SEO statistics).
 
• SEA: compare orders of magnitude (conversion rates, CPA), particularly if you mix paid and organic acquisition (see SEA statistics).
 
• GEO: factor in the impact of AI answers (zero-click, traffic declines, the growing importance of post-click quality) and build measurement around key events (see GEO statistics).

Going further with Incremys

Centralising GA4 and Google Search Console via API to connect content, SEO/GEO visibility and ROI

Incremys can centralise Google Analytics (GA4) and Google Search Console via API, making it easier to connect "before the click" data (queries, impressions, pages) with "after the click" data (engagement, key events, conversions). The aim is to support a content- and ROI-led view without multiplying exports, and without losing consistency in naming conventions and segmentation.

FAQ on Google Analytics goals, GA4 key events and conversions

How do you set up goals in Google Analytics?

In GA4, you no longer configure a goal in the same way as in Universal Analytics. Start with the business action you want to measure (for example, a form submission), instrument the corresponding event (standard or custom) with useful parameters (page, CTA, and so on), then mark it as a key event so it is treated as a conversion.

Do goals still exist in GA4?

Not in the UA sense. The "Goals" feature from Universal Analytics has been replaced by an event-based model: the operational equivalent of a goal is an event that you choose to mark as a key event.

What is the difference between an event, a conversion and a key event in GA4?

An event is a measured interaction (click, scroll, form submission, purchase). A key event is an event you promote to conversion status. In GA4, a conversion therefore corresponds, in practice, to counting an event that has been marked as a key event.

How do you migrate a Universal Analytics destination goal to GA4?

If your UA goal relied on a thank-you page, recreate the logic in GA4 by triggering an event when the confirmation URL is displayed (or, better still, when submission success is confirmed server-side). Add parameters (form name, originating page) and mark the event as a key event.

How do you track form submissions reliably without duplicates?

Trigger the event only on confirmed success (server-side validation or a verified success message), prevent double firing (on refresh, via double tags, or duplicate listeners), and confirm in Real-time reports that the event fires only once per submission. Then segment by landing page and source/medium to assess SEO and GEO effectiveness.

How do you set up a duration or engagement goal without biasing your analysis?

Avoid converting time alone into a conversion signal. Prefer interaction-led engagement signals (meaningful scroll, internal click, video watch) or a combined rule such as engagement time plus a specific action. Segment by device, as behaviour differs significantly between mobile and desktop users.

How do you measure an e-commerce goal in GA4 (purchase, basket, checkout)?

Measure the full journey: add_to_cart, begin_checkout, purchase, then mark purchase as the primary key event. Analyse drop-offs at each step (UX friction, costs, delivery concerns) and value (revenue) by source/medium.

Which goals should you prioritise for a B2B lead-generation site (MQL/SQL)?

Prioritise 1 to 2 macro-conversions (contact form submission, booked meeting), then highly intention-led micro-conversions (pricing page click, demo request click, form start, high-value document download). The aim is to accelerate SEO optimisation loops without relying solely on the final conversion.

How do you analyse the SEO impact of a piece of content on GA4 key events?

Start with organic landing pages, then review engagement, micro-conversions and macro-conversions triggered within the session or broader journey. Identify content that assists conversions (proof pages, comparisons) even when it is not the last page before conversion.

How do you distinguish direct traffic from referral traffic when analysing conversions?

Do not take these channels at face value. Investigate sudden increases, check UTM coverage, review redirects and cross-domain configuration, then use the dedicated diagnostics: direct and referral.

How do you measure the GEO impact on your goals (traffic from generative AI answers)?

Measure what is attributable in GA4 (sessions, engagement, key events by source/medium where available) and complement with Google Search Console for pre-click visibility (impressions, queries, pages). Keep context in mind: zero-click can account for 60% of searches (Squid Impact, 2025), so post-click quality becomes a central criterion.

Why do my GA4 conversions not match my CRM or other data sources?

Discrepancies typically stem from differences in definitions (lead versus MQL), de-duplication logic, time windows, consent (under-measurement) and counting rules (a GA4 event does not always correspond to a CRM "transaction"). Focus on comparing trends over a stable collection period and document every tracking change.

Which SEO, SEA and GEO statistics should you consult to contextualise performance?

For a data-driven view, rely on solid reference points covering click dynamics, conversion rates and AI trends: SEO statistics, SEA statistics and GEO statistics.

To keep structuring your measurement approach and your SEO/GEO strategy, explore further analysis on the Incremys blog.