How does "E-commerce" differ from "Service / Lead gen" and how does it affect my score?

In step 4 of the wizard, you choose your website's business model. This isn't a survey question - the choice genuinely changes the scope of the audit: it decides which checkpoints are included in the examination and how the Quality score is calculated.

Why we ask at all

GA4 measures an online store in a completely different way than a service website. A store lives on purchase events: view_item, add_to_cart, begin_checkout, purchase, product data and transaction values. A service site lives on leads: form submissions, phone and email clicks, paths to contact. An audit that checked a law firm for the completeness of purchase event parameters would generate noise - dozens of "errors" that have nothing to do with the business, and a deflated score that tells you nothing.

That's why the checkpoint catalog has three scopes: checkpoints shared by everyone, e-commerce-only checkpoints, and service-site-only checkpoints. Your choice in step 4 decides which set applies.

What we check in each variant

E-commerce - you sell products online.

On top of the shared checkpoints, the audit covers the full e-commerce category, including:

•presence and correctness of the key purchase events across the whole path (from product view to purchase),

•transaction data completeness: order values, taxes, shipping costs, presence and uniqueness of transaction_id (deduplication),

•product data quality in the items[] array: identifiers, categories, brands, quantities,

•discount code tracking (the coupon parameter),

•purchase funnel analysis - where users abandon the process and whether it's a UX problem or a measurement failure.

Service / Lead gen - you generate leads or sell services.

The audit skips the store-only checkpoints and concentrates on what determines lead measurability: conversion event configuration (e.g. generate_lead with a value parameter), form submissions, call tracking and conversion paths.

Shared checkpoints - always checked, regardless of the choice: the property's administrative settings (currency, time zone, data retention), tag implementation, event naming and recording continuity, channels and attribution, data quality (bots, duplicates, tracking gaps), integrations (Google Ads, Search Console, BigQuery), technical limits, and privacy with Consent Mode v2.

How this affects the Quality score

Checkpoints outside your scope receive the "N/A" status and are entirely neutral for the score - they neither lower nor raise it. This keeps the Quality score of a service site and a store equally meaningful, even though they're computed from different checkpoint sets. The practical consequence: don't compare a store's score directly with a service site's score - compare a project's score against its own history.

What to choose in borderline cases

•A store with a blog module - E-commerce. Sales drive the revenue; the blog is a reach channel.

•A service business with a small add-on store (e.g. a law firm selling e-books) - pick the model matching your primary revenue source. If e-commerce is marginal, the service variant gives a clearer report.

•A marketplace / platform with payments - E-commerce; the transaction path is the heart of the measurement.

•You generate leads online but close sales offline - Service / Lead gen. What counts is what the website measures, not where the deal is signed.

Can I change the business type later?

You set the business type once, in step 4 of the wizard - it can't currently be edited in an existing project's settings. If you pick the wrong model, the simplest fix is to start a fresh project with the right type. Keep in mind that changing the model changes the set of checkpoints applied, so scores from before and after the change shouldn't be compared directly.

Related articles:

•What do the 62 checkpoints cover?

•What do the statuses mean: Pass, Warning, Error, N/A?

•How is the Quality score calculated?