Analytics platforms report several hundred metrics. Most businesses look at three, act on none, and would be better served by five specific numbers and a habit of checking them.
Key takeaways
- A metric is only worth tracking if a plausible reading of it would change what you do next week.
- Sessions and pageviews are volume, not performance. They rise and fall for reasons you did not cause.
- Segment by acquisition channel before drawing any conclusion, site-wide averages hide the entire story.
- Bounce rate and time on page are the two most confidently misread numbers in the industry.
On this page
The test a metric has to pass
Before tracking anything, write down what you would do if the number doubled and what you would do if it halved. If both answers are “nothing”, it is not a metric, it is decoration.
This one filter removes most of a typical dashboard. It also exposes the uncomfortable case where a number would change your decision but you have no way to influence it: interesting, but not actionable.
The five that earn their place
Almost every business can run on conversion rate, cost per acquisition, channel mix, landing page performance and returning visitor share. Everything else is diagnostic detail you reach for once one of these moves.
| Metric | Tells you | Decision it drives |
|---|---|---|
| Conversion rate by channel | Which traffic is worth having | Where to spend more or stop |
| Cost per acquisition | What a customer costs to win | Whether the channel is viable |
| Channel mix over time | How dependent you are | Where the concentration risk is |
| Landing page performance | Which entry points work | What to fix or replicate |
| Returning visitor share | Whether anyone comes back | Retention and content investment |
Note that four of the five are ratios or comparisons. Absolute counts are almost never useful on their own, because they conflate how much traffic arrived with how well the site handled it.
The metrics that mislead
Bounce rate and time on page are not quality measures, and treating them as such produces confidently wrong decisions. Both are artefacts of how measurement works rather than descriptions of user satisfaction.
Bounce rate
A high bounce rate on a page that answers the question completely is a success. Somebody searched for your opening hours, found them, and left satisfied. The same number on a category page is a failure. The metric cannot tell the two apart; only you can, by knowing what the page is for.
Time on page
Time is measured between events. If someone reads for six minutes and then closes the tab, there is no closing event, so the visit frequently records as zero seconds. Long times can equally mean the visitor was confused, or made a cup of tea. It is a weak signal in both directions.
Site-wide averages
An average conversion rate across all channels is close to meaningless. Branded search converts several times better than cold display traffic, so the average mostly tracks how your traffic mix shifted, not how your site performed. Always segment before concluding.
Getting a setup you can trust
Most analytics problems are data collection problems, and they are invisible unless you look for them. A dashboard built on broken tracking is worse than no dashboard, because it produces confident decisions from noise.
- Define your conversions before installing anything. Two or three real ones beat fifteen vanity events.
- Exclude internal traffic. Your own team browsing the site can distort a small business’s numbers substantially.
- Filter known bot and spam referrals, then re-check quarterly.
- Check that campaign tags survive every redirect. One redirect that strips parameters can silently reattribute an entire channel to direct traffic.
- Verify events fire once, not twice. Duplicated conversion events are common and quietly halve your apparent cost per acquisition.
- Write down what each event means. In six months nobody will remember what
form_submit_2was.
How often to look
Weekly for spend, monthly for performance, quarterly for strategy. Checking daily produces reactions to noise, and small sites are almost entirely noise at a daily resolution.
A site with forty conversions a month sees swings of thirty per cent from ordinary randomness. Reacting to those is not analysis, it is superstition with a dashboard. Give any change enough time to separate itself from the variance before drawing a conclusion.
The purpose of a report is to shorten the argument about what to do next. If it does not do that, it is a screensaver.
Serhii Yelbaiev, Well Web Marketing
Frequently asked questions
Usually not at first. One well-configured analytics platform plus Search Console covers most small businesses. Add a session recorder only when you have a specific question about behaviour that numbers cannot answer.
They always will. Ad platforms count conversions against the click that led to them and use their own attribution window; analytics assigns credit by its own model. Compare each source against itself over time rather than against the other.
The only useful benchmark is your own previous figure for the same channel. Published industry averages combine businesses with nothing in common and mostly measure differences in traffic quality.
Enough conversions that ordinary variance cannot explain the change: usually a few hundred, which for many small sites means several months rather than several weeks.
Sources and method
Reflects analytics implementation and reporting work for small and medium clients. Bounce rate and engagement time behaviour describe how event-based analytics platforms record sessions. Last reviewed February 2026.