How Much Does a Website Cost? A B2B Software Buyer's Breakdown

A website's price is not one number, and the published guides put the range anywhere from free to well past six figures. That spread is real rather than marketing: the phrase "a website" describes a free template and a custom-built marketing platform equally well. This guide is for a founder or first marketing hire at a seed to Series A software company who has to commission a site and wants a defensible budget before the first sales call. Below: the published ranges with their sources attached, the reason two quotes for one brief can land an order of magnitude apart, and a method for working out your own number in about an hour.

The short version

• Page count is the wrong unit. A 40-page site built from six templates is less work than a 12-page site with twelve bespoke layouts, and the quote follows the work.
• The published ranges are wide but real. WebFX puts agency design at $3,000 to $30,000 and up. Forbes Advisor puts a do-it-yourself build at $0 to $450 and professional design at $1,500 and up.
• We read two of the page-one guides in full. WebFX prices design by the page, at $1,000 to $10,000 for up to 250 pages, and Forbes Advisor scopes its headline averages to "most small businesses," which is a different buyer than a software company.
• Route by constraint, not by budget. The question is what you must still be able to do in eighteen months without a vendor in the way.
• The ongoing bill is two different things under one label: keeping the site working, and continuing to build it. Only the second one is expensive.

What does a website cost in 2026?

Across the sources we reviewed, anywhere from nothing to well past $100,000, and the build route explains most of that spread. Forbes Advisor opens its 2026 guide by saying website costs vary "from free to hundreds of thousands of dollars," and WebFX puts the whole-project range with an agency or freelancer at $1,000 to $145,000. The table collects the published figures from both guides. They are those publishers' numbers rather than ours, and this article does not publish a GrowthHasten build price.

Build route	Published range	Source
Do it yourself on a website builder	$0 to $450 for the initial build	Forbes Advisor
Website builder subscription	$17 to $500 per month	WebFX
Named builder plans	Wix $17 to $159 per month, Squarespace $25 to $139 per month, WordPress free to $70 per month plus hosting	Forbes Advisor
Freelance designer	$500 to $10,000 and up	WebFX
Professional design, provider unspecified	$1,500 and up	Forbes Advisor
Agency design	$3,000 to $30,000 and up	WebFX
Whole project with an agency or freelancer	$1,000 to $145,000	WebFX

Two details matter more than the numbers themselves. WebFX's website cost guide prices "individual page design" at $1,000 to $10,000 for up to 250 pages, which is pricing literally by the page. Forbes Advisor's 2026 guide, audited in January 2026, introduces its headline averages with the phrase "for most small businesses, an informational non-e-commerce website," which tells you precisely who the averages were assembled for.

Neither guide is wrong. Both answer a real question carefully. But WebFX prices the unit of work as a page, Forbes Advisor prices a small business's informational site, and a B2B software marketing site is neither of those things.

Why do two quotes for the same brief differ by an order of magnitude?

Because the brief is almost never the same brief. "A 20-page marketing site" can describe six reusable templates or twenty bespoke layouts, and those are different projects wearing the same sentence. The drivers, roughly in the order they move the number:

• Build route: a subscription builder, a freelancer, an agency and an in-house team are four different cost structures before anyone looks at your requirements.
• Unique templates: the count of distinct page layouts the site needs, which is the subject of the next two sections.
• How much of that is bespoke: WebFX prices stock templates at $20 to $300 and up each, and Forbes Advisor puts template costs from free to well over $1,000. Original layout design sits on a different scale from either.
• Custom functionality: WebFX puts database integration at $2,000 to $25,000 and a CMS build at $2,000 to over $100,000 for larger enterprise sites.
• Integrations: every system the site has to talk to adds scoping, build and a place for the project to stall while someone finds an API key.
• Content production: words, diagrams and photography. Check whether the quote includes them, because a site cannot launch without them whether or not they are priced. Forbes Advisor puts photography between $50 for a single image and $10,000 for a full day of product shooting.
• Approval surface: the number of people who can send a design back. Ask how many revision rounds the quote covers and what a further round costs, because that is where an agreed scope turns into a change request.

For how far identical-sounding work can be priced apart, Forbes Advisor's Liz Masoner writes that she has seen agencies charge well over $1,500 to add text and change out stock images on a template, and has also seen agencies do deep customization and mobile optimization for far less. Same deliverable name. Opposite amount of work.

Why count templates instead of pages?

Because the template is the unit of work and the page is not. A new template costs design, front-end build, responsive states, a content model in the CMS, and QA across browsers. A new page on a template you already own costs content entry. Those are not the same expense, and only one of them scales with the number of URLs.

