Did Your PR Actually Bring in Customers? A Simple Tracking Setup for Small Businesses

You land a feature in a trade publication, share it on LinkedIn, and a fortnight later someone asks, “Did that actually do anything?” You check your website numbers, see a small bump somewhere in the middle of the month, and realise you have no honest way to say whether the article caused it.

Most small businesses stop measuring PR at the coverage itself: the clipping, the screenshot, the number of shares. That feels like a result, but it doesn’t tell you whether anyone visited your site, rang you or filled in a form because of it. This post sets out a tracking setup I would put in place for a small business in an afternoon, using free tools, and where AI helps and where it can’t.

I have written before about why agencies struggle to prove PR value. This is the practical, do-it-yourself companion: what to set up, in what order, and what to expect.

What does it mean to track PR results?

Tracking PR results means connecting a piece of coverage to something your business can count: visits to your website, enquiries, sign-ups or sales. You can’t measure everything, because many readers will see the article and search for you later without clicking anything. The aim is a reasonable, repeatable estimate, not perfect attribution.

Think of it in two layers. The first is what you can see directly: clicks from a link in an article or post to your site. The second is what you infer: a rise in searches for your business name, or customers who say “I read about you in…” when they call.

You need both. The direct layer is easy to count but undercounts. The inferred layer is looser but catches the readers who never click.

How much does it cost to set up?

For a basic setup, the software cost can be zero. Google Analytics 4 is Google’s analytics product, and standard properties are free to use.

The real cost is your time. Plan for about two to three hours for the first setup and roughly 30 minutes a month after that.

The setup, step by step

Here is the workflow, with rough times. Do it before your next piece of coverage goes live, not after.

Step one: Install Google Analytics 4 (45 to 60 minutes)

If your website already has GA4, skip to step two. If it doesn’t, create a property at analytics.google.com and add the tracking tag to your site. Most website platforms offer a paste-in field or a plugin for this. If you’re not comfortable editing your site, this is a good job to hand to your web developer, as it’s a 20-minute task for them.

While you’re in there, check the data retention setting. Google’s help page says standard properties can keep event data for 2 or 14 months (Google Analytics help, checked 5 November 2026). Choose 14 months so that you can compare this year’s coverage against next year’s.

Step two: Decide what counts as a result (15 minutes)

Pick one or two actions on your site that matter: a contact form submission, a booking, a phone number tap, an email sign-up. In GA4 these are called key events. Ask your developer or follow Google’s help pages to mark them.

Without this, you’ll only know that people visited. With it, you can see whether the people who arrived from a specific article did anything useful.

Step three: Tag the links you control (20 minutes)

A tagged link is an ordinary link with a few extra words on the end that tell GA4 where the visitor came from. Google’s help page lists the standard parameters, including utm_source, utm_medium and utm_campaign, and states that parameter values are case sensitive, so “utm_source=google” is different from “utm_source=Google” (Google Analytics help, checked 5 November 2026).

That case-sensitivity detail is the most common way small businesses end up with messy data. If one person writes “LinkedIn” and another writes “linkedin”, you’ll see two separate rows. Decide your naming now and write it down:

1. utm_source: where the link lives (linkedin, newsletter, facebook, podcast-show-notes).
2. utm_medium: the type of channel (social, email, referral).
3. utm_campaign: the PR campaign, in lower case with hyphens, such as spring-launch-2026.

Use these on links you place yourself: your LinkedIn posts about the coverage, your newsletter, your email signature and your social bios.

You can’t tag a link inside someone else’s article. A journalist will link to your site however they like, usually without any tags, and you shouldn’t ask them to add tracking. For those, you rely on the referral data in the next step.

Step four: Check where visitors come from (10 minutes, then monthly)

In GA4, the Acquisition section includes a Traffic acquisition report, where Google says the tagged parameter values are visible, including session source, medium and campaign (Google Analytics help, checked 5 November 2026). Untagged clicks from an article will normally appear as referral traffic from the publication’s domain.

Look at the dates. If an article went live on a Tuesday, check the visits from that domain on Tuesday and Wednesday. Write the figure down, even if it’s small. A trade publication sending 40 well-matched visitors can be worth more than a large site sending 400 who bounce.

