Website marketing strategy — where the customers actually come from

Marketing a firm online comes down in practice to four channels: unpaid search results, paid search, social media, and referrals from other sites. Each works at a different speed, costs a different amount, and leaves something different behind when you stop paying.

Below is a map of those four, and after it the thing most guides skip: with the measurement a small firm typically has, almost half the traffic goes into a bucket nobody knows anything about. We show it on our own numbers, because without that, comparing channels is guesswork.

Four channels — the map to start with

Before the numbers, a comparison on one screen. The "what is left" column is the least discussed when a budget is set, and it decides whether the spend is a cost or an investment.

Channel names say where a visit came from. They do not say what the channel is good for. That distinction decides the budget.

Search, unpaid results (16.1% for us). The only channel where you meet a person at the moment they are looking for a solution themselves. Two features you cannot get around: it works with a delay measured in months, and it does not start until the site is in the index. How long that takes and what it depends on is set out in how long SEO takes.

Search, paid (13.2%). The same moment and intent, bought and immediate. The advantage: it works from day one. The drawback: it stops the day you stop paying, and leaves nothing behind. This is the channel for testing hypotheses and for periods when you need enquiries now rather than in six months.

Social media (10.7% unpaid, 4.9% paid). You reach people who were not looking for anything. Hence traffic that can be plentiful with few contacts in it — not a flaw in the channel but its nature. The risk of resting your whole presence on somebody else's platform we cover separately.

Referrals from other sites (5.7%). Directories, articles, mentions, trade portals. The least impressive channel and the most underrated, because it works long after you have forgotten about it.

Note one thing: all four channels together are a little over half the traffic. The rest is a bucket that cannot be attributed. Any marketing plan ignoring that rests on half the data and pretends it is the whole.

Marketing on no budget — what to do before you spend

Before you spend the first euro, it is worth exhausting what is free — in a small firm that is usually where the largest unused reserve sits.

The Google Business Profile — the cheapest channel you have

Worth listing separately, because it is not its own line in the table above, and for firms working locally it is sometimes more important than the site.

The profile and the site do not compete; they do two jobs on the same path. The profile decides whether anybody sees your name in local results. The site decides what happens next, once the name has appeared and the customer wants to check who they are dealing with. A firm with a good profile and no site collects calls from people who could verify nothing; a firm with a site and no profile is absent at the moment the decision is taken.

For budgeting that gives a simple order: if you work locally and the profile is unfinished, that is the cheapest action on the list — free, and doable in one evening. What to fill in and how to run it is in a separate guide to the Google Business Profile.

Referrals — you cannot buy them, but you can increase them

Since the largest share of traffic is in practice referrals, it is worth asking what can be done beyond waiting. Three things, all free and all neglected.

Ask at the moment the customer is pleased. Not a month later by email — then, as they take delivery and tell you themselves that it is good. It is the only moment when the request is not awkward, and the only one at which most people actually do something.

Make referring technically easy. Somebody who wants to recommend you needs something to send: a short web address, a profile with a code, a ready link to leave a review. If it requires digging your name out of a six-month-old inbox, it will not happen — and not because they did not want to.

Turn the referral into a review that stays. The conversation disappears; a review in Google stays and works on later customers who will never meet the first one. It is the only way we know of making a one-off referral work more than once. How to ask and how to reply we describe separately.

Keep in mind what follows from the first part of this article: you will not see the effect of those three in any report. Referral visits land in "direct" and mix with the rest. The only measure left is the question you ask new customers — which is why the checklist below weights it as highly as any event in the reporting.

Why the map of channels is not enough

Most guides to marketing a firm online end here. They list the channels, describe what each does, and leave the reader to decide which to spend on.

The trouble is that the decision cannot be taken without answering a question those guides do not ask: can you see which channel worked? Below is what that looks like in reality — on our own data, because we have no way of verifying anyone else's.

