What should a business website actually cost?

Ask three people to quote a website and you will get numbers that are ten times apart, with no explanation of why. That is not because two of them are dishonest. It is because "a website" is not a unit of anything — it is a word covering everything from a one-page card to a grocery store that takes 400 orders a week. Here is what actually moves the number, what the ranges really look like, and where the costs land after launch.

Why the same request produces wildly different quotes

Four things are happening at once when quotes diverge.

The scopes are different, even though the request was the same. One quote assumes you supply finished text and photographs. Another includes writing every page and a half-day of photography. One includes a year of hosting; another leaves you to arrange it. Neither said so, so they look comparable and are not.

The overhead behind the quote is different. A solo developer working from home and a twelve-person agency with a project manager, an account manager, and an office are pricing the same deliverable through very different cost structures. Bigger is not automatically better, and cheaper is not automatically worse — but you are paying for whichever one you hire.

Vagueness gets priced as risk. If you send a two-line brief, a careful quoter has to assume the worst case, because the alternative is eating the difference. The single cheapest thing you can do to lower a quote is to describe the project precisely: how many pages, whether it sells, who writes the words, whether photos exist.

Some quotes are for a template, and some are for a design. That distinction deserves its own section.

What actually drives the cost

Unique page designs, not the number of pages

Owners count pages. Builders count templates — distinct layouts that have to be designed, built, and made to work on a phone. A site with a home page, an about page, a services page, a contact page, and eight individual service pages is twelve pages but roughly five templates, because those eight service pages share one layout. Adding a thirteenth service page after launch costs almost nothing. Adding a thirteenth layout costs real money.

This is the single most useful thing to understand before you read a quote. If a quote is priced "per page" with no definition of a page, you are being quoted by someone who has not thought about it, and the number will move later.

Whether it takes money

A site that sells is a different category of work, not a bigger version of the same one. It needs a product catalog, tax handling, shipping or delivery logic, a payment connection, order emails, a refund path, and a checkout that survives contact with a real customer on a three-year-old phone. It also needs an owner who can run it: someone has to mark items out of stock at 6 a.m.

A brochure site that goes down for an hour is an inconvenience. A store that goes down for an hour on erev yom tov is lost revenue and angry phone calls, so stores carry more testing, more monitoring, and more support — all of which is cost.

Custom functionality

Everything a site does beyond "show pages and take a normal order" is a separate build. In this line of work the common ones are specific and familiar: delivery zones drawn by zip code with different minimums and fees, a cutoff that stops orders before Shabbos and reopens after, meat priced by the pound where the final charge follows actual weight, pickup windows before yom tov with a cap per slot, a Yiddish or Hebrew version of the site that reads right to left, a wholesale price list only logged-in customers see.

Each of these is a few hours to a few days. None of them are exotic. But each one is the difference between configuring software and writing software, and that is where hours accumulate.

Content and photography — the line owners forget

The most common reason a project runs late and over budget is that nobody decided who was writing the words. A quote that does not name a content owner is not a complete quote.

Text is the cheaper half. Professional copywriting is commonly quoted somewhere in the low hundreds of dollars per page, less for short pages, and many owners write acceptable text themselves if someone gives them a structure and edits it. Photography is where the number gets serious, and it is the line that separates a store that sells from a store that does not. Product photography is commonly priced per image, and a real catalog is hundreds of images — a grocery with 800 items is a photography project with a website attached, not the other way around. Stock photos will not save you here, because nobody stocks a photo of your cholent.

The honest way to handle this is in stages: launch with your top sellers shot properly and the long tail carried by plain, consistent catalog shots, then improve. Photographing food and seforim for a heimish store covers what "shot properly" means and what you can genuinely do yourself with a phone.

Revisions and how decisions get made

The cost driver most under your control is your own approval process. Two rounds of consolidated feedback from one decision maker is a normal, cheap project. Six rounds of contradictory feedback from an owner, a brother-in-law, and a customer who "does graphics" is the same project at double the hours.

A good quote states a number of revision rounds and defines what a round is: one consolidated list, at one stage, from one person. Beware in both directions here. No stated limit means the price has padding in it or the project has no end. "Unlimited revisions" means the same thing, more cheerfully.

A template filled in versus a site that was designed

Most of the price gap between a $900 site and a $6,000 site is here, and it is worth being blunt about it.

A template filled in means someone bought or picked a pre-made layout, replaced the placeholder text with your text and the placeholder photos with your photos, changed the colors, and launched. This is legitimate work. It takes skill to do well, it is fast, and for many small businesses the result is perfectly good. What you are not getting is a layout built around what your business needs to say. The template decided that the home page has three feature boxes and a testimonial band, so your home page has three feature boxes and a testimonial band, whether or not that is how a customer decides to trust a caterer.

