How Much Should a Website Cost in 2026?

How much should a website cost? The honest answer spans from under $1,000 a year to well past $250,000, and either end can be the correct number for the right business. The difference is not how polished the result looks — it is who does the work, who carries the risk, and what support looks like after launch. This guide splits 2026 pricing into four clear bands so you can place your project before you open a single quote.

We scope and build websites for owners every week: a landscaping company that needs booked jobs, a home-goods store that needs a reliable checkout, a coffee roaster that needs wholesale inquiries. Whatever the business, the same three questions decide whether the money was well spent — what does each band actually buy, where do budgets really blow up, and how do you compare two quotes that look nothing alike? We answer all three below, with line items, hidden costs and payback math in the open. Ranges marked illustrative are modelled from typical agency engagements rather than a single survey.

Here is the short version before we dig into the bands.

Key Highlights

• Essentially every 2026 website quote falls into one of four bands: $0–$1,000 per year template builds, $2,000–$15,000 freelancer builds, $15,000–$75,000 agency builds and $75,000–$250,000+ custom or enterprise work (illustrative ranges from typical engagements).
• What you pay extra for between bands is mostly risk transfer and seniority — discovery before design, named accountability, and someone to call when launch week goes sideways.
• In an honest build budget, development runs 45–55% of the total, design and discovery together about a third, and QA roughly 8–10% (illustrative line items below).
• Ongoing costs are the number buyers forget: plan 15–20% of the build cost per year for maintenance plus $50–$500 per month for hosting and licenses.
• The three-year tail is real: a $45,000 agency build typically costs about $75,000 over three years all-in (illustrative model).
• Quality pays measurably: Deloitte's Milliseconds Make Millions study found a 0.1-second speed improvement lifted retail conversions by about 8%.
• Credibility is visual: Stanford's web credibility research found about 75% of users judge a company's trustworthiness on the design of its website.
• Same-spec quotes can spread 2–3x between vendor tiers — line items and written assumptions are the only fair way to compare them.

Where Website Budgets Sit in 2026: The Four Cost Bands

Every serious quote lands in one of four bands, and the bands are defined by who does the work and who carries the risk — not by industry, platform or how the vendor describes itself. Think of a band as a promise about process: what gets discovered before design begins, who signs off on quality, and who answers the phone when launch week goes sideways. Place yourself in a band first, then compare offers inside it. Comparing across bands is the most common way buyers talk themselves into the wrong purchase.

The ranges are wide because scope, not vendor greed, drives most of the spread. A five-page brochure site and a 3,000-SKU store with inventory sync are both "websites" the way a canoe and a cargo ship are both boats. The bands also hold their shape across markets: a $45,000 agency project in a major US metro buys roughly the same scope as one in a smaller city, with hourly rates doing most of the adjusting.

Key takeaway: Decide the band before the budget. If your site has to generate revenue — bookings, orders, qualified leads — the honest floor is Band 3, and everything cheaper is a trade you should make deliberately, not by accident.

Band 1: DIY and Template Platforms — Under $1,000 a Year

Band 1 is the world of hosted site builders and theme marketplaces, where the software is rented for roughly $20–$100 per month and the labor is yours. Subscription pricing is public and stable, so the real cost is time: plan 40–80 hours of owner effort for a first pass that looks professional, which is what we typically see clients spend before they call for help. Add a domain, a premium theme and a few apps, and a realistic first-year all-in lands between $300 and $1,200 (illustrative).

The hidden price is the ceiling. Templates flatten differences — your landscaping company gets the same skeleton as the competitor across town — and advanced needs such as custom booking logic, member areas or complex catalogs either strain the platform or arrive as paid add-ons with monthly fees. For a services business whose conversion path is a phone call, that ceiling rarely matters in year one.

• Fits: you need credibility and contact; content is simple; you can invest your own evenings.
• Strains: custom quoting or booking flows, real inventory, SEO ambitions beyond local search.
• Watch: renewal pricing jumps after promo years, and "free" plans often carry platform branding.