This is the traffic on www.digitalvantage.pl, our own company site, over the last twelve months, split by channel, as Google Analytics sees it.

The scale matters less than the proportions, but we give it so it is clear what we are talking about: 6,016 sessions over a year. A services firm's site, not a portal. If your numbers are of that order, the conclusions below apply to you directly.

The same data month by month

An annual share smooths out everything interesting. The same traffic by month looks different and says more.

Two things are worth reading out of this, and both teach more about measurement than about the market.

The largest line is the least stable. Direct visits rise to 576 a month, then fall to thirty-odd. The channel we know least about moves the total most — so a budget planned on it would be planned on the weakest number available.

April 2026 is where comparability breaks. Paid campaigns started then and too much changed at our end to read both stretches as one trend. We mark the break rather than smoothing it, because this is exactly what is worth spotting in somebody else's report: a jump or a collapse in the data more often means a change in the measurement than a change in the market. Before drawing a conclusion from your own chart, check what was changing on your side.

"Direct visits" is not a channel

This is the commonest misunderstanding in conversations about marketing, and worth closing at the start.

A visit lands in this category when the browser passed no information about the source. It does not mean "the user typed the address from memory". At least four different things end up there:

• A typed address or a bookmark — the only case that matches the name of the channel.
• A link clicked in an email, in a messenger, or in a PDF. A mail client is not a web page and has nothing to pass on.
• A link from a mobile app that does not pass a source.
• A move from HTTPS to a site partly served over HTTP — the source information is then deliberately stripped.

Worth looking at how Google defines the channel. The documentation on default channel groups describes it in plain language as traffic "via a saved link or by entering your URL" — but the matching rule it gives is different: source equal to "(direct)" and medium set to "(not set)" or "(none)". Not the same thing. The first sentence describes a user's intention, the second missing data. Everything meeting the second condition lands in the channel, whatever the user did.

The second point changes how the whole table reads. A recommendation from a friend, sent by text or over a messenger, looks in the reporting exactly like typing the address from memory. The largest bar on the chart is in large part not "direct traffic" — it is referral traffic nobody can count.

The practical conclusion is uncomfortable: the channel that brings most customers to most small firms is the one you cannot scale by buying.

Three things the table does not show at all

The figures above are a lower bound, for three independent reasons.

Cookie consent. The measurement sees only users who accepted the banner. The rest are invisible — not "approximated", absent. Every figure in the table is understated by an unknown but non-zero amount.

Phone calls. The most expensive measurement mistake we know, because we made it on our own site. For a while our report showed zero form submissions — and that was true. What was untrue was the conclusion that there were no enquiries. People were ringing, and a phone call leaves no trace in any web report. If there is a phone number on the site, some contacts are certainly not being counted.

Searches for your own name. Somebody heard about you from a customer, typed the company name into a search engine and arrived. In the report that looks like an SEO success. In reality it was a referral, and the search engine acted as a phone book. That one mechanism can credit SEO with effects it did not cause — and equally hide how well referrals work.

Cost per enquiry — the only number that lets you compare channels

Visits from different channels are not comparable, because they come from people interested to different degrees. A hundred visits from search and a hundred from a Facebook post are entirely different things. What is comparable is the cost of one enquiry, worked out per channel.

The arithmetic is simple and fits in one table. Fill it in with your own figures for the last quarter — a quarter, not a month, because at small numbers a month is noise.

Two notes on this calculation, without which it produces nonsense.

Your own time is a cost. A "free" channel run for four hours a week is not free — it is paid for with time that could have gone on a job. Enter that honestly, or social always wins the comparison.

A channel with no attributed contacts does not mean "ineffective". It means "unmeasured". Before writing one off, check whether it had any way of showing up — see the checklist below.

In practice this calculation usually ends in one of two discoveries: the most expensive channel turns out to be the only one bringing customers rather than enquiries, or the cheapest turns out to cost more than advertising once time is counted.

