Where the numbers come from
Every research and market claim on this site is below, with its source and its limits. If a borrowed number is not here, it is not on the site. Prices and workshop facts are simply stated where they live. The clearly marked concept pages are fictional businesses with illustrative demo data.
The price and the recommendation
The base is 690 € and three working days for a site of four to six sections, written and built. Everything else is an addition on top of that base, not a separate thing: a catalogue adds 690 €, so a site with a catalogue comes to 1380 €; an enquiry bot adds 400 €; a payment provider 345 €. Beyond the six included sections each one adds 80 €, and every two of them add a day. Each language past the first adds 105 € and a day. If your text is already written the work is 70 € smaller; if your photographs are usable, 45 € smaller. A rush job costs 30 % more and takes half the days, rounded up.
One caveat about the days, in the same spirit as the rest of this page. The prices above are decided; the durations are calculated, not measured. They come from how long each part takes on my own trial runs, added together — no client project has yet gone through the whole cycle end to end. I hold to them as promises, and I will say so plainly the first time one slips. When there are enough real runs to speak from measurement, this paragraph is replaced by the numbers.
Additions bought together as a set cost less than the same additions bought one by one, because they are built in one pass — that saving is shown beside the set and it is the only discount here. Work that cannot be priced this way, such as an internal tool built to your own specification, is not guessed at: a written task study is priced on its own, and the build is priced inside it.
What this cannot tell you: it is a price and a duration, not a forecast. It does not estimate your revenue and it does not promise enquiries. There is no range and no upper end — the recommendation names one exact figure before any conversation, and that figure is held for fourteen days.
Where 690 € comes from
From a fresh review of the Serbian market on 28 August 2026: 33 providers with published prices, plus 47 priced classified ads. The median of the classifieds is 170 €. The median starting price of 15 studios is 205 €. The median business package of five pages or more is 270 €. So 690 € is not “the top of the freelance band” — it is about two and a half times the median published studio price.
That comparison is not fair as it stands, because those prices do not include the same things. Adding what the market charges separately — copywriting 120 €, a .rs domain 20 € a year, hosting 50 € a year — the median comes to 417.5 € for the same delivered result. Against that, 690 € is about 65 % more, not two and a half times.
What this cannot tell you: prices move, and these were read on 28 August 2026. Cheaper exists — the floor of this market is 29 €, and a working one-page site is genuinely sold at 100 €. I am not the cheapest option and I am not going to pretend otherwise.
Read again on 29 August, and the answer changed. The figures above come from small providers and classifieds — the cheap end of this market. Taking published 2026 price reviews for the same scope, a five-to-ten-page presentation site in Serbia is quoted anywhere from 300 € to 2 500 €. The middle of that span sits between 599 and 800 €, depending on whether you take the low edge of each quoted range or the high one.
Against that reading, 690 € sits in the middle — no premium at all. Both readings are honest and they disagree.
Read a third time on 29 August, and the honest answer is that there is no answer. Take the same eight published reviews and work out the median three defensible ways — by the bottom of each quoted range, by the middle, by the top — and you get 300 €, 1 013 € and 1 500 €. Those differ by a factor of five. A number that swings fivefold with the method is not a measurement of anything.
So this page stops claiming a position. What the data does support is narrow and I will not stretch it: 690 € falls inside the published spread of 200 to 5 000 €. It is not an outlier in either direction. Whether it is expensive depends entirely on who you put next to it — and the five things below are what you are actually deciding between, which is a better question than the median anyway.
What that 417.5 € does not include, checked across the same 33 providers:
- A named number of revision rounds. Not one of the 21 studios publishes it. Three providers out of 33 do, and all three are under 450 €.
- Copy written for you. The market position is almost uniform: “we insert your texts”, or literally “we can put lorem ipsum there”. Written copy is sold as an extra at 120 €, or 6.5 € a page for filling in what you wrote.
- A deadline stated as a number of days. Eight of the 21 studios publish a deadline at all; two state it as a firm number rather than a range.
- The price staying still. “From” is not a price. One provider advertises the same package at 147, 149, 169, 249 and 295 € on different landing pages.
- Seeing the work before paying for it. Free work before payment exists on this market only below 100 €.
That is what you are paying the difference for. If those five things do not matter to you, the market genuinely has cheaper options and this page is telling you where they are.
Price against time
Three working days is not the fastest on this market, and it would be dishonest to imply otherwise. A one-page site is genuinely sold at 24 to 48 hours: 99 € in a day, 100 € in two. The median published deadline for a business site of five pages or more is about seven and a half working days, so nine days for the Front desk set is slightly slower than the middle of the market. The one place the timing really is ahead is the largest set: thirteen days against a market median of twenty-one.
What is unusual is not the speed but the shape of the promise. Of the 21 studios read, eight publish a deadline at all, and only two state it as a firm number of days rather than a range. Not one of the 21 studios publishes how many rounds of changes are included; three providers out of the 33 do, and all three sit below 450 €.
The deadline figures rest on published price lists read on 28 August 2026 — most providers name a price and no deadline at all.
What I checked and had to take down
This page used to say that in the 400–900 € range no verified package promises a finished site in under seven days. Rechecking the market on 28 August 2026 proved that wrong: a Belgrade studio sells a presentation site in three to five working days from 300 €, another in five to seven from 399 €, and a third names seven days at 427 €. The claim is gone. It was convenient and it was not true, and a page about where numbers come from is the worst possible place to keep one.
A correction, made on 28 August 2026. This page first published that median as 405 € and the premium as 70 %. Both were wrong: with fourteen sorted values the median is the mean of the seventh and eighth, and I averaged the sixth and seventh. The right figures are 417.5 € and 65 %. An outside reader found it, not me — which is the whole argument for publishing the working in the first place.
The share of people who walked away
52 % of people who searched for a local business walked away from one they had found; of those, 32 % left because the price was unclear and 18 % because they could not find a website (BrightLocal 2026 consumer survey, a panel of 1227 consumers in the United States).
What this cannot tell you: it is a survey of consumers, not a measurement of your business. It says nothing about whether anyone is searching for you at all — a site converts the people already looking, it does not find new ones.
Numbers I removed
These are commonly quoted and I checked them. They did not hold, so they are not on the site.
- “97 % of people search online for a local business” — no primary source found.
- “80 % of attention is above the fold” — the original measurement used 21 people in 2010. The same lab recalculated it at 57 % with 120 people.
- “A 100 ms delay costs Amazon 1 % of sales” — a conference slide with no published data behind it.
- “People read 20 % of a page” — not measured but calculated, by dividing a regression coefficient by an assumed reading speed, from 25 mostly academic Europeans browsing in 2005.
- “Fewer options convert better” — two meta-analyses (n=5036 and n=7202) put the average effect at 0.02, which is zero.
If you find something here that is wrong, tell me and I will change it or take it down.