Run the two extremes. A 40-page site built from six templates, say a home page, a product page, a solution page, a blog index, a blog post and a legal page, is six design-and-build jobs plus thirty-four content jobs. A 12-page site where every page has its own layout is twelve design-and-build jobs. The second site is smaller and costs more, which is exactly the result a page-count quote cannot express.

This also explains a quote you may have already received and found confusing. An agency that prices by the page will quote the 40-page site higher. An agency that prices by the layout will quote it lower. Both are being honest about their own cost structure.

When this reframe does not help: if the site genuinely is a set of one-off campaign landing pages with no shared shell, then pages and templates are the same number and page count is the honest unit. The reframe earns its keep the moment a site has repetition in it, which a blog index, a blog post layout and a set of solution pages already supply.

How do you count the templates you actually need?

In about an hour, with a document and no tools. The output is a number you can put in front of a vendor, and it changes what the first call is about.

1. List page types, not pages: write down the kinds of page the site needs at launch. "Customer story" is one line, not nine.
2. Mark each one bespoke or variant: a variant reuses an existing layout with different content or a different component order. Bespoke means it needs its own design. Be strict, because almost everything wants to be bespoke on the first pass.
3. Count the bespoke ones separately: this is the number that drives the quote. The variant count drives the content bill instead, which is a different budget line.
4. Sanity-check against a site you admire: open their /sitemap.xml or walk their navigation, and sort what you find into the same page types. The gap between their URL count and their layout count is the thing you are trying to see.
5. Cut before you quote, not after: every bespoke layout you remove at this stage is removed for free. Every one you remove after a quote is a change request.

A plausible launch set for a seed-stage software product looks like this: home, product, one solution or use-case template, pricing, customer story, blog index, blog post, docs or legal, and contact. That is nine page types, of which perhaps three genuinely need original design. If a vendor's quote treats all nine as bespoke, that gap is the conversation to have.

What does a B2B software marketing site need that a brochure site does not?

Five things, and each one is a constraint on the build route rather than a feature to add later.

1. Control of the URL structure: you will eventually want /integrations/, /compare/ and /docs/ to be real paths you own, with the ability to add hundreds of pages under one of them without a platform deciding the shape.
2. HTML that does not depend on client-side rendering: Google's JavaScript SEO basics documentation separates crawling, rendering and indexing into three phases, warns that a page can wait in the render queue for "a few seconds, but it can take longer than that," and recommends pre-rendering outright on the grounds that "not all bots can run JavaScript." That last clause now covers AI answer engines as well as search crawlers.
3. A content layer the marketing team can run without engineering: if publishing a page requires a pull request, the content program runs at engineering's cadence. That is a second-year operating cost rather than a build cost, so look for it outside the build quote.
4. A path to the product or the CRM: gated content, trial signup, in-app links and lead routing all assume the site can talk to something else.
5. A performance budget somebody enforces: Google's Core Web Vitals reference on web.dev sets the good thresholds at an LCP within 2.5 seconds, an INP of 200 milliseconds or less, and a CLS of 0.1 or less, each measured at the 75th percentile of page loads. A budget is only real if someone can fail a build against it.

The architecture behind points one, two and five is set out at length in what makes a website search-ready. Treat all five as a checklist to run against a vendor's proposal, not as a wish list to hand over.

Which build route is right for a software company?

Our recommendation is to choose by which of those five constraints already binds, and to treat price as the second question. A route that satisfies every constraint you have today and none of the ones you will have next year is the expensive option, because it commits you to paying for two builds instead of one.

Route	Choose it when	Where it stops
Website builder	None of the five constraints binds yet. You need a credible site quickly, no integrations, and marketing can publish on its own inside the builder.	URL control and rendering. Custom paths and server-rendered output are the platform's decision, not yours.
Freelance designer or developer	One or two constraints bind, the scope is a small template set, and someone on your side can specify and review the work.	Continuity. A content layer and a performance budget both need an owner after launch, and a project engagement does not supply one.
Agency	Three or more constraints bind at once, or the build has to land alongside a launch and needs design, engineering and content in one plan.	Cost and pace of change. Every post-launch page goes through a queue you do not control unless the handover is written into the contract.
In-house team	The site is a product surface, integration with the application is continuous, and you already employ front-end engineers who are not fully committed to the product.	Opportunity cost. Engineering time on the marketing site is engineering time not on the product, which is the most expensive currency a seed-stage company has.

