How to Turn One-Time Visitors Into Repeat Customers

Most advice about repeat custom starts with a loyalty card and ends with a newsletter, and never once mentions your website. That is a missed opportunity, because your site keeps a quiet record of who came back for a second look and who never returned, and it does that without asking anybody to fill in a form or carry a stamped card in their wallet.

The trouble is that analytics tools tend to report one big number — visitors — and leave you to guess what is inside it. A month with 1,200 visits could be 1,200 strangers who will never think of you again, or it could be 900 strangers and 300 visits from people who have already decided you are worth remembering. Those are two completely different businesses, and they need two completely different weeks of work. This post is about finding the second group, understanding what they do differently, and giving them somewhere to go next.

There are two figures worth pulling out of your stats before anything else. The first is returning visitors: people who have been to your site before, on the same device and browser, and have come back. The second is direct traffic: people who arrived without clicking a link from somewhere else. They typed your web address, used a bookmark, or tapped a link they had saved.

Both are signs of intent rather than accident. Someone who found you through a search for "gluten free birthday cake near me" may well have been comparing five bakeries. Someone who typed your address into their phone at 8am on a Saturday has already chosen you and is checking your opening hours. The first person is a maybe. The second is nearly a customer, and they need almost nothing from you except for the information to be correct and easy to find.

Take a rough example. A bakery has 1,200 visits in a month. Of those, 210 are returning visitors and 340 arrive directly. Those overlap heavily — a returning visitor often arrives directly — but together they suggest that somewhere between a fifth and a quarter of the month's traffic is people who already know the name. That is a group worth serving properly, and it is usually a lot cheaper to serve than the strangers. Understanding how a visit is counted in the first place helps here, and the definitions of sessions and unique visitors are worth five minutes of reading so the numbers stop feeling arbitrary.

Before you get too pleased with a rising direct number, here is the honest limitation. Direct is partly a bin for visits that could not be traced. If someone sends your link in a WhatsApp message, or clicks it inside a desktop email program, or opens it from a PDF quote you sent, the browser often does not pass along where it came from. Those visits land in direct alongside the genuine bookmark-and-type crowd.

So do not read direct traffic as a pure loyalty score. Read it as a trend with a known wobble in it. If your direct traffic climbs steadily over three months while you are not running any ads, and your returning visitors climb with it, something real is happening: word of mouth, repeat custom, people remembering your name. If direct jumps in a single day and nothing else moved, someone probably shared your link in a group chat. There is more detail on what does and does not end up in that bucket in the notes on direct traffic.

The practical version: never judge either number on a single week. Look at a rolling three-month line. Loyalty moves slowly, which is exactly why it is worth having.

Once you know you have returners, the useful question is what they came back for. Two things tell you most of it: how many pages they look at, and which page they arrive on.

First-time visitors usually wander. They read the about page, look at photographs, check whether you are local, and leave. Returning visitors are efficient and slightly boring. They go straight to one page — the menu, the price list, the opening hours, the contact form — do the one thing they came for, and go. That is why a returning visitor can look worse on paper: fewer pages, shorter visit, a higher share of visits that end on the page they landed on. It is not disengagement. It is someone who knows their way around.

The number to compare is pages per session, and the comparison is between groups rather than against some industry benchmark. If your returners look at 1.4 pages and your new visitors look at 3.2, that is a normal shape. If returners are looking at four or five pages and bouncing between them, that is usually a sign they cannot find the thing they came back for, which is a navigation problem you can fix in an afternoon.

There is nearly always one page that returning visitors go to more than any other. For a bakery it might be a Friday specials post. For a solicitor it is often a fees page. For a plumber it is the phone number and the emergency callout section. For a freelance designer it is the portfolio.

That page is the single most valuable piece of your website, and it is also the page most likely to be quietly broken, because you stopped looking at it two years ago. The opening hours are last Christmas's. The downloadable menu is a PDF that takes eleven seconds to open on a phone. The contact form sends to an email address nobody checks. A link in the footer goes nowhere.

It is worth running a proper look over your site rather than trusting your memory of it. Our free website report card checks the mechanical things — loading speed, broken links, pages nothing links to, the basics search engines look at — and gives you a short list of what is wrong. Fix whatever it flags on the page your returners use first, and leave the rest for later. The people who already came back deserve the first repair.

A second visit needs something on the other end of it. This is where most small sites stall, because the standard advice is "start a newsletter", and most owners either hate the idea or start one and abandon it by week five. A newsletter is one option, not the option, and an abandoned one is worse than none.

What works better is a page that changes on a predictable schedule, tied to something you already do. The bakery that posts Friday's specials every Thursday afternoon has trained a few hundred people to check on Thursday evening. The solicitor who publishes one plain-English explainer a month — what actually happens at completion, what a lasting power of attorney costs you in time — gives people a reason to look again before they instruct anyone. The plumber with an honest page on what a job costs and why gets bookmarked by every landlord who reads it. The designer who posts a short note about a finished project gives past clients a reason to keep an eye on them.

The rule is consistency over ambition. Something small that appears every week beats something impressive that appears twice a year. And put the date on it, so returning visitors can see at a glance whether there is anything new.

Loyalty dies from friction more often than from indifference. A few unglamorous things help more than any content plan. Keep your web address short enough to say out loud, and then actually say it — on packaging, on the van, at the bottom of invoices, in the voicemail greeting. Do not change your page addresses when you redesign, because every bookmark and every saved link breaks silently and those visits reappear as people who could not find you. Put the phone number, address and opening hours where a returning visitor can see them within one tap on a phone, not three.

Then decide what the second visit is supposed to produce. Pick one action — a booking, a form, a phone tap, an order — and tell your analytics to count it. That is what a goal is: the specific thing you want visitors to do, counted separately from general traffic. Once it is set up, you can see whether returning visitors complete it more often than new ones, which they usually do by a wide margin. That gap is the argument for spending your time on the people who already came back rather than chasing more strangers.

Be clear-eyed about the limits. A returning visitor is recognised by device and browser, so somebody who reads you on their phone on the bus and then books on their laptop at home will look like two different people. Somebody who clears their browsing data, or uses private browsing, or gets a new phone, becomes a stranger again. Shared computers in an office muddle things in the other direction.

There is also no link between your website stats and your till or your diary unless you build one. Analytics cannot tell you that Tuesday's returning visitor is the woman who has bought a sourdough every week for four years. It can tell you that the number of people coming back is going up or down, and roughly what they look at when they do.

So treat these figures as a direction of travel rather than a headcount. If returning visits and direct visits are both rising quarter on quarter, whatever you are doing for existing customers is working. If they are flat while new visits climb, you are renting attention rather than building anything.

None of this needs a strategy day. It needs a quarter of an hour, once a week, spent on three questions: how many people came back, which page they went to, and did anything they need appear to be broken. Then one small change, and a note of what you changed so you can tell later whether it helped.

That is the job Crafty Meerkat is built for. It is website analytics with a built-in AI advisor, which means the weekly email does not just show you that returning visits moved — it names the specific pages your returners are using, flags the ones that are slow or broken, and ranks a handful of changes so you know which one to do first. You can read about how those weekly reports are put together, or just start a free trial, put the tracking snippet on your site in about five minutes, and see what next Monday's list says about the people who already came back once.