Why your business needs a website — and when it actually pays for itself

The short answer: a website pays for itself when it takes over work you or your phone are doing today — answering the questions that repeat, taking enquiries after hours, and settling the first impression before anyone calls you. If none of those three exist in your business, a website will be a cost rather than an investment, and it is better to know that before than after.

That is the whole subject of this article. Not "is it worth it", because anyone who sells websites answers that the same way. The question is whether the arithmetic works in your case — and that can be worked out before you spend anything.

The demand side: seven in ten adults

Start with how many people are on the other side, because everything else rests on it.

In the preceding twelve months 69.7% of persons aged 16–74 bought or ordered goods or services online, according to the Statistics Poland (GUS) report "Information society in Poland in 2025". A year earlier it was 67.4%, so the number of buyers grew by 2.3 percentage points. Note the base: that is a share of all persons aged 16–74, not only of internet users — a distinction worth keeping, because it is the one most commonly dropped when this figure is quoted.

That figure sits on one side of the ledger. On the other sits what companies do about it: 67.5% of Polish enterprises have their own website, and 47.8% have a social media account. So one firm in three in Poland still has no website.

That pairing says two things at once, and both matter.

A website has stopped being a differentiator. Two competitors in three have one, so having one is not a sales argument — it is a condition of being in the conversation at all. Nobody will choose you because you have a website. Some customers will drop out because you do not have one and there is nowhere to check whether you are real.

One third of the market manages without. That is not statistical noise, and it is not a pile of firms that "haven't got round to it". Those are working businesses whose sales run through another channel — referrals, tenders, a contract with a single buyer, or a counter. For some of them a website genuinely is not the first need, and we will come back to that at the end.

There is a third, less obvious reading, and it concerns pace. With demand growing by more than two percentage points a year and two thirds of firms already on the supply side, the direction is clear and the pace is slow. A year's delay is not a disaster, but there is nothing to wait for either: the conditions will not change enough to make the decision look different.

But not everyone, and not equally

The average is a poor adviser if your customers are not a cross-section of the country. The same data split by place shows where this arithmetic looks different.

Place matters, but less than it used to. Among people in cities 73.8% bought online, in rural areas 63.7%; the highest share is in the Mazowieckie macroregion (82.1%), the lowest in the Eastern macroregion (54.2%). The city–rural gap narrowed from 11.1 to 10.1 percentage points in a year. Whatever else is true of a business operating thirty kilometres from a small town, most of its customers still buy online.

Age matters too. Among people aged 65–74 only 12.3% have basic digital skills (GUS). If your customer is eighty and ordering heating oil or a boiler repair, that does not mean a website is pointless; it means their daughter or their grandson will be the one who lands on it, and they are its actual reader. The site stops persuading the buyer and starts supplying arguments to an intermediary.

That distinction changes the project. A site for an intermediary has to be easy to forward, summarisable in one sentence, and has to contain what somebody checking a contractor on behalf of someone else is looking for: a number, a scope, prices or a range, and evidence that the firm genuinely operates. A site for a direct buyer can instead lead to a basket or a booking.

Selling to businesses is a third case. If your customers are companies rather than consumers, the figures above say little about you — and the question becomes not "do my customers buy online" but "do they check suppliers online before calling". The answer there is effectively always yes, but the consequence differs: the site is not selling a product, it is selling credibility ahead of the first conversation.

What a website actually does when it works

Three things. Each can be checked in your own business without any tool — you only have to recall the last month.

Worth saying up front what is not on that list. It does not include "building a brand", "a professional image" or "an online presence" — not because those are untrue, but because none of them can be checked or counted, which makes them unfit as grounds for spending money. The three below are fit, because each leaves a trace: in the inbox, in the calendar, or in the conversations that did not happen.

It takes enquiries when you are not there

The most underrated function and the easiest to verify. A building firm we work with gets most of its enquiries between 8 and 11 in the evening. People get home, notice the plaster coming off the bathroom wall, and look for a contractor immediately — not at nine the next morning. A phone does not answer at that hour. A form does.

The mechanism has two floors. The first is obvious: an enquiry that would have arrived at ten at night and evaporated now lands in the inbox. The second is less visible — an enquiry written in the evening is a different thing from a daytime call. Someone typing has time to describe the problem, attach a photograph and give dimensions. The conversation starts from specifics instead of from "hello, do you do bathrooms?".

