Google Analytics 4 for small business: what to set up, what to watch and what to ignore
Basic settings, key events, five useful reports and what to ignore in GA4, explained for people who are not analysts.
Which business KPIs belong in a monthly marketing report, which vanity metrics to leave out and how to turn numbers into decisions.
A useful monthly marketing report has a one-page decision summary and a handful of business KPIs: revenue or qualified enquiries, cost of a new customer, conversion rate and margin after ad spend. Each figure is set against last month and against target, explained briefly, and followed by two or three actions for the month ahead. Impressions, likes and followers appear only as context.
A report does not exist to prove that work was done. It exists to answer three questions: what happened, why it happened and what we do differently next month. If you cannot decide anything after reading it, it was an archive of numbers, not a report.
Reports often suffer from two problems: twenty-page documents with a chart for every platform and no conclusion, and reports that show only what went up. An honest report also shows what fell, what could not be measured and what remains unclear.
As a principle, the report starts from the goals of the business, not from what the platforms display. If the target is more qualified quote requests, the spine of the report is the number and cost of those requests, not the likes on the last few posts. Everything else serves as explanation.
“Vanity” describes metrics that look good but say nothing about money: easy to raise, hard to tie to results. They are not useless, because they can explain a change, but they should not sit at the centre of the report.
| Area | Business KPIs (at the centre) | Context metrics (in the appendix) |
|---|---|---|
| Sales | Revenue, number of orders, average order value, margin | Product views, add-to-basket events |
| Lead generation | Qualified enquiries, cost per qualified enquiry, enquiry-to-customer conversion rate | Total form submissions, clicks |
| Paid ads | Cost of a new customer, ROAS, attributed revenue | Impressions, CTR, cost per thousand impressions (CPM) |
| Website | Conversion rate, conversions by source | Sessions, page views, bounce rate |
| Social media | Visits to the site, relevant enquiries or messages | Likes, followers, reach |
| Revenue or enquiries from email, unsubscribes, click rate | Open rate (indicative) |
A simple rule: if a number can rise without revenue rising, it is a context metric. Follower counts can climb after a giveaway without a single extra customer. Email open rate is more fragile still, because privacy features in mail apps can inflate it, so treat it as a hint, not as proof.
There is no universal list. Choose between five and eight indicators, each of which you can explain in one sentence. More than that usually means you have not chosen.
A hypothetical example: a shop spends 800 EUR on ads in a month and gets 40 new customers. The cost of a new customer is 20 EUR. If the average order value is 45 EUR and gross margin is 40%, one order leaves 18 EUR. On the first order the shop loses 2 EUR, so the report must also show how often customers return. Otherwise the 20 EUR figure does not tell you whether it is good or bad. For calculations like this, use the LTV and CAC calculator.
A short report gets read. A long one gets filed. The structure below fits into four to six pages or slides, plus an appendix.
The numbers must come from the right source: sales from the shop or the CRM, not from the ad platform; traffic and conversions from GA4; costs from the ad accounts. If measurement is not stable, fix that first, with the help of our guide to GA4 for small business owners.
A monthly report does not replace frequent checks. Each rhythm has a different purpose.
Is spend within budget? Do the forms still send emails? Have any ads been disapproved or any tracking errors appeared? This is not analysis, it is hygiene. It takes fifteen minutes and prevents weeks of wasted money.
This is where the report in this article belongs: what we reallocate, what we stop, what we test. A month holds enough data to show trends at medium volumes, but it can still mislead at low volumes or in seasonal businesses.
Every three months you look at trends, acquisition cost over time, customer value and whether the goals still make sense. A short-term plan gives you the rhythm; see the 90-day marketing plan.
For campaigns with long sales cycles, one month’s figures do not tell the whole story: money spent today may bring sales in two months. In that case the report also shows an intermediate indicator, such as meetings obtained.
To see where you stand today on measurement, content, ads and website, you can run a quick digital maturity scan: it highlights the weak points worth tracking in the report.
CAC (customer acquisition cost) is marketing spend divided by the number of new customers won in the same period. CPL (cost per lead) is spend divided by the number of enquiries. ROAS (return on ad spend) is revenue attributed to ads divided by ad spend; it ignores margin, so a “good” ROAS can hide a loss. Conversion rate is the percentage of sessions or visitors that completed a desired action; see also the conversion rate calculator. Average order value is revenue divided by the number of orders. LTV (lifetime value) is the revenue or margin a customer brings over the whole relationship, estimated with caution.
Put the definitions in the appendix and use them the same way every month. A KPI whose definition keeps changing (one month an “enquiry” means a form, the next a form plus a call) makes comparisons worthless.
If the report contains data about individuals (names, emails, phone numbers), keep it in aggregate form or restrict access. This text is informative and is not legal advice; for the processing of personal data, ask a lawyer or your data protection officer.
With the clients we work with, the report starts from two questions asked at the beginning: what is a good customer, and how much can you pay for a new one. The KPIs come out of those answers, and the remaining figures go in the appendix. We say openly what we cannot measure and what we only estimate roughly.
On measurement we work within our conversion rate optimisation service, where we check tracking first and the numbers second. If you already work with an agency and want a benchmark for what you should be receiving, read how to choose a marketing agency.
Usually between five and eight. Each should be explainable in one sentence and tied to a business goal. If you have more, some are probably context metrics and belong in the appendix. More figures do not mean more clarity, and a crowded panel makes the important changes harder to spot.
There is no single answer for everyone. For most businesses, the most useful is the cost of a new customer set against the margin that customer leaves. For a shop it may be margin after ads; for a service business, the cost of a qualified enquiry. Choose the indicator that links spend to profit.
The person who decides on budget and the person who carries out the actions. If it goes only to someone who cannot decide, the report gets lost. Email the summary and the actions, and present the full report in a short meeting, so that there are questions and decisions, not just an acknowledgement of receipt.
Say clearly what is missing and why, instead of estimating in silence. Mark approximate figures and note how you will fix the gap, for example by setting up measurement or linking your CRM. An honest report with highlighted gaps is more useful than one that looks complete but rests on assumptions.
It depends on how many sources you have and how automatically the data is collected. With a dashboard connected to the sources, collection can take little time, while the valuable part, interpretation and decisions, needs attention. If preparing it takes days, simplify the report or automate collection.
Yes, especially if you have few channels. Start from a simple template with your chosen KPIs and fill it in on the same date every month. An agency or consultant helps when measurement is complicated or you want an outside view. Even then, keep access to your own accounts and data.
Laborator
Lab tools that help you put this guide into practice.
Blog
Basic settings, key events, five useful reports and what to ignore in GA4, explained for people who are not analysts.
Three phases, clear deliverables and a short list of KPIs for each month, so you know exactly what to do in your first 90 days.
15 questions before the contract, warning signs, account ownership and reporting: how to pick an agency without paying on trust alone.
Tell us where you are now and what you want to achieve — you will get a concrete proposal with services, timeline and a transparent budget.