Our landscaping example is genuinely fine here: services page, project gallery, reviews, a contact form, done in a weekend. The mistake is not starting in Band 1 — it is staying there after the site's job changes from "exist credibly" to "generate revenue."

Key takeaway: Band 1 is a legitimate starting point, not a failure — budget your hours honestly, and know in advance which two features would force you into Band 2 or 3.

Band 2: Freelancers and Small Studios — $2,000 to $15,000

Band 2 buys custom work from one or two accountable humans, usually on proven rails such as WordPress or a lightweight headless setup. Typical 2026 US freelance rates run $45–$85 per hour, with small studios a step above; multiply by a realistic 60–150 hours and you get the band's range (illustrative, charted below).

The economics are attractive and so is the speed — a single decision-maker means fast iteration and no committee. The trade is concentration of risk: one person's availability, workload or illness becomes your timeline. The strongest Band 2 outcomes share three traits: a written scope, a design direction the owner signs off before build starts, and a handover package that documents hosting, passwords and how to make changes.

What the range buys in practice: a custom-designed site of 5–15 pages, mobile-first, with a contact flow, baseline SEO setup and perhaps a blog or an embedded booking widget. For the coffee roaster in our examples, Band 2 covers a brand site with product pages that feed a wholesale inquiry form. It does not cover subscription logic or ERP sync — and a good freelancer will say so up front.

Key takeaway: Band 2 is the best value per dollar in the market — if you supply clear scope and the vendor supplies documentation. Buy the handover package even when it costs extra; it is the cheapest insurance in web.

Band 3: Professional Agency Builds — $15,000 to $75,000

Band 3 is where a site stops being a brochure and starts being infrastructure: discovery workshops, user flows, custom design, a real CMS, integrations with your operations, performance and accessibility budgets, QA across devices, and a launch plan. You are paying for a team — strategy, design, engineering, QA, project management — and for a process that survives contact with complexity.

Two patterns in the table deserve attention. Development dominates at roughly half of both totals, which is why scope control beats rate negotiation every time. Discovery and design together run about a third, which is why quotes that quietly skip discovery usually cost more in rework than they saved up front. When the build includes unusual machinery — a headless storefront, a booking engine, a product configurator — the range bends upward; our headless ecommerce storefront case study and our booking engine for a retreat business show what that machinery looks like in delivery.

For the home-goods store in our examples, Band 3 is the realistic floor once orders matter: inventory accuracy, checkout reliability, tax and shipping logic, and page speed all become revenue questions rather than cosmetics. A store doing meaningful volume cannot afford Band 2 checkout behavior during a holiday sale.

Key takeaway: Agency pricing is really process pricing. Demand the phase list — discovery, UX, design, build, QA, launch — and you will instantly understand the number, and spot the quotes that quietly skip phases.

Band 4: Custom and Enterprise — $75,000 to $250,000+

Band 4 starts when the website effectively is the business: customer portals, marketplaces, membership platforms, multi-location rollouts, and systems that must integrate with ERP, CRM or payments infrastructure. At this level budgets are shaped by engineering depth, integration count and compliance requirements — not by page count. A regional property portal lives in this band by definition; we documented the shape of that work in our property portal platform case study.

Three things change structurally in Band 4. First, discovery becomes its own paid phase — often four to eight weeks of defining roles, data flows and edge cases before pixels move. Second, non-functional requirements enter the contract: uptime targets, load thresholds, accessibility standards (WCAG 2.2 is the common reference) and security review. Third, governance appears on both sides — a product owner on your staff, change control, and a regular steering rhythm.

Not every business needs Band 4, and a good vendor will tell you when Band 3 plus a roadmap beats a big-bang build. The honest test is simple: if the system touches money flow, regulated data or thousands of concurrent users, you are in Band 4 territory; if it mainly presents and converts, you are probably not.

