How much does a business website cost in 2026

A slow, ugly site that does not convert is costing you more than the cheapest quote you are about to accept. Run the math. If your site gets 2,000 visitors a month and converts at 1 percent instead of 3 percent, you are leaving 40 leads on the floor every month. At a 200 dollar close value that is 8,000 dollars a month gone. The website did not cost you 3,000 dollars. It cost you 96,000 dollars a year in pipeline you never saw.

So the right question is not how much does a business website cost. It is what the wrong website is costing you while you shop around. Most operators anchor on the build quote and ignore the leak. The leak is the bigger number, every time. Below are real 2026 ranges, a table by site type, the actual cost driver, and an honest answer about when the cheap option is the correct one.

The real 2026 price ranges

Here are the bands operators pay this year. These are market ranges, not quotes, and they move with scope and who builds it. A solo freelancer, a 6-person boutique, and a 50-person agency will quote the same brief at wildly different numbers. Anchor on the band that matches the job you need done, not on the cheapest name in the inbox.

Those bands track with what research firms report on digital project spend, and the spread is wide for one reason. The cost driver is not the page count. Work on the economics of design, summarized by outlets like Harvard Business Review and consulting research from McKinsey, keeps landing on the same point. Design that drives behavior pays for itself. Design that does not is a sunk cost no matter how cheap it was.

Notice what each band is pricing. The brochure band pays for a templated layout, a few pages, and a contact form. The SMB band pays for original copy, custom sections, and basic analytics. The agency bands pay for discovery, bespoke design, a CMS your team can edit, and integrations with whatever runs your business. Ecommerce sits in its own lane because the cart, the payment flow, and the catalog carry technical weight a content site never touches. When a quote lands far outside the band for the job, ask which line item is doing the work. Usually it is process overhead, not output you can see on the page.

The real cost driver is conversion, not page count

Two sites can have the same 8 pages, the same stock photos, and the same CMS. One closes 1 percent of traffic. One closes 4 percent. The second is worth 4 times the first, and it did not cost 4 times as much to build. You are not buying pages. You are buying conversion rate, and conversion rate is a function of structure, speed, and copy.

This is why a 1,500 dollar template build and a 30,000 dollar custom build can produce the same revenue, or wildly different revenue. The agency that charges 30,000 dollars and ships a beautiful site that loads in 6 seconds and buries the CTA below three scroll lengths has sold you nothing. The price tells you about their cost structure, not your return. Speed alone moves the number. Every extra second of load time shaves measurable conversion off the top, which is why we deploy on infrastructure built for it rather than a shared host that times out under load.

Speed is not a vanity metric. Google publishes its own thresholds for the Core Web Vitals, and the page experience guidance on Google for Developers treats load speed and layout stability as ranking and behavior signals, not nice-to-haves. A site that fails those thresholds loses traffic at the search layer before a visitor sees the CTA, then loses a second slice to slow rendering once they arrive. You pay twice for the same mistake. The cheap host that saved you 40 dollars a month caused both leaks.

Want to see your own number before you spend a dollar? Our revenue leak heatmap shows where traffic is dying on your current site, page by page. The free Closed Loop Audit at /quiz puts a dollar figure on the gap so you can compare it against any quote you receive. Bring that number to every sales call and watch how fast the conversation changes.

A worked example: the quote is the wrong number to compare

Run two quotes through the same business and the comparison flips. Say you have a B2B service company doing 3,000 visitors a month, an average deal worth 1,200 dollars, and a current site closing leads at 1.5 percent. That is 45 leads a month, and at a 25 percent close rate on those leads, roughly 11 customers. Hold the traffic flat and change nothing but the site.

Quote A is a 4,000 dollar template build. It looks clean, loads fine, and keeps conversion at 1.5 percent. Eleven customers a month, same as before. The site earns its keep but moves nothing.

Quote B is a 14,000 dollar conversion-first build that lifts you to 3 percent. That is 90 leads a month, about 22 customers, 11 more than before. At 1,200 dollars a deal that is 13,200 dollars in new monthly revenue, or 158,400 dollars across the year, from a build that cost 10,000 dollars more. The expensive quote is the cheap quote once you price the output instead of the invoice. The mistake is comparing 4,000 against 14,000 as if those were the only numbers in the room. They are the two smallest numbers in the room.

This is why we push the audit before the quote. If you do not know your conversion rate and your deal value, you cannot tell which quote is cheaper. You are guessing with the sticker price as your only data point, and the sticker is the line item least correlated with your return.

The named answer: Website / Vibe Code

Here is what we ship instead of the project-plus-retainer model. Our Website / Vibe Code build delivers a high-conversion site or landing page in 7 days at a fixed price. Not 6 to 8 weeks. Not a 15,000 dollar project followed by a 2,000 dollar monthly retainer you can never cancel.

The stack is boring on purpose, because boring ships fast and stays up. We design in Figma, build in Webflow or Framer, and deploy on Vercel. The term for shipping fast with AI in the loop is vibe coding, and you can read the definition on Wikipedia if it is new to you. The point is not the tooling. The point is that fixed scope plus fixed price removes the open-ended invoice that makes traditional builds so expensive.

