A spike in website visits looks promising—until you see that most people landed on an old post about toddler shoes, while you’re trying to help families find your photo book ordering page. The visits are real. They just don’t answer your question. A few numbers read together will tell you more than a dashboard full of big ones.
First, decide what each important page is there to do. A parenting post might answer a practical question. A photo book page might explain the format and help someone request details. The contact page should let people get in touch. Once you know the job, you can choose numbers that relate to it.
Set up a measurement baseline first
Use a privacy-conscious analytics setup that meets your local requirements. Check that it records your main pages and the actions you care about. Exclude your own visits if you can; a parent editing a post between bedtime and the second bedtime can generate a surprising amount of traffic. If visitors’ consent choices limit tracking, your reports will show only what your setup can measure.
Choose a consistent period, such as the last 28 days, and compare it with the 28 days before that. Jot down anything that might explain a change: a new post, an email to subscribers, a school holiday, or a broken order form. Otherwise, it’s easy to give a rise or fall the wrong explanation.
The core metrics and what they can tell you
1. Visits and people
Visits, often called sessions, are browsing periods. Users are an estimate of distinct visitors, calculated according to the tool you use. One person may make several visits or show up as more than one user across devices. Neither figure is a headcount.
Watch how the figures change over time, then look at the pages people visit. A quieter month in which more readers reach a useful guide or product page may do more for your site than a viral week of unrelated traffic. On a small site, use several weeks of data rather than treating one busy afternoon as a trend.
2. Traffic sources
Source reports show how visitors arrived: through search, a newsletter, another website, social media, or a direct visit. These labels aren’t perfect explanations. “Direct,” for instance, doesn’t prove someone typed your address; visits with missing referral information can end up there too.
Look at sources alongside page goals. If search brings people to a family blog post, check whether it answers the question they came with. If a newsletter sends only a handful of people to your photo book page but several make inquiries, those visits may be especially useful. Add campaign tags to links you control, such as newsletter buttons, to make later reports easier to read.
3. Landing pages
A landing page is the first page someone sees during a visit. It could be a years-old post, not your homepage. Check your top landing pages each month. On a family blog, an old article can keep helping new parents long after you wrote it.
Check that each page delivers what its title promises. If a guide to preserving children’s artwork draws visitors, make sure the instructions still work and the photos load. You don’t need to steer every reader toward a sale. Sometimes a clear answer is the whole point of the page.
4. Engagement, with care
Analytics tools define engaged visits differently. One may count time on the page; another may also count a second page view or an event. Read your tool’s definition before comparing reports. Time on page has limits too: someone might leave a helpful page open while making lunch, while another reader gets the answer in 30 seconds.
Use engagement figures to find pages worth checking, not to grade readers. A sudden drop on one page might mean a slow-loading image, a confusing layout, or a change in tracking. Open the page on your phone before deciding the writing is at fault. If the layout is hard to follow, the blog’s guide to making posts easier to scan is a place to start.
5. Actions that match the page’s purpose
A meaningful action might be a completed contact form, a newsletter signup, a download, or the confirmation page after an order. Where possible, track the completed action. A click on “Send” doesn’t necessarily mean an inquiry arrived: the form could fail, or a required field could prevent submission.
For a product page, you might record both visits and completed inquiries. For a practical parenting guide, a visit to related instructions might be useful to know about. But not every tap needs to become a “conversion.” Pick two or three actions you’d genuinely notice if they stopped happening.
Read the numbers together, not separately
A monthly check can fit in a small table. Keep your definitions and date range the same each time so the comparisons mean something.
| Record | Question to ask |
|---|---|
| Visits and top sources | Where did measured visitors come from? |
| Top landing pages | Which pages welcomed them first? |
| Engagement on key pages | Is anything unusual enough to inspect? |
| Completed actions | Could visitors finish the task the page offers? |
Then look at one relationship between the figures. Say your photo book page had 100 measured visits last month and 50 this month, while completed inquiries held steady at four. That doesn’t prove the page improved: traffic may have come from different sources, and the counts are small. It does suggest checking where the visits changed before rewriting the page. If inquiries drop to zero, test the form yourself before deciding demand has fallen.
Keep privacy and small numbers in view
Analytics show patterns, not individual families. Don’t collect names, children’s details, or private messages as analytics events. Limit access to reports and collect only the data you need to answer a practical question. If you publish family stories, a visitor’s path through them tells you nothing reliable about that family’s circumstances.
Percentages swing sharply on a small site. One inquiry from ten visits is 10%; two inquiries double that rate. Keep the raw counts beside the percentages, and don’t call a single week a trend. Record tracking changes as well: a new consent banner can change your figures even if visitors behave exactly as before.
For your first monthly review, write down the top three landing pages, where their measured visits came from, and how many contact forms were completed. Then submit a test form from your phone. Check that the message arrives and that the completed action appears in your report.