The Five Numbers to Check Every Monday Morning

Two hundred numbers, and none of them tell you what to do

Open any analytics dashboard and it will show you roughly two hundred numbers. None of them tell you what to change on Monday. So most owners either look at everything for an hour and change nothing, or stop looking altogether.

Five numbers, ten minutes, every Monday morning with your tea. That is the whole practice.

The five numbers

1. Visitors — did anybody arrive?

How many different people reached your site last week. Find it free in Google Analytics 4 under Reports → Acquisition → Overview → Users, or in your hosting statistics if analytics is not set up yet.

What good looks like: for a new local site, anything above twenty in month one — and rising week on week matters far more than any particular figure. By month six, a hundred or more a week is a healthy base.

2. Top source — where did they come from?

Search, social, direct or referral. GA4 → Acquisition → Traffic acquisition, then look at the largest channel. Check your Google Business Profile insights separately, because map visitors often never appear in your website analytics at all.

What good looks like: search and direct both present. If a hundred percent comes from one place, you are standing on one leg.

3. Top page — what did they look at?

GA4 → Reports → Engagement → Pages and screens, sorted by views.

What good looks like: your home page or your main service page. If a blog post is winning, that is genuinely good — but check that it links clearly to the page where somebody can actually hire you.

4. Actions — did anybody do the thing?

Form sent, WhatsApp click, call button pressed. Count form submissions in your form plugin’s entries. For the first few months you can honestly count WhatsApp and call clicks by hand in a notebook.

What good looks like: for a local service business, five to fifteen actions a week on modest traffic. Zero for two weeks running with traffic above fifty means something is broken.

5. Action rate — what proportion acted?

Actions divided by visitors, times one hundred.

What good looks like: two to five percent for a home page. A focused landing page should do better — five to ten percent, sometimes more when the traffic already knows you.

The fix-first rule

This is the part that saves you from wasted months. Find your weakest number in funnel order — arrivals, then engagement, then action — and change exactly one thing.

• No page work while traffic is under twenty. Nobody knows the site exists. That is almost never a design problem, and redesigning an empty shop changes nothing.
• No traffic work while action rate is under one percent. More visitors simply leak faster. Fix the page first.

Some worked examples of the one fix:

• Visitors under twenty and flat → complete your Google Business Profile fully and add ten real photographs. That is the whole job this week.
• Visitors rising, actions zero → send yourself a test enquiry from your phone. If it arrives, move the button above the fold and add a WhatsApp route.
• Action rate under one percent → put one specific testimonial and a “from” price directly above the main button.
• All traffic from one source → publish one question-and-answer post this week, based on a real customer question.
• Top page is not the page you expected → add your standard call-to-action to the bottom of that page today. You have just found the room people were already standing in.

The Monday ritual, minute by minute

• 0–2: open GA4 and set the range to last week — Monday to Sunday, not “last seven days”, so weeks stay comparable.
• 2–5: read off visitors, top source and top page. Do not click into anything else, however tempting.
• 5–7: count actions.
• 7–9: name the weakest number in funnel order, in one word.
• 9–10: decide the single change you will make this week, and when you will make it. Write it down as an action with a place and a day — “move the proof strip above the fold on the home page, Wednesday evening”. Then close the sheet.

Five traps that make people misread their own numbers

• Counting your own visits. You visit your site more than anybody. In week one that can be half the traffic. Filter your own traffic in GA4, or at minimum stop refreshing the page to admire it.
• Judging in the first week. Weekly numbers wobble for reasons that have nothing to do with you — a festival, exam week, heavy rain, one WhatsApp forward. Never draw a conclusion from fewer than three rows.
• Chasing traffic when the leak is the page. More visitors feels like progress and is easy to buy. It does not help a page that does not persuade.
• Changing five things at once. The numbers move, everybody celebrates, and nobody knows what caused it. One change per week, written down before you make it.
• Treating a zero as failure instead of a fault. Two weeks of zero actions with decent traffic is almost always a broken form or an untappable phone number — not a weak offer. Test before you redesign.

Why this works

One measured change a week is fifty-two a year, and you will know which ones worked. That is a different business from the one that redesigns the whole site every eighteen months and still cannot say whether it helped.