Sessions with a success
6.4%32 ÷ 500 × 100
The share of sessions with at least one occurrence of the chosen success event. A session counts once even when the event occurs more than once within it.
Independent software guidance for creators and small teams.
MEASUREMENT FOR SMALL MARKETING TEAMS
Start with Google Analytics 4 for channel and conversion reporting, Microsoft Clarity to investigate page friction, or PostHog for product funnels and controlled experiments. These roles can complement one another; you do not need all three to answer one question.
Documentation checked September 9, 2026. This is a source-based selection guide, not a hands-on vendor benchmark. The examples below are hypothetical, not ToolMerit traffic or experiment results.
A SHORTLIST BY JOB, NOT A UNIVERSAL WINNER
For a newsletter site, the first useful question may be which channels produce successful sign-ups. For an app, it may be where new accounts fail to activate. For a landing page, it may be why people repeatedly click an element that does nothing.
Scroll the comparison sideways. Keyboard users can focus it and use the arrow keys.
| Tool | Useful starting job | Cost basis | Important boundary |
|---|---|---|---|
| Google Analytics 4Channel & key-event reporting | Compare acquisition channels with recorded outcomes. Its Traffic acquisition report connects session traffic sources with metrics including session key event rate. | The standard Analytics offering is free. Budget time for event setup, consent handling, reporting and maintenance; a free account is not a finished measurement system. | A tag loading successfully does not prove the right outcome was recorded. Define the event, validate it and check report scope before judging a channel. |
| Microsoft ClarityHeatmaps & session replay | Inspect captured interactions around an obstructed form, confusing navigation or dead clicks. Heatmaps, recordings and funnels help form a specific usability hypothesis. | Microsoft lists Clarity as a free service without traffic limits in its official FAQ. There is still review and data-governance work. | Replays are observational evidence, not proof that a change caused more sales. Review masking and consent before capture. Microsoft excludes sites and apps targeting people under 18. |
| PostHogProduct funnels & experiments | Analyze instrumented product steps and test a variant against a control. Its experiment system combines randomized allocation, exposure tracking and outcome metrics. | The monthly free allowances include 1 million analytics events and 5,000 replay recordings. These are separate meters; experiments are billed with feature flags. Check the meters you enable before adding paid usage. | Useful experiments need a measurable outcome, reliable exposure events and enough observations. Adding an SDK or looking at two percentages does not establish a winner. |
THE EVENT CONTRACT
For a lead form, agree on the success condition with the person responsible for the form and the person using the report. A practical contract is to record a lead when the application confirms a valid submission was accepted, not when the visitor presses Submit.
The visitor tried to send the form. Validation or a network error may still prevent success.
The application confirms receipt. Record the successful lead once under the agreed event rule.
A later business check establishes fit. Track it separately; it is not implied by receipt.
Google documents generate_lead for lead acquisition and qualify_lead for qualification. Choose which event should count as a key event. Do not assign invented revenue to a form submission or send form text, email addresses or other personal details as ordinary analytics parameters.
Keep the trigger owner, event name, success rule and exclusions in a shared note. If both a tag manager and hard-coded script send the same event, agree on one implementation path. A retry, reload or second click must not silently create another successful lead for the same accepted submission.
READ THE RATE BEFORE THE CHART
Suppose one landing page records 500 sessions, 40 occurrences of your chosen success event, and 32 sessions containing that event. Some sessions contain repeats. The two calculations answer different questions:
32 ÷ 500 × 100
The share of sessions with at least one occurrence of the chosen success event. A session counts once even when the event occurs more than once within it.
40 ÷ 500 × 100
Eight event occurrences per 100 sessions. Calling this an 8% converting-session rate would overstate what the example establishes.
Google's metric definitions distinguish the rate for any key event from the rate for a specific key event. Use the same selected event, date range, time zone and eligible session population on both sides of a comparison. Neither calculation above tells you how many unique people became customers.
Use the conversion calculator after choosing a consistent numerator and denominator. For campaign links, the UTM builder helps keep naming consistent; tagging a URL does not validate the destination's event collection.
A SMALL ACCEPTANCE RECORD
These are proposed checks for your implementation, not tests ToolMerit has run on these vendors. Use a test environment and harmless test data where possible. Record the observed result and the person responsible for any failure.
A confirmed valid submission should produce the agreed success event. A validation error or server rejection should not. Check the application result and the event payload, not only a thank-you message.
Exercise a double click, a retry and a thank-you-page reload. Verify the intended counting rule. Check whether a plugin, tag manager and theme are sending overlapping events; never assume the dashboard automatically removes duplicates.
Check accepted, refused and withdrawn choices against the intended collection behavior. Review replay output with synthetic data. Clarity's masking changes are not retroactive and can take up to an hour to apply to new recordings. Fix masking before collecting real sensitive content.
Document the chosen event, counting unit, eligible population and report dates. Assign an owner to recheck collection after form, checkout, theme or consent changes. For experiments, also verify that assignment and exposure are recorded as designed.
Installing every available tracker is not an acceptance criterion. The smallest useful setup is the one that answers the current question with understandable data. Our performance and privacy tools guide covers related loading and consent decisions; no tool on this page is a blanket compliance guarantee.
RELATED MEASUREMENT DECISIONS
Use these focused guides for a specific decision. Attribution, affiliate tracking and ad-cost calculations need their own definitions; they should not be treated as interchangeable conversion reports.
Compare affiliate tracking platforms by billing connection, attribution rules, refund handling, partner recruitment, payouts, and program complexity.
Compare GA4 replacements by reporting job, deployment, privacy model, funnels, replay, price, and the migration work each route requires.
Build a B2B SaaS search system across keyword research, buyer decisions, product proof, technical SEO, authority, pipeline, and measurement.
Compare attribution systems by commercial motion, identity unit, authoritative outcome, model, evidence drill-down, activation, and pricing basis.
Compare B2B data systems by the fit, timing, reachability, verification, workflow, credit, and governance evidence they can deliver.
Choose an ad-operations layer by what it can safely observe, edit, automate, create, or reconcile across live advertising platforms.
Solve for CPM, total advertising cost, or impressions from two known values and expose the unit economics behind the result.