It filters out the questions you answer for the hundredth time

A section with prices, scope and lead times is not decoration — it is a filter. Instead of ten conversations of which eight end in "that's too expensive for us", you get two conversations with people who had already seen the price and called anyway.

The effect runs the other way too, and it is worth saying plainly: if you are afraid to show prices, the site will filter out nobody. It will generate the same conversations as before, only later. A range, a "from" figure, a sample project priced out — any of those forms will do. "Price on application" will not, because it is a sentence carrying no information.

This is the one of the three whose value can be priced directly. If eight fifteen-minute conversations a week leave your calendar, that is two hours a week — around ten working days a year.

It settles the first impression before you make it

A furniture maker whose site we rebuilt recorded roughly 40% more enquiries with an unchanged offer and unchanged prices. Only the layout and the presentation changed.

That is not design magic. It means the previous site was costing them enquiries they never saw — because people were closing it before the form, and a closed tab leaves no trace in an inbox. This is the hardest part of the conversation: the cost of a bad website is invisible from inside the business. You see what arrived, never what fell away on the road.

A channel you own and a channel you rent

There is one more reason that rarely comes up in conversation and that decides matters over a few years. A website is not the only place a business can be present — it is the only one that belongs to it.

The figures sit closer together than you would expect. 18.3% of enterprises made e-sales (2024 data, unlike the other figures in this text, which are 2025), most often through their own websites or apps (12.9%), then through online trading platforms (10.2%), and through EDI 3.3%.

The difference between these channels does not surface while everything is going well. It surfaces on the day the channel's owner changes the rules — raises the commission, reorders the ranking, trims the reach, or suspends an account for a reason no form will explain. That is when it becomes clear whose channel it was.

The easiest way to see it is to ask what exactly disappears along with the channel. With a suspended profile you lose not only reach but the entire history of conversations with customers and the only list of people who ever got in touch. With a marketplace you lose the reviews you spent years earning, because they belong to the account rather than to the business. With your own site the equivalent risk is the domain alone — which is why the single ownership question to put to a contractor is: whose name is it registered in?

This is not an argument against marketplaces or social platforms. Most firms use them alongside a site and that is sensible — social media reach people search engines do not, and a marketplace brings traffic you could not build yourself. The argument is different: your own site is the only place where the address, the content and the contact list belong to you — which makes it what remains when the other channels change their terms.

One practical consequence, worth writing down: if you are building presence in a channel you do not control, make sure it leads somewhere you do. A profile that routes only to private messages builds nothing durable.

The arithmetic: what the site has to bring in

Here begins the part most articles about "the benefits of a website" skip, because it needs numbers instead of adjectives.

The mechanism is simple and has four quantities. You know three of them already, or will within a month; the fourth is in the contract.

The return is: visits × contact rate × job value × close rate, minus running cost.

Two of those need a comment, because they are where the usual mistakes live.

The contact rate. A well-built business website turns 2% to 5% of visits into contact. That is the range we see among service clients — shops and landing pages work differently and have ranges of their own. We settle that separately, under conversion rate.

The close rate. An enquiry is not a job. If you close one in three, put in a third — without it the arithmetic comes out three times too favourable, and that is the most common way such calculations lie.

An example with actual proportions, though deliberately without amounts: a car workshop we work with had a thousand visits in a month, and forty of them ended in a booked service — 4%, the upper part of the range. Multiply that by the average value of a job and you have the revenue attributable to the site. Here the close rate was one, because a booked service is the job; in a business where a quote and a negotiation sit between the enquiry and the contract, that multiplier has to be added.

We give no currency amounts on purpose. Ours would be the wrong market's, and a figure converted at an exchange rate is worse than no figure at all, because it sounds credible. Your own average job value is the only one that belongs in this calculation.

If the result comes out negative, the problem is almost never the site itself. The problem is that nobody visits it — and that is a job for search visibility, not for another redesign.

How you will know it has paid for itself

The arithmetic above is a forecast. Below are three measurements that will tell you, after a few months, whether it held. All are free and none needs a specialist.

