DEV Community: Roman Mikhailov (Роман Михайлов)

How much does a business website cost, and what really drives the price

Roman Mikhailov (Роман Михайлов) — Tue, 20 Oct 2026 08:59:14 +0000

Ask five developers how much a website costs and you will get five answers: a few hundred dollars, a thousand, three thousand, ten thousand, and "it depends". The funny part is that they are all right. Each of them is pricing something different. One assembles a page from ready-made blocks. Another writes the copy and connects your CRM. A third designs everything from scratch, adds online payments and integrates with your accounting system.

So the real question is not "how much does a website cost", but "what kind of website does my business need right now, and what exactly is included". Let's go through it step by step.

Start with the stage of your business, not the website

The classic mistake of a new business: ordering a big, beautiful website before it is clear what you sell and to whom. Three months later the offer changes, and the expensive site has to be rebuilt.

A more sensible order:

You are still testing the niche. A free business listing on maps, an ad on a classifieds site or a marketplace, a simple page. You need to test demand, not design.
First steady sales. A landing page or a small site. Now you have real customer questions, cases and prices.
Several stable services. A multi-page site where each service has its own page.
Many products, but deals are discussed manually. A catalog without online checkout.
Customers buy fully online. An online store with cart, payment and delivery.
The business process itself is a software product. Only here does custom development make sense.

Part of the first step costs nothing. In Russia, where I work, adding a company to Yandex Business is free, and it even lets you generate a simple mobile-friendly site from your listing data. The same idea works with free map listings in other countries.

One more thing. At the start, a website usually does not "bring customers". It turns interest that already exists into trust. Someone saw your ad, your map listing, or got a recommendation, and now they check: who you are, what it costs, whether there are real projects, and whether the terms are clear.

Types of websites and what they cost

A landing page is built around one offer and one action: send a request, call, book. While your niche is still moving, this is the most flexible format: you can rewrite the offer without rebuilding dozens of pages.

A multi-page service site makes sense when the offer has settled and different services need different arguments. Instead of one "renovation" page you get "new builds", "bathrooms", "design project", "prices", "examples". Search engines like this too: one clear page per need.

A catalog looks like a store, but ends with "get a quote" instead of "pay". Good for custom furniture, equipment, B2B products.

An online store adds what beginners often forget: cart, checkout, delivery options, payments, notifications, order statuses, receipts, returns, stock, product import.

Custom development is for genuinely non-standard processes: configurators, client portals, complex pricing, platforms.

As for numbers, international reference points look like this: Upwork currently shows a median rate for web developers of about $30 per hour, with a typical range of $15-50. A Shopify industry overview from September 2026 suggests roughly $3,000-15,000 for a professional small-business site of about ten pages. That is a platform vendor's estimate, so treat it as a rough guide.

The Russian market shows how slippery the word "average" is. One industry rating of web agencies found an average landing page price around 48,800 rubles among WordPress agencies. Another rating, of "affordable studios", came out at about 154,600. Same type of site, more than three times the difference, simply because different companies were counted. At the same time, a real landing page project on Tilda with a mobile version and leads sent to a spreadsheet, Telegram and email was published on a Russian freelance board for 15,000 rubles and four days of work.

The lesson: "a website costs X on average" means very little. Compare scope, not totals.

What really drives the price

The number of pages matters, but not linearly. Five similar service pages can cost less than one complex calculator. Price goes up with:

custom design instead of ready-made blocks: prototype, designer, approvals;
every page with a unique structure;
catalog and filters;
CRM integration: fields, statuses, duplicates;
online booking: external service, calendar, notifications;
online payments: payment provider, refunds, possibly fiscal receipts;
chatbots;
SEO, copywriting, photo and video;
migrating an old site while keeping its URLs.

The cost after launch

Comparing only the build price is risky. A $500 site can end up cheaper over two years than a $150 one if the cheaper one depends on paid plugins, or every small change costs money.

Website builders charge a subscription. Tilda's basic plan in Russia, for example, is about 6,000 rubles a year when paid annually.
WordPress itself is free and open source. You pay for hosting, the domain and, above all, people.
A domain is not bought forever. It is a time-limited right that must be renewed. Look at the renewal price, not the first-year promo: in March 2026 one Russian registrar listed a .RU domain at 226 rubles for registration and 1,199 for renewal, more than five times higher.

With self-hosted sites, the expensive part is not the software but maintenance and technical debt: updates, plugins, security, backups. Google has described a case where attackers injected hidden links, pages and scripts into a restaurant's WordPress site, and the owner had to clean it up and request a review. That does not make WordPress bad. It means a self-hosted site cannot be launched and forgotten.

