The figures commonly quoted for websites generally sit somewhere between two and five percent, and that range is assembled from ecommerce sites where a conversion is a purchase, lead generation sites where it is a form submission, and everything in between. A local service business receiving highly qualified traffic from a search for its exact service might reasonably convert at fifteen percent. A business receiving broad informational traffic might convert at half a percent and be performing well.
What matters is your own rate over time, and the same rate segmented by source. That second comparison is where the useful information sits. Traffic from a search for your service will convert at a completely different rate from social traffic, which is not a criticism of social traffic. It means the two are doing different jobs, and averaging them produces a number that describes neither.
Define what you are counting before measuring it, and be honest about which conversions are worth anything. A newsletter signup and a request for a quote are both conversions and one of them is a customer in three weeks. Businesses that count everything as a conversion report an impressive rate and cannot tell whether the business is growing.
The comparison worth making is against yourself last quarter, with the traffic mix held roughly constant. A rising conversion rate alongside falling traffic frequently means you lost the low intent visitors rather than that the page improved, and a falling rate alongside rising traffic can mean a successful campaign reaching people earlier in their decision. Reading either number alone produces the wrong conclusion.
For a first year business the sample size is usually the real constraint. A page receiving two hundred visitors a month and producing six enquiries has a rate of three percent, and the difference between six and nine next month is noise rather than improvement. Drawing conclusions from small numbers is how businesses convince themselves that a change worked, and the honest position is that most small sites cannot measure incremental change reliably.
What you can do at low volume is watch for the obvious. A page with meaningful traffic and no conversions at all is telling you something. A form nobody completes is telling you something. Those are visible without statistics, and they are the failures worth finding.
Improve it by removing rather than adding, which is consistently the higher return direction. Fewer form fields, one action rather than four, the phone number visible without scrolling, and the answer to the question the page exists to answer near the top. Each of those is known to work and requires no testing to justify at your volume.
Then check the rate on a phone separately, because it is frequently materially lower than desktop and the gap is where most of the available improvement sits. Most visitors arrive on a phone, and a mobile rate at half the desktop rate means the majority of your traffic is landing on the worse version of your business.
Record the number monthly alongside the traffic and the source mix, even before it is reliable, because the series is what eventually becomes useful. A single month tells you almost nothing at small volume, and eighteen months of the same figure recorded consistently reveals trends that were invisible at the time. Starting the record costs nothing and cannot be done retrospectively.
Compare the rate against your own best month rather than against any external figure, since that is the only benchmark built from your traffic, your offer, and your market.