Where enquiries come from. One question added to the form or asked in the first call: "how did you hear about us?". "From Google" and "from a referral, but I checked the site first" are two different things and both count in favour — the second means the site did not bring the contact, it saved it.

How many conversations disappeared. Compare the number of calls asking about price before the launch and after. A drop is the success, not the failure — that is precisely the work the site was meant to take over.

Whether the enquiries are better. An enquiry where the client already knows what it costs and how long it takes closes faster than the same enquiry before the site existed. That is an effect no visitor statistic shows, and one that genuinely changes a month.

A sensible horizon is three to six months from launch — and only on the assumption that somebody is bringing traffic to the site. Publishing it starts none of these three measurements. How long that takes and what it depends on is set out in how long SEO takes.

Three things that ruin this arithmetic

We see them in the same order, project after project.

A form with eight fields. Every extra question reduces the number of forms sent. Three fields will do to start: name, email, message — you can ask the rest in your reply. "How did you hear about us", "budget" and "preferred date" look innocent, and each is one more reason to leave sending until later, which means never.

No price, not even a range. A visitor who finds no information about cost mostly does not ask — they leave. The objection "we can't give a price, every project is different" is true and beside the point: nobody is asking for a final figure, only for an order of magnitude that lets them decide whether to call at all.

A site nobody looks at after launch. An out-of-date offer and a phone number from two years ago cost more than having no site, because they undermine credibility instead of building it. It is the only one of the three that deepens on its own — which is why, when budgeting, maintenance is a more important line than the last ten per cent of design quality.

When a website is not the first expense

We have kept to the honest version of this conversation, so let us finish it. There are situations where we advise against starting with a website — and this is where that one business in five comes back.

All your revenue comes from one channel that works. A referral network, tenders, a contract with a single buyer. Then the first question is not "which website" but whether you want to change that channel at all — a site will not strengthen a channel that does not use it. It is worth knowing, though, that one channel is one risk, and that this is exactly the argument made above about rented channels, at a different scale.

The offer is not settled. A website fixes whatever you write on it. If the offer will be different in three months, you will pay twice — once for the copy and once for rewriting it. Settle what you sell, to whom, at what price and what you do not do, before anyone starts writing pages.

You operate purely locally and have an active Google listing that brings calls. Then the first step is to get that listing right and the site comes later, as the place the listing points to. The reverse order costs the same and works more slowly. Our guide to the Google profile covers it.

Your customers are not online to the degree that would justify it. This is rare and worth checking rather than assuming — but with an audience in the bracket where barely half buy online, and with sales happening entirely over a counter, the first unit of budget may have a better home. Usually the answer is still a website, only a smaller one: a brochure site rather than a full one.

In every other case the question is not "whether" but "which kind" — a different decision, set out in which website makes sense for which business.

How long before it starts paying

Three thresholds, in this order, because confusing them is the most common cause of disappointment in the first quarter.

The first threshold is being indexed, not ranking. Before a site appears in results for anything at all, a search engine has to notice it exists. That is days or weeks, and it cannot be bought — it can, however, be broken, by leaving a staging environment's indexing block on the live site.

The second threshold is your own names. The company name, the name plus a city, a service plus a city in less crowded trades. The effect is often visible within weeks, because there is barely any competition.

The third threshold is phrases somebody is fighting over. Months, and only with regular work. Anyone promising otherwise is promising something they do not control.

The practical conclusion: the first enquiries usually come through channels you already have — the Google listing, a referral, the signature in your email — and search adds to it later. Which is why a verdict of "the site didn't pay for itself" delivered after six weeks measures patience rather than the site.

What is not in this article

Five things, deliberately, so as not to pretend we are settling more than we are.

• Amounts. What it costs and what the bill is made of is kept in the cost section, where the figures are maintained.
• The choice of type. Brochure site, business website, full service or portal — a separate decision and a separate article.
• The minimum scope. What has to be on a company's first site: brochure website.
• The campaign page. A different construction and different measures: landing page.
• How to bring traffic in. The largest part of the answer to "why didn't this pay off" lives in the marketing and SEO section, not in the design of the site.

The shortest summary we can give: a website pays for itself when it has something to take over. If your business has questions that repeat, enquiries outside working hours and customers who check a supplier before calling, the arithmetic will work. If none of those three exists, no website will create them, and it is better to start elsewhere.