Where the hidden costs are

Before you sign, check whether the price includes:

domain, hosting, SSL;
CMS license, paid themes and plugins;
source files and the mobile version;
content upload and migration;
images and their licenses;
forms, notifications, CRM, booking, payments;
backups and security;
analytics and privacy documents;
basic technical SEO;
bug fixes after acceptance and training on how to edit the site.

The most dangerous line in any quote: "turnkey website" with no appendix explaining what the key actually opens.

Where extra money is worth it

If you have a little more budget, spend it on things that change decisions rather than on animations:

Real photos of your work, team and premises instead of stock images.
A clear offer and honest answers to customer objections.
Speed on mobile. Swappie reported a 42% increase in mobile revenue after working on Core Web Vitals. A study published by Google covering 37 brands and about 30 million sessions linked a 0.1-second improvement in mobile speed with an 8.4% rise in retail conversions. These are specific cases, not promises, but the direction is clear.

If your niche changes

My name is Roman Mikhailov, I build websites for businesses at RM Systems, and the main advice I would give at the start is this: the first version of your site should be cheap to throw away. You should not mind rewriting it after twenty conversations with real customers.

A rough guide to how much a rework costs relative to the original build:

new offer in the same framework: about 20-40%;
new copy, photos and some blocks: 30-60%;
a different audience and structure: 60-100%;
services turning into an online store: 80-150%.

This is a budgeting heuristic, not market data. Changing text on a site builder is cheap. Changing the business model is not.

The domain matters too. If you register something like "bathroom-renovation-city" and six months later start selling kitchens, you either live with the wrong name or move. Moving a site that already ranks in search is a separate project with redirects for every old URL. Until your niche is settled, pick a neutral brand name.

Don't forget the law

As soon as your site collects a name, phone number or email, privacy law applies. Rules differ by country, so check yours before launch, not after. Russia is a good example of how strict this has become: since September 2025 consent to personal data processing must be a separate document, not a line under the "Send" button, and since July 2025 data on Russian citizens collected online must first be stored in databases inside Russia. If your site sells online, consumer protection rules for distance selling and payment receipts apply as well.

This is not something to hand to a web developer as "we'll add the standard checkbox". For e-commerce, healthcare, finance or children's services, talk to a lawyer.

SEO: basic hygiene now, campaigns later

Basic technical setup should be part of any new site, even if you never plan to "do SEO": proper titles, indexable pages, clean URLs, a sitemap, one main domain version, a solid mobile layout, connected webmaster tools and analytics. That is hygiene, not promotion.

Full SEO campaigns are a different story. Agencies commonly say first noticeable results take 3-6 months, and a fair judgement of payback takes 9-12. Google states plainly that nobody can guarantee a number one ranking. So for the first 3-6 months, do not build your financial model on leads from organic search.

Will the site pay off?

A hypothetical example, not a forecast:

the site costs $1,000;
upkeep is $100 a year;
each new customer leaves you $250 after direct costs.

Then about five genuinely new customers brought by the site cover the investment. If one in four qualified leads becomes a customer, you need roughly 20 good leads.

Now SEO at $1,000 a month. Six months is $6,000, so it needs about 24 new customers just to break even. In a niche where a customer is worth $2,500 that is a bargain. In a niche where a customer is worth $50 it barely makes sense.

Count the whole chain, not visits: where the person came from → did they leave a request → was it a real lead → did they buy → how much did you earn.

How to hire, and keep control of your site

The domain and every account are yours. Register the domain in your own or your company's name, with your own email. Create the site builder, hosting, analytics and CRM accounts yourself and give the contractor access, never the other way round.

Write the rights into the contract. "I paid, so it's mine" does not always hold. Russian law, for example, treats commissioned works, works created by the author directly and works created while performing other services differently. So the contract should state that after payment you receive the exclusive rights to the code, design and copy created for the project, and the contractor should list what does not transfer: licensed fonts, stock photos, paid plugins, ready-made templates.

Compare like with like. Give every candidate the same short brief: who the site is for, what action you want, which pages, where leads go, which integrations, how many revision rounds, and what is out of scope.

Check live projects. Open them on your phone, send a test request, and ask what exactly the contractor did on each one. Before a big contract, pay for a small first stage, like a homepage prototype.

Red flags:

a guaranteed number one ranking;
the domain and hosting registered only to the contractor;
refusal to hand over admin access;
a "turnkey site" with no detailed scope;
100% prepayment with no milestones;
"SEO included" with no list of tasks;
no clear answer on how to take your site with you if you part ways.

