Exporting a PDF from Google Analytics takes a few clicks. Getting the person who receives it to understand it is the hard part. Most web analytics reports that land in a client’s inbox are a dashboard screenshot with no explanation, so the reader checks whether traffic went up, shrugs, and closes the file.
A useful web analytics report works differently. It picks the few numbers that connect to a business goal, compares them with an earlier period, and says in plain language what changed and what to do next. That is the step that turns website analytics into decisions about content, campaigns and design.
This guide covers what a website analytics report should include, how a report differs from a dashboard, which Google Analytics 4 (GA4) metrics changed meaning after Universal Analytics was retired, how to read a report quickly, and a one-page template you can reuse for every client or project.
The short version:
- A web analytics report covers a fixed period, compares it with an earlier one and says what to do next. A dashboard only shows the numbers.
- Five to eight metrics tied to one business goal are more useful than a 20-page export.
- GA4 redefined bounce rate and renamed conversions to key events, so old Universal Analytics benchmarks no longer apply.
- Since May 2026, GA4 shows visits from ChatGPT, Gemini and other chatbots in a separate AI Assistant channel.
- GA4 and Looker Studio cover most reporting needs for free.
Table of contents
- What is a web analytics report?
- Dashboard vs report: what’s the difference?
- What to include in a web analytics report
- GA4 terms that confuse clients
- How to read a web analytics report in five minutes
- How to build a web analytics report people actually read
- Web analytics report template: a one-page monthly structure
- Tools for web analytics reporting
- Common web analytics reporting mistakes
- Web analytics report questions: metrics, GA4 exports and reporting frequency
- Start with one page
What is a web analytics report?
A web analytics report is a summary of how a website performed during a set period, written for a specific reader. It takes data from tools such as GA4 and Google Search Console, narrows it to the metrics that matter for that reader’s goals, compares them with a previous period or a target, and explains the result.
The explanation is what the reader is paying for. A table showing 4,210 sessions says very little on its own. A sentence such as “organic sessions rose 18% after the new service pages went live, and those visitors requested quotes twice as often” tells the owner where to put next month’s budget.
Reports only work when the data behind them is collected the same way every month. If tracking breaks for a week, or someone changes which events count as leads, the comparison falls apart. Consistent tracking comes first. The report comes second.
Dashboard vs report: what’s the difference?
People use the two words interchangeably, but they do different jobs. A dashboard is a live screen you check to monitor performance. A report is a document sent on a schedule that interprets what happened during a period and recommends what to do about it.
| Dashboard | Report | |
|---|---|---|
| Main job | Monitoring what is happening now | Explaining what happened and what to do next |
| Data | Live or refreshed daily | A fixed date range compared with an earlier one |
| Typical reader | Marketers and site managers who check numbers often | Owners, clients and managers who make decisions |
| Level of detail | Many metrics, filters and drill-downs | A short list of metrics with written context |
| Format | Interactive screen | PDF, email, slides or a shared doc |
| Example tools | GA4 Reports snapshot, Looker Studio | GA4 scheduled emails, Looker Studio PDF, a written summary |
Most sites need both. The dashboard catches problems early, like a checkout page that suddenly stops recording purchases. The report is what you bring to the monthly conversation. If your report looks exactly like your dashboard, it is probably missing the explanation.
What to include in a web analytics report
Choose metrics by asking what the website is for. A lead generation site cares about forms and calls. An online store cares about revenue and checkout drop-off. A blog cares about returning readers and newsletter signups. The sections below cover what most website analytics reports need, and you can cut any section that doesn’t connect to the goal.
Traffic and acquisition
Start with how many people arrived and where they came from. In GA4, the Traffic acquisition report groups sessions by channel, such as Organic Search, Direct, Referral, Organic Social, Paid Search, Email and AI Assistant. Show the top channels, the change from last period, and one line on why the biggest movement happened.