Step five: Build a one-page monthly report (30 to 40 minutes)

You can use Looker Studio to build a simple dashboard that reads from GA4, or a spreadsheet if you prefer. Keep it to a handful of lines: coverage published, visits from that coverage, key events from those visits, and a note on anything you can’t explain.

An AI assistant can help summarise the numbers once you’ve pasted them in. Try something like this:

I run a small business in Australia. Below are my website numbers for last month and a list of media coverage published. Write a plain-English summary of 150 words: what appeared to change after each piece of coverage, what I can’t say with confidence, and one thing to try next month. Do not claim any piece of coverage caused a change unless the dates and numbers clearly support it. Numbers: [paste here]. Coverage list: [paste here].

Read the output with a sceptical eye. The “do not claim” instruction matters, because AI tools like to tell tidy stories, and a tidy story about causation is exactly what you’re trying to avoid.

A worked example

Take a Brisbane physiotherapy clinic, which I’ll use as an illustration (it is a made-up example, not a client). The owner is quoted in a regional newspaper’s online story about desk-job injuries on a Tuesday. She shares the story on LinkedIn with a tagged link and mentions it in her Thursday newsletter, also tagged.

At the end of the month she looks at the report. The newspaper sent 31 visits, her tagged LinkedIn post sent 18, and the newsletter link sent 44. Seven of those people booked a first appointment through the website form. In the same fortnight, searches for the clinic’s name in her Google Business Profile insights rose modestly.

What can she say honestly? That tagged and referral traffic from the coverage produced a measurable number of bookings, and that the newspaper story probably also lifted people who searched for her afterwards. What she can’t say is how many of those bookings would have happened anyway, because she doesn’t have a control group.

That honest, slightly hedged statement is far more useful in a conversation with a business partner or accountant than a screenshot and a guess.

Which tools should a small business use?

For most small businesses, the answer is GA4 plus a spreadsheet, and nothing else for the first three months. Add a shortener or dashboard only when you find the manual steps annoying.

If you want to compare options, here is how I would think about each:

• GA4 alone. Free, shows referral and tagged traffic, and has a learning curve. It’s a sensible starting point.
• A shortener such as Bitly. Useful if you share links in print or on a podcast, where a long tagged URL is awkward. On the Free plan you’re limited to 5 short links a month, so check the current limits before relying on it.
• Media monitoring tools. These find coverage rather than measure visits. My guide to media monitoring on a real budget covers that side.

What not to do

A few traps are worth naming, because I see them regularly.

• Don’t change your tag names halfway through. Switching from “newsletter” to “email” in month three splits your history in two.
• Don’t count every spike as PR. A school holiday, a competitor closing or a seasonal pattern can move your numbers as much as an article can.
• Don’t track people you shouldn’t. If you add extra tracking tools or collect personal information, your privacy policy needs to reflect that. Australian privacy rules can apply to small businesses in some circumstances, and I’m not a lawyer, so check with a professional if you’re unsure.
• Don’t build a dashboard before you have data. An empty dashboard looks impressive and tells you nothing.
• Don’t ask a journalist to add tracking links. It looks odd, and it can cost you goodwill.

What AI can and can’t do here

AI is good at the tidying-up jobs in this process. It can draft your naming convention, turn a messy export into a readable summary, and suggest questions to ask next month.

It can’t see inside your analytics unless you paste the figures in, and it can’t tell you why people visited. It will also happily produce a confident explanation for a bump that was really a public holiday. Treat any causal claim from an AI tool as a hypothesis to check against dates.

It also can’t replace the human signal: asking new customers how they heard about you. I add a “How did you hear about us?” field to enquiry forms for every client, and the free-text answers often mention a podcast, a paper or a friend that no analytics tool picked up.

If you want help getting coverage worth measuring in the first place, the PR Team trial is a way to try the AI-Stories PR Team for free, and the AI-Stories page has the current plans. And if your site or analytics are already tangled, it may be worth paying a developer for an hour rather than losing a weekend to it.

Where should you start this week?

Start with the smallest version. Check that GA4 is installed, mark one key event, and agree your three naming rules. Then tag the next link you share about any coverage.

If you do only that, in three months you’ll have a record most small businesses never build. For tracking the coverage itself, the press page guide is a useful companion, because a good press page gives journalists somewhere to land that you can also count.