Five Website Metrics Worth Watching (And Four to Ignore)

Too Much Data, Not Enough Information

Open Google Analytics and you are handed several hundred metrics, almost all of which are irrelevant to whether your business is doing well. The natural response is to fixate on the one number that is easiest to understand — sessions — and treat it as a proxy for success.

It is a poor proxy, and it has become a worse one. As we covered in our piece on zero-click searches, a growing share of valuable search activity never produces a visit at all. Traffic can fall while the business grows. Traffic can rise while nothing improves.

Here is a shorter list.

The Five Worth Watching

1. Conversions, defined honestly

A conversion is something that moves someone closer to being a customer. For most Northern Ireland SMEs that is a form submission, a phone call, a quote request, a booking, or a sale.

The trick is defining it honestly. A newsletter signup is a conversion; it is not the same conversion as a £4,000 enquiry. Track them separately or you will optimise your website into a very efficient producer of things that do not pay wages.

Set up two or three goals, no more:

• Primary — the thing that makes you money. Enquiry, booking, purchase.
• Secondary — a genuine step toward it. Pricing page view, quote calculator use, phone number tap.
• Ignore everything else.

If you track one number on your website, track this one.

2. Conversion rate by landing page

Total conversion rate is mildly interesting. Conversion rate broken down by the page people arrived on is where the money is.

This is the report that tells you a service page converts at 6% while your homepage converts at 0.8%, which tells you to spend your effort and your ad budget accordingly. It tells you that a blog post pulling 400 visits a month converts nobody, and another pulling 40 produces two enquiries.

Small businesses consistently discover the same thing here: a couple of pages do nearly all the work. Find them, then improve them rather than building new ones. Our conversion rate optimisation guide covers what to change once you know where to look.

3. Search Console impressions and average position

Analytics tells you about people who arrived. Search Console tells you about people who nearly did.

Two numbers matter:

Impressions — how often you appeared in results. If impressions are climbing, your visibility is growing even if clicks are not. If impressions are falling, you have a real problem regardless of what your traffic looks like this week.

Queries with high impressions and low clicks — these are your cheapest wins. You are already ranking; people are choosing not to click. Almost always this is a title tag and meta description problem, and rewriting them takes ten minutes.

4. Google Business Profile actions

For any business with a local customer base, this is where a large chunk of your value shows up, and it is invisible in your website analytics.

Calls, direction requests, website clicks, messages. Someone who finds you in the local pack, sees your rating and rings you has converted without ever loading your site. Our Google Business Profile guide explains how to grow these; the point here is simply to count them.

If your website traffic is flat but your profile calls doubled, you had a good quarter.

5. Enquiry source, self-reported

The least sophisticated metric on this list and frequently the most accurate. Add one question to your enquiry form or ask it on the phone: how did you hear about us?

Attribution modelling struggles with the reality of local business: someone sees you on Instagram, forgets, searches your name three weeks later, and lands via branded search. Analytics credits organic search. The customer would tell you Instagram. Both are true; only one is actionable.

The Four to Ignore

Bounce rate

Bounce rate has always been a bad metric for small business sites and it is now largely meaningless in GA4, where engagement is measured differently. A visitor who lands on your contact page, reads your phone number, rings you and leaves is a bounce and a customer.

Look at conversions instead. If people are converting, you do not care what the bounce rate is.

Total pageviews

Pageviews measure how many pages someone had to look at. That is not obviously good. A site where people find what they need on the first page will show fewer pageviews than one where they have to hunt.

Time on page

Time on page is heavily distorted by how it is measured — a visitor who opens your site in a background tab and goes for lunch records forty minutes of rapt attention. Even when accurate, longer is not automatically better. Ten seconds on a page listing your opening hours is a success.

The exception: long-form content where you genuinely want people to read. Even then, use it as a rough signal, not a target.

Keyword rankings for their own sake

Ranking reports are seductive because they go up and to the right when you want them to. But rankings vary by device, location and search history, and a number one position for a term nobody searches, or nobody buys from, is worth nothing.

Track rankings for a handful of terms you have decided actually matter commercially. Ignore the rest of the report. Our Belfast local SEO guide covers how to choose them.

A Practical Setup

For a small business, the whole thing should take about twenty minutes a month.

Once, at the start:

1. Verify your site in Google Search Console
2. Set up GA4 with two or three conversion events
3. Add "how did you hear about us?" to your enquiry form
4. Note your starting numbers somewhere you will find again

Monthly, twenty minutes:

1. Conversions this month vs last month vs the same month last year
2. Top five landing pages by conversions
3. Search Console: impressions trend, and any query with impressions but no clicks
4. Google Business Profile: calls, directions, website clicks
5. Read the self-reported sources from actual enquiries

Quarterly, an hour:

Look at the trend rather than the month. Months are noisy — a bank holiday, a bad week of weather, one large customer. Quarters tell you something.

Comparing Year on Year, Not Month on Month

Seasonality will mislead you constantly if you compare consecutive months. A Belfast retailer comparing December to November concludes they are a genius. Comparing January to December concludes they are finished. Neither is true.

Compare like for like: this September against last September. It is the only comparison that controls for the fact that your business has a shape to its year.

What Good Looks Like

There is no universal benchmark, and anyone quoting one without knowing your sector is guessing. But as rough orientation for a small local business site:

• A service business converting 2–5% of visitors into enquiries is doing well
• An e-commerce site converting 1–3% is normal; above 3% is good
• Impressions growing quarter on quarter matters more than any single month's traffic
• If more than half your enquiries name a source your analytics never credits, your marketing is working in ways you are not measuring

The Underlying Point

The purpose of measurement is to decide what to do next. If a number does not change a decision, it does not need watching. Most website analytics is watched, screenshotted, put in a report and acted on by nobody.

Pick five numbers, look at them monthly, and spend the time you save on the website itself.

For help setting up measurement that tells you something useful — or a free website audit that starts from your actual data — get in touch with Amigo Studios. We work with Belfast businesses on websites built to convert, not just to be visited.