The open-ended invoice is the real cost trap. A traditional build quotes you a project number, then bills change requests against it, then hands you a retainer to keep the lights on. Each step looks reasonable in isolation. Stacked over two years, the retainer alone often outruns the original build. Fixed scope kills that drift. You agree on the pages, the sections, and the outcome up front, and the price does not move because someone added a meeting to the calendar.

The 7-day timeline is not a gimmick. It is the same logic we apply to SaaS teams in our breakdown of the 7-day website for SaaS startups. Fixed scope, fixed price, conversion-first structure, live in a week. You see the build, you sign off, it ships. The receipts are in our case studies, not in a slide deck.

What to ask before you sign any quote

Most operators evaluate a web quote on price and portfolio screenshots. Both are weak signals. The portfolio shows you the prettiest work, not the work that converted, and the price tells you about the vendor, not your return. Here are the questions that actually separate a build that pays for itself from one that drains the account.

Is the scope fixed, and what happens when it changes?

Ask for the page list and the section list in writing. Then ask the exact dollar cost of adding one more page after sign-off. A vendor who cannot answer that on the spot is selling you an open meter, and the meter runs in their favor. A fixed-scope shop has the number ready because it lives in the contract.

What load speed do you guarantee, and on what host?

Speed is the cheapest conversion lever and the first thing budget builds sacrifice. Ask for a target against the Core Web Vitals thresholds and ask where the site will be hosted. A shared host with a slow time to first byte will cost you traffic no copy can win back.

Do I own the site, and can I leave?

Some builds lock you into a proprietary platform or a mandatory retainer with no clean exit. Ask whether you own the files, whether you can move hosts, and whether the maintenance plan is optional. If leaving requires a rebuild, the quote is hiding its true price in the lock-in.

How will we measure whether this worked?

If nobody is instrumenting conversion, nobody can tell you whether the spend paid off. Ask what gets tracked, where the numbers land, and who reviews them after launch. A build with no measurement plan is decoration with an invoice attached.

Common mistakes that inflate the real cost

The sticker is rarely where operators lose money on a website. The losses hide in decisions that feel sensible at signing and compound for years. These are the ones we see most.

The first is buying on price alone. A 1,500 dollar site that keeps conversion flat is not a saving if a 12,000 dollar site would have doubled your pipeline. The worked example above is the median outcome for a business that depends on its site for leads.

The second is paying for design that does not change behavior. Award-shelf visuals, animations, and a hero video feel like value because they are visible. None of them move conversion if the structure buries the offer and the CTA. You can spend the whole budget on the part of the site that earns nothing.

The third is ignoring the run cost. The build is a one-time number. Hosting, security, platform fees, and maintenance are forever numbers, and they compound. A cheap build on an expensive platform can cost more across three years than a fair build on a lean stack.

The fourth is skipping instrumentation. A site with no analytics and no lead attribution cannot tell you what is working. You end up redesigning on opinion instead of data, which means paying for a second build to fix the guesswork in the first.

When the cheap site is the right call

Here is the part most agencies will not tell you. A 1,500 dollar brochure site is fine. If you do not need the site to generate pipeline, do not overpay for conversion engineering you will never use. Cost discipline runs both directions, and the cheapest correct answer is still the correct answer.

Who the cheap build fits

If you are a local plumber whose leads come from word of mouth and a Google Business Profile, a clean 5-page brochure site is correct. You need legitimacy, hours, and a phone number that works on mobile. Spending 25,000 dollars there is setting money on fire. Buy the template, point a domain at it, and put the saved capital into something that moves leads. The same logic holds for a tradesperson or a single-location restaurant whose calendar fills from referrals. The site is a business card with a map on it, and a business card does not need a conversion funnel.

Who this is NOT for

The Website / Vibe Code build is not for you if the site is decoration. It is for operators who run paid traffic, track CAC and LTV, and need the site to carry pipeline. If a half-point lift in conversion changes your quarter, you are our buyer. If it does not, buy the brochure and move on with a clear conscience. We would sooner tell you to keep your money than sell you engineering your business model does not reward.

Total cost of ownership, not the sticker

The quote is the smallest number you will pay. Budget for the full picture: the build, plus 1,100 to 5,000 dollars a year for hosting and security, plus a maintenance plan if your content changes often. A WordPress site carries plugin and update overhead that compounds over years. A Shopify store carries platform fees charged against your revenue, which scale up as you grow.

Map it across three years and the picture sharpens. A 4,000 dollar build with a 2,000 dollar monthly retainer is 76,000 dollars by year three, and most of that is the retainer, not the site. A fixed-scope build with optional maintenance on a lean host lands at a fraction of that with the same or better conversion. The retainer is where traditional agency pricing does its quiet damage.

And if you are stacking automations on top, you will spend again. We see this constantly, and we documented it in our piece on how agency retainers bleed mid-market operators dry. Tools like Make and n8n can wire your forms to your CRM cheaply, but only if the site was built to feed them in the first place. A pretty site with no instrumentation is a dead end you pay to maintain.

The honest summary: pick the band that matches the job. A 3,000 dollar site that closes is cheaper than a 30,000 dollar site that does not. Get the number that matters first, then decide. Book it through our contact page when you want to ship something that pays for itself.