A designed site starts from your business: what a customer needs to know, in what order, to pick up the phone or place the order. The structure is chosen, then drawn, then built. It costs more because there are more hours in it and the hours require judgment.

The test for which one you are being quoted is simple: ask to see the site the design is based on. If there is a theme or template underneath, a straight answer is a good sign, not a bad one. If the answer is evasive, you have learned something.

What the ranges actually look like

These are the ranges you will encounter in the American small-business market. They are observations, not our prices, and they move over time. What matters is the shape: cost tracks scope, not luck.

Timeframes follow the same shape. A small brochure site is a two-to-six-week project. A store is more often six to twelve weeks. In both cases the bottleneck is almost never the code — it is waiting for text, photos, and a decision.

What the cheap end genuinely gets you

Under about $1,500, you should expect a competent, mobile-friendly, template-based site of a handful of pages, using content you provide, that loads fast and gives people your address, your hours, and a way to reach you. That is a real product and it is not a lesser choice. For a brand-new business, it is frequently the correct choice, because you do not yet know what your website needs to do. Six months of taking phone calls will teach you more about that than any planning meeting.

What you should not expect at that price is strategy, original photography, written copy, ongoing support, or unusual functionality. If a quote at that level promises all of those, the promise is the problem.

What the expensive end is buying

Above roughly $10,000, you are mostly not buying prettier. You are buying thinking, and you are buying risk transfer. Specifically: someone who works out what the site has to accomplish before drawing anything; original photography and written copy; functionality that matches how your business actually runs rather than how the software assumed; performance work, because a store that loads slowly on a phone loses orders; accessibility done deliberately rather than hoped for; testing on real devices; and a named person who answers the phone when checkout breaks. That last item is worth more than most owners expect until the first time they need it.

The ongoing costs, itemized honestly

This is where owners get surprised, because the build price gets negotiated carefully and the monthly costs get waved through. Over five years, the monthly column is frequently larger than the build.

Two ongoing items deserve extra attention.

Payment processing is a percentage, so it scales with success. At $8,000 a month in online sales, roughly 3% is about $240 a month — more than hosting, email, and plugins combined. It is unavoidable and it is not a rip-off, but it belongs in your arithmetic when you compare platforms. Watch for platform fees stacked on top of processing: some hosted platforms add an extra percentage if you use an outside payment provider instead of their own.

"Maintenance" is the vaguest word in this industry. Ask exactly what it covers. A reasonable definition names the parts: software and security updates, off-site backups with a tested restore, uptime monitoring, an SSL certificate that renews itself, and a stated number of small content changes per month with a response time for genuine emergencies. If the answer is "we keep an eye on things," you are paying a subscription for goodwill.

Wix, Squarespace, and Shopify: a fair account

These platforms are good software, and pretending otherwise would be dishonest. They are the right answer more often than most developers admit.

They are genuinely right when: the business is new and unproven and needs to exist online this week; the site is a handful of pages with no unusual logic; the owner wants to change text and prices constantly and does not want to email anyone to do it; or the alternative is a badly built custom site, because a $600 custom site is worse in every respect than a $20-a-month Squarespace. Shopify in particular is excellent at the thing it does: if you sell a normal catalog of physical goods to consumers, with normal shipping, Shopify will do it well and you should think hard before building something else.

The costs are real and worth doing on paper. Website builder plans generally run in the region of $16–$40 a month for a business tier, e-commerce tiers higher; Shopify's entry plan has sat around $39 a month, with apps stacked on top. Check current pricing before deciding — these change. Then multiply. A store on $39 plus $60 of apps is roughly $100 a month, about $6,000 over five years, and the number rises as you add apps. A built site on $25-a-month hosting is about $1,500 over the same five years. That does not automatically favor building — the built site had a larger up-front cost, and your own time has value — but it is the comparison that matters, and almost nobody runs it.

They stop being right when you need something they will not do. This is where heimish businesses hit the wall earlier than most, because the requirements are specific: delivery zones by zip with per-zone minimums, an ordering pause for Shabbos and yom tov in the right time zone, brisket priced by actual weight, a properly typeset Yiddish or Hebrew right-to-left version. On a hosted platform each of these becomes an app subscription, then a workaround, then a compromise you explain to customers. When the workaround list gets long enough that the site is fighting you every week, you have already paid for a build.

And there is the exit. If you leave Wix or Squarespace, you take your domain, your text, and your images. You do not take the site. Shopify is better — products, customers, and orders export as spreadsheet files, and theme files are yours — but the apps and their data do not come with you. Factor a rebuild into any move.

