Digital Marketing
GA4 Implementation
GA4 works out of the box and measures almost nothing useful in that state. The value is entirely in the configuration.
GA4 implementation configures events, conversions and the settings most installations skip. The Nexclick sets it up so the reports mean what you think they mean, because a default GA4 install measures pageviews well and almost nothing else that matters commercially.
Is this you?
What usually prompts the call
- GA4 was installed by pasting a tag and nothing has been configured since.
- Your reports show sessions and bounce rate and nothing about enquiries.
- Nobody trusts the data enough to make a decision from it.
- You migrated from Universal Analytics and the reports no longer answer your questions.
What we do
The actual deliverables
Things that appear on an invoice, not adjectives.
- Event architecture designed first
- A naming convention and parameter structure decided before implementation. GA4 event data is difficult to restructure retrospectively, so this matters more than in Universal Analytics.
- Conversions mapped to business events
- Marked as key events with values where they exist, so reporting reflects commercial outcomes rather than form interactions.
- The settings almost everyone misses
- Data retention extended from the default, internal traffic excluded, unwanted referrals filtered, cross-domain configured, reporting identity set deliberately.
- Enhanced measurement reviewed rather than accepted
- GA4 auto-tracks scrolls, outbound clicks and file downloads by default. Some of it is useful and some inflates engagement metrics misleadingly.
- Ecommerce tracking where relevant
- Full purchase funnel with item-level data, so category and product performance is visible rather than a single revenue figure.
- Custom dimensions and audiences
- The segments you actually need — by service interest, lead quality, customer type — registered properly so they appear in reports.
- Useful default reports
- Explorations built for the questions your team asks, because the standard GA4 interface answers few of them without configuration.
Checklist
The GA4 settings nearly everyone misses
Open your own property and check these. Most installations we inherit have configured fewer than half, and the first three are the ones that cost you data you cannot get back.
- 01Data retention extended from the 2-month default to 14 months.
- 02Internal traffic from your office and your agency excluded.
- 03Unwanted referrals — payment gateways, booking platforms — filtered out.
- 04Cross-domain measurement configured if you use a separate checkout or booking domain.
- 05Reporting identity set deliberately rather than left on the default.
- 06Key events marked, with values where a value exists.
- 07Enhanced measurement reviewed rather than left fully on.
- 08Session timeout adjusted if your typical visit is unusually long or short.
- 09Google Ads and Search Console linked.
- 10Custom dimensions registered, not just sent as parameters.
- 11BigQuery export enabled if volume justifies it — free, and it preserves raw data.
- 12Consent mode configured so declined visitors are handled correctly.
- 13Attribution model chosen deliberately rather than left at default.
- 14Debug traffic excluded from reporting.
- 15Someone other than a departed employee owns admin access.
How it works
Step by step, with timeframes
Timeframes are typical rather than guaranteed, and they assume we get account access and approvals when we ask.
- 01Week 1
Design the event structure
Naming, parameters and what each event means, agreed in writing. Difficult to change once data has accumulated.
- 02Week 1–2
Implement
Events, conversions, settings and ecommerce configured through Tag Manager, tested in preview and DebugView before publishing.
- 03Week 2–3
Verify against reality
Real submissions and transactions checked end to end. The most common failure is an event that fires in testing and not in production.
- 04Week 3–4
Build reports and document
Explorations for the questions your team actually asks, plus written documentation of the measurement model.
What you get
Reporting and ownership
- A documented event and parameter structure, so the data stays coherent as the site grows.
- Conversions marked as key events with values, reflecting commercial outcomes.
- The overlooked settings configured — retention, internal traffic, cross-domain, reporting identity.
- Explorations built for the questions your team actually asks.
- Verification against real submissions, not just preview mode.
Tools and platforms
- Google Analytics 4
- Google Tag Manager
- GA4 DebugView
- BigQuery export where volume justifies it
- Looker Studio
Timeline
How long this actually takes
Three to four weeks. Two things worth knowing. GA4 data retention defaults to two months on the free tier and can be extended to fourteen — changing it does not apply retrospectively, so it is the first setting to fix and the one most commonly missed. And GA4 event structure is genuinely hard to change once data has accumulated, which is why the design stage comes before implementation rather than emerging from it.
Pricing model
Fixed-price project
Fixed price. Frequently bought alongside the wider tracking setup, since GA4 alone leaves ad platform conversion tracking and consent unaddressed.
Questions
GA4 Implementation questions
Why does GA4 not show what Universal Analytics did?
Because the data model changed fundamentally — everything is an event rather than a pageview with dimensions. Some familiar metrics do not exist, others are calculated differently. Reports have to be rebuilt around questions rather than translated from the old interface.
What is the most costly setting to leave at default?
Data retention. It defaults to two months on the free tier, and extending it to fourteen does not apply retrospectively — data older than the retention period is gone permanently. It is the first thing to change and among the most commonly missed.
Should we enable BigQuery export?
Yes, if you have meaningful volume. It is free at reasonable scale, preserves raw event data beyond the retention limit, and allows analysis the interface cannot do. It costs nothing to enable and cannot be applied retrospectively either.
Is enhanced measurement useful?
Partly. Outbound clicks and file downloads are genuinely useful. Scroll tracking inflates engaged sessions in a way that flatters the reports without meaning much. We review each rather than leaving the lot switched on by default.
How do we get GA4 to report on enquiries rather than sessions?
By defining conversions against real business events and marking them as key events with values. The default install measures pageviews well and commercial outcomes not at all, which is why so many teams conclude GA4 is useless when it is simply unconfigured.
Can event structure be changed later?
With difficulty. Historic data keeps its original structure, so a renamed event splits your history in two. That is why the design stage comes first — it is one of the few genuinely hard-to-reverse decisions in analytics.
Last reviewed 28 July 2026.
Tell us what you are trying to fix
A 20-minute call, no pitch deck. The Nexclick will tell you what we would do, roughly what it costs, and whether we are the right people for it.