Key takeaway: In Band 4 you are buying engineering insurance as much as features — discovery, QA share and compliance are not line items to shave; they are the reason the band exists.

Hidden Costs: Where Budgets Actually Blow Up

Most budget blowups are not price changes — they are costs that never appeared in the quote. The recurring offenders, with the sizes we see in mid-market US practice (illustrative), are easy to list and easy to forget:

Two of these deserve a hard look. Accessibility is the sleeper: thousands of ADA-related web lawsuits are filed in the US each year (annual legal trackings such as UsableNet's consistently count in the thousands), and retrofitting costs multiples of building it in. Migration is the classic launch-week ambush — old URLs, indexed pages and accumulated SEO equity evaporate without a redirect plan, and recovery is far slower than prevention, as our how long SEO takes guide explains.

Put maintenance at 15–20% of build cost per year, add hosting and licenses, and the tail emerges: the $45,000 services site from the last section costs roughly $75,000 over three years. Budget the tail at decision time, not in year two when the invoice arrives. A build you cannot afford to maintain is a build you cannot afford.

Key takeaway: Ask every bidder what is excluded — the exclusions list predicts blowups better than the price does. Then judge affordability on the three-year total, not the invoice.

How to Compare Quotes Fairly — and Frame the ROI

Quotes for the same stated project routinely differ by 2–3x because they answer different hidden questions: one includes discovery, another assumes you supply content, a third prices maintenance and a fourth ignores it. The fix is to normalize before comparing. Send every bidder the same one-page brief and require the same quote structure.

Then frame ROI the way a CFO would. A rebuild earns money through conversion, speed and trust, and each has a measurable value. In the illustrative model below, a 10% conversion-rate lift on a store with an $80 average order value and a 2% baseline conversion rate adds revenue that scales directly with traffic.

Read that chart against the three-year cost line from the hidden-costs section and the payback logic appears. At 25,000 monthly visits, a $45,000 rebuild that produces a 10% conversion lift pays for itself inside a year; at 5,000 visits the same lift needs several years, and the smarter first spend is SEO and content, not design. That is what fair comparison ultimately means: judge each quote against the revenue mechanics of your specific site, not against the other bids.

Key takeaway: Fair comparison is mechanical — same brief, same quote structure, same three-year lens. Then rank bids against your traffic and conversion math, and let the vendor who understands that math win.

What Your Business Should Do

Budgets are only useful if they change what you do on Monday. Here is the 90-day plan we hand clients, in three sprints:

Days 1–30: Define before you quote.

• Write a one-page brief: goal, conversion action, must-have features, integrations, content owner.
• Score your own readiness — do you have copy, photography, brand assets, and access to domains and analytics?
• Set a three-year budget ceiling, including 15–20% annual maintenance, before hearing any price.

Days 31–60: Collect and normalize quotes.

• Send the same brief to three to five vendors across at most two bands; require the line-item structure from the last section.
• Check references on each vendor's two most recent launches of similar size, not the showcase piece.
• Score quotes on three-year cost, team and assumptions — then interview the shortlist on process, not price.

Days 61–90: Decide, contract, plan the tail.

• Contract with milestone payments tied to phases, IP assigned to you, and a named change-order rate.
• Put QA share, accessibility conformance and a redirect plan into the statement of work.
• Book the 90-day post-launch review now: speed, rankings, conversions and the next scoped iteration.

The businesses that overpay for websites almost always skipped the brief; the businesses that under-spend almost always skipped the tail. Both mistakes are free to avoid, starting this week.

The Bottom Line

There is no single right price for a website in 2026 — there is the right band for your goal, honest line items inside it, and a three-year number you can defend internally. Get those three right and the quoting process turns from guesswork into procurement. If you want a second pair of eyes on your brief or the quotes already on your desk, that is exactly what we do all day: Book a free strategy call and we will pressure-test your scope together — or send us a message with your questions and we will respond within one business day.