Web design services explained: what you're actually paying for.

Any agency that knows what it is doing will move you through seven phases. The names vary. The work does not. Skipping a phase always costs you more than it saved.

• Discovery. Interviews with stakeholders, customers, sometimes competitors. Goals, audiences, success criteria. Output: a written brief.
• Information architecture. The site map, the navigation logic, the content hierarchy. Output: an annotated sitemap and content outline.
• Wireframing. Low-fidelity layouts of each template, showing what goes where without colors or imagery. Output: clickable wireframe prototype.
• Design system. Typography, color, spacing, components. The pattern language that every page is built from. Output: a documented design system.
• Comps (visual design). Polished, full-fidelity mockups of the key templates. Output: design files for every major page.
• Build. Front-end and back-end development, CMS configuration, integrations. Output: a staging site.
• QA and launch. Cross-browser testing, accessibility audit, performance testing, content migration, DNS cutover. Output: a working production site.

A good project also includes an eighth, post-launch phase: warranty, analytics setup, training, and ongoing care.

Discovery is the phase clients hate paying for because the deliverable is a document, not a picture. It is also the phase that determines whether the rest of the project succeeds. Without real discovery, the design team is guessing at what the business needs and the buyer is paying for those guesses to be wrong.

A real discovery phase includes stakeholder interviews, customer interviews (yes, with actual customers), a competitive scan, an audit of the existing site analytics, and a written brief that everyone signs off on before design begins. Two to four weeks on a mid-sized project. Nielsen Norman Group's research on discovery practices remains the canonical reference for what good discovery looks like.

The structure of your site (what pages exist, how they nest, what links go where) is decided in this phase. It is also the cheapest thing to get right and the most expensive thing to get wrong. A bad sitemap costs you organic traffic, customer confusion, and a redesign three years later.

The deliverable for this phase is an annotated sitemap that shows every page on the site, the relationships between them, and the primary content that will live on each. If your agency hands you a sitemap with vague labels like "About" and "Services" and nothing else, push back. Real IA is granular.

The temptation is to skip wireframes and go straight to polished visual design. The reason every experienced design team refuses is simple: it is far cheaper to argue about page layout in grayscale than after the visual design is built.

Wireframes force the conversation to be about content priority, not color. Once you have approved the wireframe, the visual design phase is faster and the comp revisions are smaller, because the underlying argument is already settled.

Pricing for web design services falls into rough tiers. These ranges assume a competent agency, original work (not a template flip), and a full project (not a one-page landing site).

• Small business / SMB ($8K to $25K). A 10 to 30 page site, light discovery, design built on an existing system (often a customized template), build on a mainstream CMS (WordPress, Webflow, Shopify). Two to three months timeline.
• Mid-market ($30K to $90K). A 30 to 100 page site, real discovery, custom design system, custom CMS configuration, light integrations. Three to five months timeline.
• Enterprise ($120K to $500K+). A 100+ page site or a complex web app, full research phase, custom design system documented for ongoing use, deep integrations (CRM, ERP, marketing automation), staged rollout. Six to twelve months timeline.

Below these ranges you are usually buying a template flip with a logo on it. Above these ranges you are usually paying for an agency's overhead, not your project. The middle of each band is the honest center of gravity.

Every shortcut in a web project costs more than it saves. The math is consistent enough to be useful as a rule.

• Skip discovery: expect to redesign at least one major template after launch because nobody understood the audience.
• Skip information architecture: expect to lose 20 to 40 percent of your organic traffic in the months after launch as Google reindexes the new structure.
• Skip wireframing: expect twice the design revisions and a launch date that slips by four to eight weeks.
• Skip a real QA phase: expect to launch with broken forms, broken redirects, and accessibility issues that the team will fix in the months after going live.

The shortcuts are not free. They are loans, with interest.

A useful set of questions for any agency you are considering:

• Walk me through your discovery process. Who do you interview and what do you produce?
• How do you handle content? Are you writing it, or are we?
• What CMS will we be on and why? What is the long-term cost of ownership?
• What does your QA process include? Specifically, accessibility, performance, cross-browser.
• What happens in the first 90 days after launch? Is that included or extra?
• Can I talk to two clients whose sites you launched 12 to 18 months ago?

The last question is the most useful. A site looks good at launch and reveals its actual quality a year in. Agencies whose 18-month-old work still looks current and performs well are the ones worth hiring. Smashing Magazine's archive on web project management has decades of writing from practitioners on what separates good agency engagements from bad ones.

The patterns are remarkably consistent. If you spot more than two of these, slow down.

• The agency wants to skip discovery and "just start designing." They are trying to bill billable hours faster.
• The proposal is one page and one fixed number with no phase breakdown. You will not know what you bought until you see the invoice.
• The first comp is delivered without any wireframe approval. You will spend twice the revision cycles getting to the right answer.
• The agency does not ask about your existing analytics, content, or SEO performance. They are planning to bulldoze whatever you have.
• The team that pitches you is not the team that does the work. Find out who will actually be on the project before you sign.

Web design is one of the few professional services where the gap between the best and worst providers is enormous, and the buyer often cannot tell from the pitch. The questions above are how you tell. The A List Apart archive on agency practice is a useful second opinion on what good engagements look like from the inside.

A good website is a five-year asset. The hours and dollars you spend on the project are amortized across every visitor, every lead, every sale for that whole period. Shorting the project to save 20 percent up front almost never produces 20 percent better returns. It usually produces a site you redesign in three years instead of five.

The right question is not "what does this cost." The right question is "what will this be worth in 2029."