The order of channels depends on the trade

The same set of channels arranges itself differently depending on how the customer decides. Three arrangements we see most often:

Local services with an urgent need — plumber, electrician, repair service, builder. The customer has a problem now and looks for the nearest contractor who answers the phone. Here the Google Business Profile and local results win, and the site confirms the firm is real. Paid search works but can be expensive, because everybody in the area bids on the same phrases.

B2B services with a long decision — projects, consulting, contracting for firms. Weeks pass between first contact and contract, and more than one person decides. Here content and unpaid results do the work, because somebody reads, compares and comes back. Referrals weigh heaviest of all the channels and are the worst represented in reports.

Products and mail order. The decision can be immediate, so advertising works from day one and social channels genuinely sell, because the product can be shown. The only one of the three in which traffic with no prior intent turns into revenue.

If your firm fits the first arrangement and the budget is going into content and ranking for national phrases, that is the commonest mistake in this list and the easiest to correct.

Check your own measurement before you spend anything

Buying traffic while the measurement is broken ends the same way every time: the money goes to the channel that looks best in the report, not the one that brings customers.

Why visits are rising and the phone stays silent

The commonest question we get after the first months of any activity, and it usually has one of three causes. Worth telling them apart, because they lead to three different decisions.

The traffic does not match the offer. People arrive because the site answers a question your customer does not ask. The classic example: a contractor writes a guide on "how to lay tiles yourself" and gains traffic from people who have just decided not to hire a contractor. Search Console shows it — if the phrases you appear for start with "how to yourself", you have your answer.

The traffic matches, but the site makes contact hard. A number hidden in the footer, a form with eight fields, nothing about price or the area covered. The traffic is right and is lost on the last metre. The cheapest of the three to fix and the most often missed, because no report shows it — only a stranger trying to get in touch does.

The contacts are there, you cannot see them. They ring. They write on a messenger. They reply to a message you sent a month ago. Before deciding a channel does not work, count every way people reach you — including the ones no tool sees.

The order of checking is always the same: the third cause first, because it is free; then the second, because it is cheap; the first last, because fixing it means changing content or campaigns.

The first three months, starting from zero

If you want an order rather than a list of possibilities, this one works at services firms and needs no budget to start.

Month one — visibility and measurement. A finished Google Business Profile, contact details fixed on the site, an event on the form, a tappable phone number, and one question asked of every new customer. None of it costs money; all of it costs a few hours.

Month two — content that answers real questions. Not a blog. Service pages describing what customers ask about on the phone: scope, area, lead time, indicative price. That material answers search queries and shortens conversations at the same time.

Month three — only now paid traffic, if at all. With working measurement and pages worth pointing at, you can buy traffic and see within two weeks what a conversation costs. Without the first two months, the same campaign produces numbers that mean nothing.

The order is dull and therefore rarely proposed. Its advantage is that each step can be abandoned without loss — while a campaign switched on at the start usually costs as much as learning on your own data.

What this means for a budget

Three conclusions, all straight from the figures above.

Do not move budget to the channel that "has the most visits". The largest line in the table is a bucket for unrecognised sources. You can only compare channels you measure the same way — and even then the sensible measure is conversations, not visits.

Before adding traffic, check what happens to it after arrival. Doubling visits at an unchanged contact rate doubles the cost of one enquiry, if the traffic was paid for. That side of the calculation is in the article on conversion optimisation, and measurable goals in website KPIs.

Do not judge a channel over less than a quarter. At a few dozen contacts a year a month is noise, not measurement — two good weeks and two poor ones can reverse the result either way. Switching a channel off after thirty days is a coin toss, only more expensive, because it cost a month of spending and usually ends with the budget moving to a channel judged just as hastily.

Count referrals by hand. One question in the first conversation — "how did you come across us?" — recorded for three months gives you knowledge you cannot buy in any tool. At most services firms that ask it, the answer looks different from the report.