How much should a business website cost?

For an SMB with 5–15 pages: €8,000–€42,000 depending on platform, complexity, and integrations. For mid-market with multilingual, multiple service lines, lead-capture: €30,000–€95,000. Enterprise: €80,000+. Pricing wildly outside these ranges (either way) usually signals a mismatch.

Do we need to rebuild our existing site, or can we improve it?

If your current site loads in under 2 seconds, ranks for the queries that matter, converts visitors at industry-typical rates, and your team can publish without engineering involvement — improve it. If two or more of those are missing, the rebuild math usually wins.

WordPress, Webflow, or custom?

We do not start new builds on WordPress in 2026. Webflow is fine for marketing sites that don't need server-side logic. Custom on Next.js for anything with depth — multilingual, e-commerce, complex content models, integrations.

How long does a real business website take to build?

8–14 weeks for a typical mid-market corporate site. Faster if the scope is genuinely tight. Slower if there's content migration, multilingual, or integrations involved. Anyone offering a serious business site in 'two weeks' is selling a template.

What's the single biggest mistake businesses make?

Optimizing for what the website looks like in a presentation to the CEO instead of for what it does when a customer arrives. The internal review meeting tends to elevate visual polish at the expense of conversion clarity.

Three rules for building a business website that earns its budget

Why three rules

There are about a hundred things to get right on a business website. Most of them matter at the margins; three of them produce the outcomes. Across the commercial builds documented in the Marketing MIX portfolio, these three are what we end every kickoff call with.

Rule 1: One question above the fold, answered in seven words

Visit your homepage. Imagine a stranger has landed there from a Google search or a referral. Within seven seconds, that stranger needs to answer one question: what is this, and is it for me?

If the answer requires scrolling, clicking, or reading past more than seven words, you've lost the meaningful half of the audience. Bounce rates above 70% on cold traffic almost always trace to a failure of above-the-fold clarity.

The fix:

One H1 in plain language. Not a clever tagline. State what you do and for whom.
One sub-headline that adds the next layer of specificity — what makes you different, or who you serve precisely.
One primary CTA that matches the audience's stage. "Book a demo" works for B2B SaaS; "Shop now" works for e-commerce; "Get a free audit" works for many service businesses.
No more than three above-the-fold elements competing for attention.

Examples of homepages that pass:

Stripe: "Payments infrastructure for the internet" — clarity in seven words.
Linear: "Linear is built for the world's fastest software teams."
Anthropic: "Build with Claude" — explicit, specific.

Examples that fail (paraphrased from real sites we've audited):

"Transforming businesses through innovative digital solutions." Nobody knows what you do.
"Welcome to [Brand Name]." Doesn't earn its space.
"Empowering tomorrow's leaders today." Pure motivational poster.

Rule 2: Speed is a brand attribute

Site speed is no longer a technical concern. It's a brand attribute that customers experience directly. A site that loads in 0.9 seconds feels like a different brand than the same site loading in 4.2 seconds — different perceived quality, different perceived seriousness, different conversion rates.

The targets that matter in 2026:

| Metric | Target | Why | |---|---|---| | Largest Contentful Paint (LCP) | < 2.5s | Google ranking factor, perceived load time | | Cumulative Layout Shift (CLS) | < 0.1 | Visual stability — shifting content feels unprofessional | | Interaction to Next Paint (INP) | < 200ms | Replaced FID in 2024; measures responsiveness | | Time to First Byte (TTFB) | < 800ms | Server response speed | | Total page weight (mobile) | < 1.5MB | Loading on slower connections |

These aren't aspirational. We hit them on every site we ship. Sites that don't lose business — to bounce rate, to ranking position, to customer perception.

What gets you there in 2026:

Server-side rendering (Next.js, Astro, or equivalent). Single-page apps with client-side rendering are increasingly the wrong default.
Modern image formats (AVIF, WebP) with proper srcset.
Edge delivery (Cloudflare, Vercel, or similar CDN-edge).
Minimal third-party JavaScript. Every analytics tool, chat widget, and tag manager has weight; audit them quarterly.
Critical CSS inlined; non-critical CSS deferred.
Font loading optimized — font-display: swap, preload critical fonts.

Rule 3: The CMS your marketing team can actually use

Every business website that ages well has one thing in common: the non-technical marketing team can update it without filing tickets. Every business website that becomes a maintenance burden has one thing in common: structural changes require engineering, so structural changes stop happening.

The CMS choice is one of the most important decisions in a build and one of the least talked about. We typically recommend:

For content-heavy sites:

Sanity — best in class for content-as-data. Real-time collaboration, structured content, generous free tier.
Contentful — older, more enterprise, more expensive. Strong API, mature ecosystem.
Strapi — open-source, self-hosted, free of vendor lock-in.

For e-commerce:

Shopify's CMS for Shopify stores (it's competent for marketing content).
Sanity + Shopify for headless commerce with deep content needs.

For marketing-team-friendly visual editing:

Webflow — when the team needs to make visual layout changes, not just text edits. Better than self-hosting WordPress for most use cases.
Framer — newer, fast-growing, especially good for design-led marketing sites.

For developers-only content:

MDX in the repo — when the team is technical, this works well and stays version-controlled.

What we avoid for new builds: WordPress (security and maintenance burden), Wix and Squarespace at any meaningful business scale, Drupal (overkill for the use cases that fit), Magento open-source (better commercial alternatives now exist).

The fourth rule we don't talk about

There's a fourth rule but it's not a rule, it's a discipline: ship and iterate. The websites that produce business outcomes are the ones that get measured, improved, measured again. The ones that don't get improvements are the ones where the launch was treated as the project, not the start.

Build the launch with a 30/60/90-day improvement plan attached. Identify what you'll measure (conversion rate per page, time-on-page, scroll-depth, form-completion rates, organic-traffic growth). Set review cadence. Make small changes weekly, larger changes monthly.

This is how a website goes from a one-time deliverable to an ongoing asset.

For our corporate-website practice: /development-of-corporate-websites. For e-commerce specifically: /online-store-development. For landing pages: /landing-page-development. For the SEO that compounds with a good site: /seo-promotion.

Written by Maksym Stepanenko, founder of Marketing MIX. Last reviewed: 2026-05-13.