Sessions and users are enough for most readers. Leave new versus returning users out unless the reader has a goal tied to loyalty, such as a membership site or a newsletter-driven blog.
Engagement rate, bounce rate and engagement time
This is where old reports go wrong most often. In GA4, a session counts as engaged when it lasts longer than 10 seconds, includes a key event, or includes at least two page views. Ten seconds is the default timer. You can raise it to as much as 60 seconds for each web data stream, and engagement rate shifts when you do. Engagement rate is the share of sessions that were engaged, and bounce rate is the opposite: the share that were not. Google explains both metrics in its GA4 engagement rate and bounce rate guide.
That definition is different from the Universal Analytics bounce rate, which counted any single-page visit as a bounce. A blog post that someone reads for three minutes before leaving was a bounce in Universal Analytics. In GA4 it is an engaged session. So when a client compares today’s bounce rate with a benchmark from an old article, they are comparing two different metrics that share a name. Say so in the report. Also, most GA4 standard reports don’t show bounce rate or engagement rate by default, so you need to customize the report to add them.
Average engagement time is the closest replacement for the old time on page. A longer time can mean people are reading. It can also mean they are lost. Read it next to key events: long engagement on a contact page with almost no form submissions usually points to a confusing page, not a successful one.
Key events (what GA4 used to call conversions)
Key events are the actions that matter to the business: a purchase, a quote request, a newsletter signup, a tap on a phone number. Google renamed GA4 conversions to key events in March 2024, and the word conversion now refers to the actions Google Ads counts. If your client sees key events in one screen and conversions in another, that is why.
Show key events by channel in every report, because that is how the reader learns which traffic is worth paying for. If you run ads, keep in mind that ad platforms undercount results they can’t see. Learning how to send conversion data back to Meta and Google Ads is worth doing before you judge a campaign on the platform’s own numbers.
Top pages and landing pages
GA4 calls page views simply views. A view is counted every time a page loads, so one visitor who reads three articles creates three views. That makes views a measure of content consumption, not of how many people visited.
The Landing page report is often more useful than a list of top pages, because it shows where sessions started and how each entry page performed on engagement and key events. A landing page with plenty of traffic and almost no key events is usually the first thing to fix.
Device breakdown
Split traffic and key events by device category: desktop, mobile and tablet. When mobile visitors engage far less than desktop visitors on the same page, the cause is usually layout or loading speed on small screens rather than the content itself. That is an action item a designer can pick up the same week.
Google Search Console data
GA4 tells you what happened after people arrived. Google Search Console tells you what happened before: which queries showed your pages in Google, how many impressions and clicks they earned, the click-through rate and the average position. Link Search Console to GA4 or pull it into Looker Studio, then list the queries that gained or lost the most clicks. Those queries are also good input for keyword research with Google Keyword Planner when you plan the next batch of content.
AI assistant traffic (ChatGPT, Gemini and other chatbots)
Since May 13, 2026, GA4 puts visits from recognized AI chatbots in their own AI Assistant channel instead of mixing them into Referral. Google’s examples include ChatGPT, Gemini, Claude, Copilot and DeepSeek, and GA4 sets the medium for these sessions to ai-assistant, which makes them easy to filter.
Two gaps are worth explaining to clients. Clicks from Google’s AI Overviews and AI Mode still count as Organic Search, so look for them in Search Console rather than in this channel. And some AI visits arrive without a referrer and land in Direct. Treat the AI Assistant number as a minimum, not a full count. If an assistant you care about still appears under Referral in your property, create a custom channel group with a rule above Referral that matches its domain.
Clicks are only part of the picture, because many people read an AI answer and never visit the site. If you want to know where chatbots mention your brand, answer engine optimization tools track those citations directly. And because the channel only appeared in mid-May 2026, add a note to the report so nobody reads its first appearance as a sudden jump in traffic.