Two of those constraints are really platform decisions, so make them before the route. Our comparison of which CMS you build on covers the trade-off between a coupled platform and an editable one, and the case for running a headless content layer sets out what the most flexible branch actually asks of you in return.

When is a website builder the right answer?

When none of the five constraints binds yet, which is a genuine position for a company that has not shipped its product. A builder gets a credible site in front of buyers in days, and Forbes Advisor's figures put that at $0 to $450 for a do-it-yourself build with hosting and apps at $15 to $150 per month, against a professional design starting at $1,500 and up.

The honest limit is that the cheap route stops being cheap the moment a constraint appears, because the fix is a migration rather than an upgrade. The awkward part is timing. The events that make a builder stop fitting, a funding round, a pricing change, a repositioning, are the same events that leave you least able to spend six weeks on a website.

The test to apply: if you can name a constraint that will bind within twelve months, price the route that satisfies it now and compare the two totals honestly. If you cannot name one, the builder is the correct answer, and spending more is a preference rather than a requirement.

What are the ongoing costs of running a website?

Small and predictable, until they are not. The published figures for the recurring items are narrow enough to budget from directly.

Item	Published range	Source
Hosting	$2 to $1,000 per month	WebFX
Hosting, entry level	$5 to $13 per month traditional, $15 to $20 per month for builder packages	Forbes Advisor
Domain	$1 to $12 per year	WebFX
Domain	$10 to $35 per year	Forbes Advisor
SSL certificate	$0 to $1,500 per year	WebFX
SSL certificate	$8 to $270 per year	Forbes Advisor
Business email	$1 to $27 per user per month	WebFX
Maintenance	$5 to $5,000 per month, or $60 to $60,000 per year	WebFX

That maintenance range spans a factor of a thousand, and it is worth explaining why rather than quoting it and moving on. One clue sits inside the source itself: WebFX's guide carries two different annual maintenance figures, $60 to $60,000 per year in its body copy and $3,600 to $50,000 per year in its key-takeaways summary and its cost cheat sheet. "Maintenance" is doing duty for two unrelated jobs.

Keeping the site working: dependency and platform updates, security patching, backups, uptime, fixing what breaks. This is cheap and fairly predictable, and managed hosting absorbs most of it. Forbes Advisor notes managed WordPress plans starting around $6 per month and observes that a dedicated website management service can instead cost several hundred dollars every month.

Continuing to build: new pages, new templates, new components, experiments, a redesign of one section. This is development work priced as a retainer, and it is where the top of the range comes from. It is not maintenance in any useful sense, and calling it maintenance is how WebFX's $5 figure and its $5,000 figure end up inside one range.

Decide which of the two you are buying before you read a maintenance quote, and budget your search program as its own line. Our breakdown of what ongoing SEO costs covers that side separately, because a static site and a site that gains pages every month have different requirements there.

What if you already have a website?

Then you are pricing a different purchase. A rebuild carries risks a first build does not: existing rankings and URLs to preserve, redirect mapping, content that has to survive the move, and a live site that cannot go down while it happens. Those risks change the scope rather than the design, so the figures above will under-describe it. We cover what a rebuild costs instead in its own guide.

What should a website quote contain?

An itemization you can argue with. A single line reading "website design and development" gives you nothing to push back on. Ask for the breakdown below, and expect a vendor who has scoped properly to have it already.

• Template list: every page type, marked bespoke or variant, with the count of each.
• Platform and hosting: what it is built on, who hosts it, who owns the account, and what the monthly figure is.
• Content: who writes it, who sources images, and what happens to the timeline if that slips.
• Integrations: named systems, not "CRM integration" as a category.
• Performance and accessibility targets: the thresholds the build has to meet before it is accepted.
• Handover: what your team can change after launch without calling anyone.
• Post-launch: the separated keep-it-working and keep-building figures, priced apart.

Three questions worth asking on the scoping call, in this order.

• How many unique templates is this quote built on: a vendor who cannot answer has not scoped it.
• What does adding ten pages do to the price, and what does adding one new layout do: the difference between those two answers tells you their real cost model.
• Who can publish a page the week after launch: the answer decides how much the site costs you in the second year.

If you want that scoping done alongside the build, our web development work starts with scope and architecture for exactly this reason.

Everything above reduces to a single habit. Count your templates before you request a quote, every time. It takes an hour, costs nothing, and converts a vague budget question into a specific scope conversation that a vendor can actually price.

This week, list the page types your site needs at launch and mark which ones need their own layout. That single list is the most useful document you can bring to a first call.