The bottom line

Until your niche is confirmed, a sensible budget looks like this: free map listings and a simple page first, a domain registered to you right away, a modest landing page on a site builder once demand is confirmed, and no custom design, complex CRM, online store or monthly SEO retainer yet.

Move to an expensive site not when the old one "looks unprofessional", but when you can say: this is my audience, this is what they need, this is why they buy from me, this is what a customer costs and this is what I earn from one. From that point on, a bigger site, SEO and a catalog stop being image spending and become an investment you can measure.

What was the biggest hidden cost in your first website? 👇

AI answers in Google and Yandex: fewer clicks, and how a site becomes a source

Roman Mikhailov (Роман Михайлов) — Sat, 17 Oct 2026 19:36:42 +0000

Search for something like "how to choose a stretch ceiling", and there is a good chance a ready-made answer appears above the links. In Google it is called AI Overviews. In Yandex, the main search engine in Russia, it is a quick answer from Alice AI.

Convenient for the user: read the answer, close the tab. For the site owner, rankings stay the same while clicks go down.

At RM Systems I build websites for businesses in Russia, so I watch both engines. Here is what these answers are, how many clicks they take, how they pick sources, and what you can check today.

What these answers are

Google shows AI Overviews above the results and has a separate AI Mode where you talk to search like a chatbot. According to Google, AI Mode is used by more than a billion people a month.

Yandex shows quick Alice AI answers above the results, with links to the pages the answer was built from. According to Yandex, in Q2 2026 quick answers appeared for about 42% of queries and reached 49.5 million people a month. Yandex estimates its own share of the Russian search market at about 70%, so for most Russian businesses Alice matters first.

How many clicks they take

The research points the same way:

Pew Research, US, 2025. Almost 69,000 searches from 900 people. With an AI summary, people clicked a regular result in 8% of cases, without it in 15%. Links inside the summary got clicks in 1% of cases.
Indian School of Business and Carnegie Mellon, 2026. A browser extension hid AI Overviews for some users. Where the summary was shown, clicks to sites were 39.8% lower.
University of Pennsylvania and Northeastern University, 2026. AI Mode cut the share of clicks to external sites by 18.8 percentage points.
Russia. According to Kokoc Performance, informational sites have lost about 30% of search traffic on average since early 2025, and from 15% to 60% in some categories.

Informational queries ("how", "what is", "why") suffer most. Commercial queries less so for now, but that is changing: according to Ashmanov & Partners, the Alice block in Yandex commercial results grew from almost zero to about 15% of queries between April and July 2026.

Google says clicks from AI summaries are "higher quality". The ISB and Carnegie Mellon experiment found no difference in click quality, so this claim is still open.

Where the answer gets its sources

Google says there is no special optimization for AI features: the page has to be indexed and eligible to show with a snippet, and normal SEO rules apply. No special files or markup "for AI" are required.

Yandex describes it the same way: search finds relevant pages, the model picks the most useful fragments and builds an answer with links. Only indexed pages are used, and well-structured, informative ones are picked more often.

Independent data is stricter. According to Ashmanov & Partners, an Alice answer relies on about 10 sources on average, and the share of sources outside the top 10 dropped from 40% to 10% between August 2025 and January 2026. For extended answers, more than 90% of pages come from the first page of results.

Short version: the AI answer is built from regular search results. If a page is not in the top 10, it is almost never in the answer.

How to check your site in 10 minutes

Ask your customers' questions. Write down 10 questions clients actually ask you, put them into Google (and Yandex, if you work in Russia). Look at what the answer says and which sites it cites.
Check indexing. In Search Console (and Yandex Webmaster) see which pages are not indexed. A page outside the index will not get into results or into an AI answer.
Use the visibility report if you have it. Yandex Webmaster has a report showing for what share of queries your site is cited in Alice answers, with example queries and competitors. Google has no separate report yet: AI Overviews traffic is counted inside regular "Web" search in Search Console.
Watch the trend. Since AI traffic is not separated anywhere, look at overall clicks per page. Usually informational articles lose traffic first, not service pages.

What to do now

Fix the basics. Important pages indexed, no old URLs returning 404, fast on mobile.
Fight for the top 10 on your main commercial queries. Without this the rest barely works.
Answer directly. Price, timing, process and who does the work should be plain text on service pages, not only inside images. An FAQ block at the bottom helps, including uncomfortable questions.
Add markup. Organization and FAQPage help search engines understand the page. It must match the visible text.
Keep company details consistent. Name, address, phone, services and description should match on the site, maps, directories and social profiles.
Do not block crawlers without a reason. In Yandex you can block the YandexAdditional robot in robots.txt to stay out of Alice answers without affecting regular results. For Google, nosnippet or max-snippet also limit the regular snippet. For a business this is usually a loss.