Notes on what changed
Add a short change log to each report: a redesign, a new tracking setup, a price change, a campaign launch, a holiday. Six months later nobody remembers why traffic jumped in March, and without the note the report loses its value as a record.
GA4 terms that confuse clients
Standard Universal Analytics properties stopped processing data on July 1, 2023, but plenty of clients still think in its terms. A small translation table in your first GA4-based report saves a lot of back-and-forth email.
| Term the client remembers | What GA4 shows now | How to explain it |
|---|---|---|
| Bounce rate (single-page visits) | Bounce rate (sessions that were not engaged) | It looks lower because the definition changed, not because the site improved overnight |
| Goals or conversions | Key events | The important actions you chose to track, such as forms, calls and purchases |
| Pageviews | Views | One view each time a page loads |
| Time on page | Average engagement time | The time your page was in the foreground of the visitor’s screen |
| Custom reports | Explorations | Flexible analysis, limited by the property’s data retention setting |
| Referral traffic from chatbots | AI Assistant channel | Visits from ChatGPT, Gemini and similar tools, now reported on their own |
How to read a web analytics report in five minutes
If you are the one receiving the report, read it in this order. It gets you to the part that affects decisions before the detail pulls you in.
- Check the date range and what it is compared with. A month compared with the previous month can mislead a seasonal business, so look for the same month last year as well.
- Read the summary and the recommended actions first. If the report has neither, ask for them before the next one is sent.
- Look at key events before traffic. More visitors with the same number of leads or sales is not progress.
- Find which channel caused the biggest change, then which landing pages that traffic reached.
- Read the notes. A redesign, a tracking fix or a new channel in GA4 explains many sudden jumps and drops.
- Finish with one question: what will we do differently next month because of this report?
How to build a web analytics report people actually read
- Write down who reads the report and the one goal it tracks. The owner of a local service business wants leads. A content team wants returning readers. Different readers need different reports.
- Choose five to eight metrics tied to that goal. If a metric would not change a decision, leave it on the dashboard.
- Compare every number with the previous period and, once you have the history, the same period last year. Seasonal businesses need the yearly comparison most.
- Open with a summary of three sentences or fewer: what went up, what went down, and what you recommend.
- Use the simplest chart that does the job. Line charts work for trends over time and bar charts for comparing channels or pages. Save pie charts for showing a few parts of one whole.
- Explain anomalies in the notes section instead of leaving the reader to guess.
- End with two or three actions, each with an owner and a due date.
- Send the report on the same day every month. In GA4, administrators can open a report, click Share this report and choose Schedule email to send it automatically as a PDF or CSV file.
The same approach works for other channels. If you also report on Instagram, LinkedIn or TikTok, the components of a social media report follow the same logic: the goal first, a short set of metrics, then decisions.
Web analytics report template: a one-page monthly structure
This layout fits on one page or one screen, whether you build it in Looker Studio, a Google Doc or a slide. The example lines are illustrative, so replace them with your own numbers.
| Section | What it shows | Example line |
|---|---|---|
| Summary | The headline result and your recommendation | Leads rose 14% month over month, mostly from organic search. Next step: add a quote form to the two blog posts that send the most traffic. |
| Traffic by channel | Sessions per channel compared with last month | Organic Search: 2,140 sessions, up 9%. AI Assistant: 96 sessions, up from 61. |
| Engagement | Engagement rate and average engagement time | Engagement rate 61%, flat compared with last month |
| Key events | Leads, purchases or signups by channel | 38 quote requests, 24 of them from Organic Search |
| Top landing pages | Entry pages with engagement and key events | Pricing page gets heavy traffic but few quote requests |
| Devices | Engagement and key events by device | Mobile engagement rate is 20 points below desktop |
| Search Console | Queries gaining or losing clicks | WordPress maintenance plan: clicks doubled after the page update |
| Notes and actions | Changes, issues, owners and dates | Contact form fixed on the 12th. Designer to retest the mobile pricing page by the 30th. |
Tools for web analytics reporting
You can build a solid report without paying for anything. GA4 standard reports can be customized, downloaded as a PDF or CSV from the Share this report menu, and scheduled by email. Looker Studio, Google’s free reporting tool formerly called Data Studio, combines GA4, Search Console and Google Sheets in one report, and it is the usual next step when you want your own layout and branding.