Red flags in a quote

• No content plan. The quote never says who writes the words or supplies the photos. This is the number one cause of a project stalling for four months.
• No revision limit — or unlimited revisions. Both mean the same thing: nobody has defined "done."
• No mention of who owns the domain and the code. Covered below. This is the one that actually hurts.
• Vague ongoing fees. "$99/month hosting and maintenance" with no itemization, and no answer to what happens to the site if you stop paying.
• Priced per page with no definition of a page.
• "SEO included." Included how? Written page titles and descriptions, a sitemap, and a Google Business Profile is a real answer. "SEO" alone is not.
• Guaranteed rankings. Nobody controls Google's results. A guarantee of position one is either meaningless small print or a sign the person will do something you would not want done.
• 100% up front. A deposit is normal and fair — a third to a half. All of it, before anything exists, is not.
• Nothing in writing. A one-page scope beats a handshake, for both sides.
• No exit terms. If you part ways in year two, what do you get and how fast?

The point that matters more than the price

This goes wrong more often than you would think, usually with no bad intent — a developer registers the domain on a client's behalf, twelve years pass, and the developer has moved to Israel and does not answer email. Recovering a domain from an unresponsive registrant is slow, sometimes expensive, and occasionally impossible. Since European privacy rules took effect in 2018, public domain lookups generally no longer show who the registrant is, so you often cannot even find out from outside. And be aware of timing: under ICANN's transfer rules, a domain is generally locked against moving to another registrar for 60 days after registration and after certain changes of registrant — so "we'll sort it out later" can mean two months of waiting at exactly the wrong moment.

The same applies to every account attached to the business: your Google Business Profile, your payment processor, your email, your analytics, your social pages. A developer should have access. A developer should not have sole access.

On the code itself, the default rule surprises people. Under U.S. copyright law, an independent contractor generally owns the copyright in what they create unless there is a signed agreement transferring it. The "work made for hire" doctrine covers commissioned work only in nine specific categories listed in the statute, and a website is not obviously one of them; the Supreme Court's 1989 decision in Community for Creative Non-Violence v. Reid settled that hiring someone does not by itself make you the author. So a clause that assigns ownership to you on final payment, in writing, is worth asking for.

Two honest caveats. First, a broad perpetual license is often good enough in practice, and some developers reasonably keep rights to reusable internal components — that is not a scam, it is how anyone who has built more than one site works. Second, third-party pieces are licensed, not owned: themes, plugins, apps, stock photos, and fonts come with their own terms, and those licenses need to be in your name too, not on the developer's account. Fonts in particular trip people up — a font license for print does not automatically permit use on a website.

What to ask before you sign

1. How many unique page designs is this, and which pages share a layout?
2. Who writes the text? Who supplies the photographs? What happens to the schedule if I am late with either?
3. How many rounds of revisions, at which stages, and what counts as one round?
4. In whose name and whose account is the domain registered? Will I have the login?
5. When it is paid for, do I own the site in writing? What is licensed rather than owned?
6. Itemize the ongoing costs. What is the total monthly, in year one and in year three?
7. What happens to the site if I stop paying the monthly?
8. Is this a template or an original design? If a template, which one?
9. Who do I call when checkout breaks on a Thursday afternoon, and how fast do you answer?
10. Will the site be built to WCAG 2.1 Level AA? That standard is what accessibility claims are measured against.
11. Can I edit my own text and prices, and will you show me how — ideally with a recording I can rewatch?
12. If we part ways, what do I receive, and how long does it take?

A developer who answers all twelve without flinching is telling you as much about the working relationship as about the price. If you want a number for your own project rather than a range, describe what you need and we will quote it properly — including the parts most quotes leave out.

The short version

• Cost tracks unique page designs, whether it sells, custom logic, content, and revisions — not the number of pages. Quotes differ mostly because the scopes differ silently.
• Roughly: $800–$3,000 for a template filled in, $3,000–$10,000 for a designed site, $5,000–$25,000 for a real store. The cheap end is a legitimate, correct choice for a brand-new business.
• Content and photography are the hidden line item. A grocery with 800 products is a photography project with a website attached.
• Do the five-year math on ongoing costs. Domain, hosting, email, app subscriptions, roughly 3% on card sales, and maintenance often add up to more than the build.
• Wix, Squarespace, and Shopify are genuinely right for many businesses — until the workaround list gets long, or the monthly quietly passes what a built site would have cost.
• Own your domain, in your name, in your account. Get code ownership in writing. That single point outranks every dollar figure on this page.

Questions about a quote you have received — yours or somebody else's — are welcome. Call 845-587-0531 or email info@bhwebsolutions.com.