If you ask a Japanese marketer "have you installed GA4?", most will say yes. A 2022 survey of 1,009 web marketing professionals in Japan found 71% already had GA4 installed[1] — and that was before Universal Analytics sunset.
If you ask the same people a follow-up question — "have you set up report automation in GA4?" — the answer drops to 11%[2].
One country. One tool. A 60-point gap between "installed" and "actually using."
This post walks through what Japanese market surveys tell us about that gap, why it exists, and what it implies for anyone shipping analytics tooling — not just in Japan. Because the pattern here isn't a Japan-specific quirk. It's the default state of any organization that treats "install GA4" as the finish line.
TL;DR
- GA4 adoption in Japan: 71%. Report automation: 11%. Full-custom setup: 23%.
- 50.7% of marketers in a 2024 Japanese survey cite "how to actually use the data" as their biggest pain point — above setup, above integrations.
- The gap between "installed" and "used" is not a training problem. It's a dashboard-shape problem: the question marketers need to answer and the screen GA4 shows them are not the same shape.
1. The gap nobody talks about
Most "is GA4 popular in your country?" conversations end at installation rate. Which is fine as a vanity number, and useless as a planning number.
Here is what the full staircase looks like in Japan, pulled from three independent surveys:
| Stage | Metric | Share |
|---|---|---|
| Installation | "GA4 is installed" | 71% (2022, Digipro)[1] |
| Basic setup | Account + tag deployment completed | 31–35% (2023, Auriz)[2] |
| Custom setup | Custom conversions / events configured | 23% (2023, Auriz)[2] |
| Full utilization | Report automation reached | 11% (2023, Auriz)[2] |
| Experience | "I find it hard to figure out how to actually use the data" | 50.7% (2024, Fujifilm BI)[3] |
The drop from 71% to 11% is the entire story. Installation is the easy step — it's a tag, it's a prop, it's a migration guide.
Everything downstream — custom events, cohort analysis, report automation — requires answering a question the tool does not ask for you: what exactly do you want to know every morning?
Most organizations never answer that question, so the tool sits at 11%.
2. What the data shows
Market context
Before the "how is GA4 being used" part, the scale worth knowing:
Japan's B2C eCommerce market hit ¥26.1 trillion (~US$170B) in 2024, up 5.1% YoY, per METI's 2024 eCommerce Market Survey[4]. The physical-goods slice alone is ¥15.2 trillion. That's the surface area where GA4 setup quality directly translates to revenue decisions.
The Auriz survey (2023, n=248)
Auriz surveyed 248 marketing professionals actively running ad operations in Japan. They asked: "Which GA4 setup steps have you completed?" (multiple choice)[2]
| Setup step | Completion rate |
|---|---|
| Created account | 35% |
| Installed GA4 tag | 31% |
| Configured custom conversions | 23% |
| Set up report automation | 11% |
Only 25% had actually completed migration from UA. 28% were mid-migration. 29% had not started. 6% had stopped using Google Analytics entirely.
So within this sample, less than a quarter were fully migrated — and within that quarter, less than half had configured report automation. Roughly 1 in 10 of the entire surveyed population had reached the point where GA4 is producing automated output daily.
The Fujifilm BI survey (2024, n=535)
A year later, Fujifilm Business Innovation surveyed 535 marketing professionals on what they felt hard about access-analytics tools (not just GA4, but GA4 is the dominant tool in the sample)[3]:
| Pain point | Share citing it |
|---|---|
| How to optimally use the data | 50.7% |
| Linking data to goals and KPIs | 49.7% |
| Deciding what to focus on | 46.0% |
| Setup for accurate data collection | 40.2% |
| Integration with external tools | 30.8% |
Two things to notice:
- The top three pain points are all about interpretation, not collection. "Setup" only shows up at rank 4.
- Asked what support they need, respondents ranked "data analysis and interpretation" (55.9%) and "translating analysis into action plans" (48.2%) at the top[3].
The problem isn't that the tool is broken. The problem is that the tool answers the wrong question.
The Digipro survey (2022, n=1,009)
The Digipro study established the 71% installation baseline back in 2022[1]. The sample was larger and less filtered-for-seniority than the Auriz one, which is why its installation rate is higher than Auriz's "migration complete" rate — the two are measuring different states (installed vs. fully migrated + configured).
3. Why this gap exists
Three reasons I keep seeing, in roughly decreasing order of scale.