What I found on my own site

I went through this list on the RM Systems site and found dozens of old URLs returning 404, plus service pages that were crawled but not indexed. Those pages cannot appear in results or in an AI answer, however useful they are.

One small but telling number: over the last four weeks my site got 17 visits directly from ChatGPT. Small, but people already arrive from AI tools, not only from search.

Does "GEO" actually work?

Honest answer: there is no separate magic. A 2026 review of 45 studies found no technique that reliably and durably lifted a site in AI answers across platforms. What works is the usual: the page answers the question precisely and ranks in the top results. So for me, visibility in AI starts with regular SEO.

Indexed pages -> top 10 on main queries -> direct answers on pages -> consistent company details everywhere -> regular visibility checks.

Technical SEO on Astro: what I learned from 400 generated URLs

Roman Mikhailov (Роман Михайлов) — Sun, 04 Oct 2026 20:23:08 +0000

I thought I had found a convenient way to expand my website: generate a service page for each city, change the location in the URL and headings, then let the sitemap introduce everything to search engines.

About 400 URLs later, I had plenty of pages. What I did not have was 400 different reasons for someone to read them.

Some pages appeared in the index, then many started disappearing. Cleaning up the structure took longer than generating it. This is what I learned from that experiment on my own website.

First, look at what you actually built

Before opening another SEO tool, I make a simple inventory: URL, title, description, H1, main section headings, and the purpose of the page.

This gives me a skeleton of the site. Are two pages answering the same question? Does the title match the content? Is there a useful difference between the regional versions?

A separate URL does not answer those questions. Neither does a different city name in the H1.

Astro made the technical part easier

I chose Astro because, in the static setup I used, the build produces HTML containing the page text, headings and links. The main content does not need a client-side application to assemble it after loading.

I can also control titles, descriptions, canonical URLs, internal links and the sitemap. That made the generated structure look tidy.

But static HTML is not a guarantee of indexing or high rankings. Astro cannot make an almost identical page useful just because it renders quickly. I still have to give the page something worth reading.

How my regional pages multiplied

I started with Russian service pages for Cheboksary, Kazan, Moscow and other cities. For a landing-page service, the addresses looked like this:

/cheboksary/catalog/ai-landing
/kazan/catalog/ai-landing
/moscow/catalog/ai-landing

The city changed in the title, headings and a few sentences. Most of the offer stayed the same.

Then I extended the idea to English pages for Almaty, Dubai, Abu Dhabi and other locations. The same template produced another set of regional versions.

At the time, automation looked like the solution. Looking back, it also made it easy to avoid asking whether each page needed to exist.

Discovery was not the same as indexing

My sitemap exposed the generated URLs to search engines. It did not make them accept every page. Google's sitemap documentation makes that distinction: a sitemap helps discovery, but does not guarantee crawling and indexing.

Initially, some pages appeared in search. A few days later, many began dropping out, especially in Yandex.

The similar regional content seemed like a plausible explanation. I cannot prove it was the only cause. Internal links, canonical settings, technical problems, demand and the usefulness of individual pages could also have affected the outcome. This was an observation on my site, not a controlled test or a confirmed search-engine penalty.

Keep the indexing controls separate

robots.txt and noindex

robots.txt controls crawler access. Blocking a URL there does not reliably remove it from Google's search results.

noindex is an indexing instruction in a meta tag or HTTP response header. Google must be able to crawl the page to see it. Blocking the page in robots.txt can prevent that. Google explains this dependency.

Canonical URLs

A canonical identifies the preferred version of duplicate or very similar content. It is not an indexing ban, and Google can choose a different canonical. I should not treat it as a replacement for removing a page. Google's canonicalisation guide describes it as a signal of preference.

What I changed

I stopped automatically expanding into more cities. I kept a general Russian version, an English version, and a focused Cheboksary section. I live in Cheboksary, which gave me a practical reason to develop that direction.

I restricted unwanted regional pages and later deleted them. I also separated the general and local catalogue content rather than changing only the city name.

Old URLs did not disappear from webmaster reports immediately. Cleaning the site and seeing search engines reflect the changes were separate stages.

Now I start with one useful page. Another regional version needs a real difference: relevant examples, conditions, services or information for that audience. I no longer generate a city page just because the template makes it possible.

Roman Mikhailov, author of RM Systems.

Adapted from my original article in Russian.