Your website is live. Which basic numbers matter?
Four measurement groups give a new site a useful baseline: visits, sources, recorded engagement and enquiry actions. Here is what each can and cannot prove.
Once a website goes live, it starts producing numbers — and the numbers can either inform you or distract you. For a small-business site, four of them do most of the useful work. Everything else can wait.
The four numbers worth watching
1. Visits. How many people came, week by week. The absolute number matters less than the direction and the pattern: a site serving a local trade might see modest, steady traffic and be doing its job perfectly. What you’re learning in the first months is your baseline — what normal looks like — so that later changes stand out.
2. Where visitors come from. Search, tagged campaigns, social links, direct visits and links from other sites. Attribution is imperfect — privacy settings, apps and untagged sharing can hide the original source — but consistent campaign links help compare which activities deliver visits and actions.
3. Which engagement events are actually recorded. A page view only proves the page loaded. If scroll depth, engaged time or another event has been deliberately configured, it can provide additional clues; it still does not prove understanding or purchase intent. State exactly which event you are looking at instead of calling every visit a “read”.
4. Enquiry actions and qualified outcomes. Record form starts, successful submissions and agreed contact clicks separately. Then let staff record which submissions became useful conversations or customers. A click cannot prove that a message was sent, and a submission cannot prove that it was a qualified lead.
What not to obsess over in month one
- Likes and follower counts. They live on social platforms and say little about whether the website converts visitors into enquiries.
- Search rankings. New sites take months to settle in search. Checking your position daily in month one is checking a pot that hasn’t started boiling — and nobody can honestly promise you a ranking anyway.
- Daily traffic wobbles. Small sites have small numbers, and small numbers swing. Twelve visits Tuesday and thirty Wednesday means nothing. Compare weeks, not days.
- Averages that hide the point. “Average time on page” across the whole site blends your best and worst pages into mush. Look page by page or not at all.
When a number justifies action
A number earns a reaction when it is sustained, explains itself, and suggests a specific change. Some worked examples:
- Enquiries started, few completed — for several weeks running. Action: shorten the form, or add the chat channel your customers actually use.
- One service page repeatedly receives more recorded engagement and qualified actions. Action: investigate the source and visitor intent before deciding whether to give that service more prominence.
- Growing search traffic lands on one article or page. Action: make sure that page has a clear path to contacting you, because strangers are entering there, not on your homepage.
- Visits are sustained while enquiry actions remain near zero. Action: review traffic relevance, the offer, information, trust signals, technical errors and contact friction. The metric identifies a gap, not its cause.
A sane routine
Choose a review rhythm that produces a meaningful sample for your business; for many small sites, monthly is more useful than daily. Write down visits, attributable sources, configured engagement events, enquiry actions and qualified outcomes next to the previous period. Small samples need cautious interpretation. If the pattern is stable and supports the business goal, the correct conclusion may be that no immediate change is needed.
Measurement must also respect applicable privacy, consent and platform rules. Collect only what serves a defined purpose, and do not send customer form content into analytics events.
This article describes general knowledge about reading website numbers — it applies whether or not you build with us.