Reason 1: GA4 is shaped for sessions, but business is shaped for revenue
GA4's default home screen leads with users, sessions, events, engagement rate. A business owner's default question is: "I spent ¥50,000 on Instagram ads yesterday. How much of that came back as revenue, and which campaigns?"
Answering that in GA4 requires building an exploration, joining event data with transaction data, and configuring channel grouping. It's possible, it just isn't the path of least resistance.
So people install GA4, see that the home screen doesn't answer their actual question, and... never come back to finish the setup. Installation rate 71%, utilization rate 11%.
Reason 2: The "11% who automate" are the ones who had someone to delegate to
Report automation in GA4 is not a one-click feature. It's a multi-step configuration: exploration templates, scheduled email, BigQuery export, Looker Studio wiring, or custom dashboard work.
Every one of those steps requires either (a) a dedicated analytics role, or (b) an external consultant. Organizations without either stop at the "I can load the default Home screen" level.
The 11% figure is less about skill and more about whether the organization has budget headroom for an analytics function at all. In a survey skewed toward SMB eCommerce operators, that headroom is rare.
Reason 3: Training content optimizes for the install stage
If you search "GA4 setup tutorial" in any language, you will find an enormous volume of material about tag deployment, migration, property creation. You will find dramatically less on "what is the weekly dashboard a store owner should build and look at every Monday."
The content market is shaped the same way installation is: rewards for the easy step, silence at the hard step. So marketers graduate from tutorials feeling confident about setup and no more confident about use.
4. What to actually do
Three things. Ordered by effort.
1. Pick one question, before you open GA4
Write down the single question your team needs answered every Monday morning. Make it concrete: "Which channel drove the most revenue last week at what AOV?" beats "How is our marketing doing?"
Now: can you answer that question in GA4 in under 60 seconds, starting from a cold login? If no, the gap isn't skill — it's configuration. Configure toward that question. Ignore the other 40 reports.
2. Instrument revenue-side events before you perfect pageview events
Most GA4 implementations spend 80% of setup budget on pageview, scroll depth, and engagement events. They spend 20% on purchase, add_to_cart, begin_checkout.
This is backwards for eCommerce. The revenue-side events are the ones that answer the question you actually asked in step 1. If purchase events are not firing cleanly with value, transaction_id, and items, no amount of downstream setup will produce a useful Monday dashboard.
3. Decide if GA4 is your reporting surface, or your raw data layer
This is the hard one.
If GA4 is your reporting surface — where decision-makers read numbers — you are signing up for the "build 3-5 custom explorations + a Looker Studio dashboard" project. That's real work. Budget for it.
If GA4 is your raw data layer — where events are collected and then read by something else — then accept that, and point that something else (BigQuery, a reverse-ETL tool, a specialized analytics product, or an internal dashboard) at GA4 output. Stop trying to read GA4 directly.
Most of the "11% gap" organizations I've talked to are stuck because they're trying to do both with the same tool and the same setup budget. Picking one lane unsticks the configuration.
So what?
The gap between "71% installed" and "11% automated" is not going to close by writing better GA4 tutorials. Every additional tutorial optimizes for the step that isn't the bottleneck.
It closes by one of two things:
- Organizations decide GA4 is worth a dedicated analytics role and invest accordingly (slow, expensive, uncommon outside larger companies)
- Something else takes the specific "read revenue by channel on Monday morning" job off GA4's plate, so GA4 can be the raw data layer it's actually good at being (faster, cheaper, increasingly common)
If you're shipping analytics tooling in 2026: the question isn't "how do we replace GA4." The question is "which slice of 'use GA4 data' can we make so trivial that the 11% figure becomes 50% for our specific use case."
That's the gap worth working on.
I'd love to hear from folks outside Japan: does the 71% / 11% pattern match what you see in your market? Or is there a country / vertical where the utilization curve actually bends upward? Drop it in the comments.
I'm building RevenueScope, a revenue-first analytics layer that sits next to GA4 and answers the "which channel drove revenue last week at what AOV" question from one dashboard — for the 89% of organizations not going to build that surface themselves.
This post originally appeared in Japanese on the RevenueScope blog. Canonical source is set accordingly.
References
- Hagakure (Digipro) — "GA4 Adoption Survey," July 2022 (n=1,009)
- Auriz — "Google Analytics 4 Utilization Survey," October 2023 (n=248)
- Fujifilm Business Innovation — "Access Analytics Tool Usage Survey," December 2024 (n=535)
- METI Japan — "FY2024 Electronic Commerce Market Survey," August 2025
Top comments (0)