On WordPress, plugins such as Site Kit by Google or MonsterInsights show the main GA4 numbers inside the admin area, which helps clients who never log in to Google Analytics. If tracking isn’t installed yet, start by learning how to connect Google Analytics to your WordPress site.
Paid reporting platforms earn their price when you manage many sites and need white-label client reports that pull analytics, ads and social data together automatically. For one or two sites they mostly add another login. And if you want to see how a competitor’s traffic compares with yours, GA4 can’t help. You need third-party website traffic analysis tools, which give estimates rather than exact numbers.
Common web analytics reporting mistakes
- Sending every metric GA4 offers. Long reports don’t get read, and the number that matters gets buried.
- Comparing GA4 bounce rate with Universal Analytics benchmarks. The name is the same, but the calculation isn’t.
- Leaving data retention at the default. Explorations keep event data for only two months unless someone changes the setting to 14 months under Admin, Data collection and modification, Data retention. Standard reports aren’t affected, but any deeper analysis is.
- Judging awareness campaigns on last-click results. Video and display ads rarely get credit in click-based reports, so they look weaker than they are. If they are part of your mix, read up on measuring upper-funnel impact before cutting the budget.
- Mistaking a tracking change for a performance change. A new consent banner, a duplicated tag, a longer engaged-session timer or the arrival of the AI Assistant channel can all move numbers overnight. Check the setup before celebrating or panicking.
- Reporting numbers without a recommendation. If the report doesn’t change what anyone does next month, it was a status update.
Web analytics report questions: metrics, GA4 exports and reporting frequency
A dashboard is a live screen for monitoring metrics as they change. A report covers a fixed period, compares it with an earlier one, and explains what changed and what to do about it. Dashboards suit people who check numbers often, while reports suit the people who make decisions.
Most reports need sessions by channel, engagement rate, key events such as leads or purchases, top landing pages, a device breakdown and Search Console queries. Keep it to five to eight metrics tied to the site’s goal, and drop anything that would not change a decision.
Yes. Open the report, click Share this report and choose Download File to save it as a PDF or CSV. Administrators can also choose Schedule email to send the report automatically as a PDF or CSV attachment.
GA4 changed the definition. Universal Analytics counted any single-page visit as a bounce. GA4 counts a bounce only when a session is not engaged, meaning it lasted 10 seconds or less with the default timer, had no key event and had fewer than two page views. The same traffic usually produces a lower bounce rate in GA4.
Open the Traffic acquisition report and look for the AI Assistant channel, which GA4 added in May 2026 for visits from recognized chatbots such as ChatGPT and Gemini. Some AI visits still land in Referral or Direct, so treat the number as a minimum and add a custom channel group for any assistant GA4 does not recognize.
Monthly works for most small business and client websites, because a month gives enough data to see a trend. Send weekly updates during launches or paid campaigns, and add a quarterly summary for owners who only want the big picture.
Not for one or two sites. GA4 and Looker Studio are free and cover most needs, including scheduled PDF emails. Paid reporting platforms save time for agencies that manage many clients and want branded reports combining analytics, ads and social data.
Start with one page
If your current report is a 20-page export, cut it to one page next month and put a written summary at the top. Then ask the reader what they did with it. Their answer tells you which metrics earn their place and which questions the report still leaves open.
GA4 menu names and metric definitions in this guide were checked against Google’s Analytics Help documentation when the article was updated. Google renames features often, so if a label in your account looks different, search the Analytics Help Center for the current name. Visualmodo has no commercial relationship with any of the tools mentioned in this guide.
