Bring a business question, not a tool question
The difference between a useful consultation and an expensive tutorial is what you walk in with. "Can you show me around GA4" gets you a tour. "Which of our channels actually produces customers, not just leads" gets you an answer.
Write down the three decisions you cannot currently make with confidence. Where should next month's budget go? Which campaigns produce customers who stay? Is the website getting better or worse at converting? Those are analytics questions wearing business clothes, and they are what a consultation should resolve.
The checks that expose a broken setup
Ask the consultant to run these in front of you. They take minutes and they tell you almost everything.
- Does the conversion count match reality? Compare last month's GA4 conversions with the enquiries your team actually received. If those two numbers disagree, everything downstream is fiction.
- Is internal traffic excluded? Your own team browsing the site should not be inflating your numbers.
- Do conversions have values attached? Counts alone cannot tell you what is worth doing more of.
- Are phone calls tracked? For most service businesses, this is the majority of the enquiries.
- Do the [recommended event names](https://support.google.com/analytics/answer/9267735) get used consistently? Inconsistent naming is how data becomes unusable a year later.
If three or more of those fail, you do not need a consultation. You need a rebuild of the tracking, and anyone who tells you otherwise is selling you an hour.
What you should leave with
- A written list of what is broken, ranked by what it costs you. Not everything needs fixing.
- A measurement plan. What the business cares about, and which events represent it.
- One report you will actually open. If it takes more than a minute to read, nobody will.
- A clear next action. With an owner and a date, or it will not happen.
What you should not leave with is a 40-page audit document. Data you do not act on is just expensive decoration, and a report that thorough is usually written to justify the fee rather than change the business.
When you do not need one
If your reports already match your bank account and you act on them, you are done, go and spend the money on traffic or conversion instead.
And if nobody in the business is going to change a decision based on better numbers, be honest about that before you book anything. Better data that nobody acts on is not an improvement. It is a more accurate way of doing the same thing.
But if you are spending real money on ads while quietly not trusting the reports, that is exactly what analytics and tracking work is for, and it is the cheapest problem on this list to fix.
The tooling question people get stuck on
A lot of consultation time gets burned on which tools to use. It is the wrong argument, and it is popular because it is easier than the right one.
The right argument is: what decisions do you need to make, and what is the minimum measurement that lets you make them confidently? Everything else is procurement.
- GA4 plus [Tag Manager](https://support.google.com/tagmanager/answer/6102821) covers most businesses. It is free, it is well documented, and it is what your ad platforms integrate with natively.
- Add call tracking if you are a service business. Without it, most of your enquiries are invisible, and no additional tool fixes that.
- Add the CRM connection when you have a sales process worth measuring. This is the upgrade that turns "leads" into "customers" in your reporting.
- Add a dashboard when, and only when, someone will actually read it.
That is nearly always sufficient. The businesses drowning in tools are rarely the ones making the best decisions; they are usually the ones who bought a solution to a measurement problem they had not defined yet.
And the point that governs all of it: data you do not act on is just expensive decoration. Adding another tool to a business that already ignores the data it has produces a more sophisticated version of ignoring it. Start with the decision. Work backwards to the smallest measurement that supports it.
What happens after the session is the actual test
A consultation is worth exactly what gets done afterwards. Most produce a document, a shared sense of enthusiasm, and no change whatsoever.
The failure is predictable: the fixes need developer time, nobody owns them, they go on a backlog, and three months later the tracking is exactly as broken as it was. Everyone remembers the session as useful and nothing was useful about it.
- Get the fix list ranked by cost of being wrong. Not everything matters. The double-counted conversion matters; the tidier event name can wait.
- Get an owner and a date against each item. Without both, nothing moves.
- Book the follow-up before you leave. Thirty minutes, four weeks later, to confirm the fixes actually landed.
- Verify, do not assume. After the changes, re-run the basic check: do recorded conversions now match the enquiries your team received? That is the only proof that counts. Google's recommended event structure tells you what good looks like, and Tag Manager is usually where the work happens.
And be honest about the last one. Data you do not act on is just expensive decoration. If your business is not going to change a decision based on better numbers, the most useful outcome of a consultation is realizing that, and spending the money on something else.
Frequently asked questions
What should I ask in a Google Analytics consultation?
Bring business questions, not tool questions. Write down the three decisions you cannot currently make with confidence (where the budget should go, which campaigns produce customers who stay, whether the site is converting better) and ask the consultant to answer those.
How do I know if my analytics setup is broken?
Compare last month's recorded conversions against the enquiries your team actually received. If the two disagree, the setup is wrong. Then check whether internal traffic is excluded, conversions carry values, phone calls are tracked, and event names are used consistently.
What should a good analytics consultation produce?
A ranked list of what is broken and what it costs you, a measurement plan tied to business outcomes, one report short enough that you will actually open it, and a clear next action with an owner and a date. Not a 40-page audit document.
Is an analytics consultation worth it for a small business?
It is worth it when you are spending real money on advertising and quietly do not trust the reports, because every budget decision depends on those numbers. It is not worth it if nobody will change what they do regardless of what the data says.