Most agencies pitch their custom website process as a tidy diagram: Discovery → Design → Develop → Launch. That diagram is accurate the way "go on vacation" describes a flight. It’s true in shape and wrong about the texture. The actual experience of building a custom website — the real time investment, what gets shipped each week, where the surprises sit — is what most owners only learn during their first project.

I’m David Campbell, founder of Nerd Stack. We’ve been running custom website builds for 9+ years, and this is the honest week-by-week version of what actually happens — written for owners about to commit to one, so the experience matches the pitch when the project starts.

Why "Discovery → Design → Develop → Launch" Is Misleading

The four-phase diagram makes a custom build sound like a sequence of discrete handoffs: discovery ends, design begins, design ends, development begins. The actual project doesn’t work that way. Phases overlap. Decisions made in week two affect deliverables in week six. Your time as the client is concentrated in specific moments, not evenly distributed. And the part of the project that produces the most anxiety for owners — the gap between “designs approved” and “site is live” — is the part the diagram glosses over the hardest.

Below is what a typical 8-week Nerd Stack custom website build actually looks like from the inside. The exact week numbers shift project to project; the pattern stays consistent.

Week 1: Discovery (Your Heaviest Week)

Discovery is where the entire rest of the project is decided. We spend 60–90 minutes with you in a structured discovery session, then several days writing up what we heard.

What actually happens: we ask about your business — your customers, your competitors, what you do that other people can’t, what success looks like in concrete terms. We push on the answers, because most businesses describe themselves in language that’s useful to them and meaningless to a prospect. By the end of the session we know who the website needs to convince, of what, and why.

What you get: a written strategy document that defines audience, positioning, goals, and scope. Both sides sign off on it. Every design decision that follows is anchored to it.

What surprises owners: how much you’ll be asked to clarify things you thought were obvious. “Who’s our customer” sounds simple. “Who, specifically, by demographic and psychographic and decision stage, and what differentiates the 30% who buy from the 70% who don’t” is harder. That’s the point — vague answers here produce generic websites later.

Week 2: Strategy and Wireframes

Now we design the bones of the site. Sitemap, page-by-page wireframes, the conversion path through the site — low-fidelity layouts that show structure, hierarchy, and flow without any visual design yet.

What actually happens: we map every key page (homepage, services, case studies, contact, and the ones specific to your business) as a clean wireframe. The wireframes are deliberately ugly — gray boxes and placeholder text — so the conversation stays on layout and conversion logic rather than “I don’t love that shade of blue.”

What you do this week: review the wireframes carefully. This is where structural changes are essentially free. Once we move to visual design, restructuring a layout costs real time.

What surprises owners: how revealing wireframes are. Without color and imagery distracting you, it becomes obvious whether the path from landing to action is clear — and where it isn’t. Most clients catch at least one structural issue here that would’ve been expensive to fix later.

Weeks 3–4: Visual Design

Now the site gets its actual look. We apply your brand — typography, color, layout patterns, imagery direction, motion — across every wireframe, producing high-fidelity Figma designs for desktop and mobile.

What actually happens: the first design round usually lands in week 3. We present desktop and mobile mockups, walk you through the decisions, and collect feedback. Round two responds to your feedback. Round three is for fine-tuning.

What you do these weeks: two or three real design reviews, each requiring 30–60 minutes of focused attention. Treat them seriously — once designs are approved, the site that gets built is the site you signed off on. Half-engaged design review is the #1 cause of post-launch “why does it look like this” surprise.

What surprises owners: how much restraint matters. Most owners enter design review wanting more — more sections, more copy, more images, more announcements crammed above the fold. Our job is to push back on the additions that hurt conversion. The websites that perform are usually less crowded than owners initially want.

Weeks 4–7: Development

Designs get built into a real, working website. This is the longest single phase and the one you’re least involved in — most of the heavy lifting happens on our side.

What actually happens: we build the site on a modern stack (Next.js by default, WordPress with custom themes where the project calls for it). Performance, SEO, and accessibility are built in from the first commit, not bolted on at the end. You get a staging URL — a private link to the site as it’s being built — so you can watch progress and flag issues in real time rather than at the end.

What you do this phase: review the staging site as pages come online. Provide final copy and content. Make any small adjustments. Mostly: stay out of the way and let the build happen.

What surprises owners: the staging URL is both a blessing and a trap. It’s great for catching issues early. It’s also tempting to keep tweaking long after substantive decisions should be locked. A good development phase ends with a site you’re proud of; a long development phase ends with a site that’s been polished into something tentative.

Week 8: Content, QA, and Launch

The final week is the most coordinated and the most boring. Everything needs to be in place before launch: content migrated, QA across every browser and device, performance tuned, redirects from the old site mapped, analytics and Search Console configured, DNS prepared.

What actually happens: we run through a long internal QA checklist. We coordinate with you on launch timing (Tuesday mornings, generally — not Friday afternoons). We flip DNS, monitor for issues, verify the old-site redirects preserve SEO equity, and verify the conversion path works end-to-end on a live URL.

What you do this week: approve the final QA pass. Pick a launch time. Tell your customers if it makes sense to. Otherwise: let us handle the coordination.

What surprises owners: launch is anticlimactic. A well-prepared launch is a non-event — the site just becomes live, traffic flows in, everything keeps working. The drama owners brace for never materializes, because all the drama has already been resolved in the previous seven weeks.

What the Diagram Doesn’t Show

Beyond the week-by-week, three things consistently surprise owners going through a first custom build:

  • Your time investment is heavily front-loaded. Weeks 1–2 are the heaviest for you. By weeks 5–6, you’re mostly waiting. That distribution catches many owners off guard — they assume the workload will be even and try to ramp up in the back half, when there’s little to do.
  • The biggest cost driver is decision quality, not scope. A clear answer on positioning and audience in week 1 saves more time over the whole project than any single feature decision. The owners who get the best sites are the ones who took the discovery session seriously, not the ones who paid the most. The industry data supports this: the Standish Group's CHAOS research found that small, well-scoped software projects succeed at roughly a 90% rate while large projects succeed less than 10% of the time — and the single biggest predictor of success is scope clarity established up front.
  • Post-launch is where the work continues. The site shipping isn’t the end — it’s the start of a different phase. First 30 days are about monitoring, performance tuning, and small adjustments based on real-user behavior. From there, ongoing maintenance and content keep the site healthy. For more on that, see our guide to website maintenance plans.

What Different Project Types Change

The 8-week timeline above is for a typical SMB custom website with 8–12 pages. Variations:

  • Simpler projects (4–6 pages): 6 weeks total. Discovery and strategy stay the same; design and development compress.
  • Larger marketing sites (15+ pages): 10–14 weeks. The structure is identical; each phase takes longer.
  • Redesigns vs. new builds: usually similar time, but discovery shifts toward auditing what’s currently working and what isn’t. Our redesign vs. refresh guide covers when each is the right call.
  • Custom web apps: different process entirely — phased builds, longer timelines, ongoing iteration. We cover that in our custom web app cost guide.

What This Process Looks Like in Practice

For real-world examples of what this process has produced, our case study portfolio includes:

  • Deseo Salon — luxury hair salon, custom service and pricing page template system, 60% lift in service page booking click-throughs
  • Reichelt Capital — private investment firm, custom WordPress build with private client portal
  • Hey Eduardo — privacy-first AI desktop product, Next.js marketing site with 4.2% visitor-to-purchase conversion
  • Overland Park Men’s Club — 130-year-old golf club, WordPress with custom tournament management

Each of those went through the process above — with the exact phases adapted to the project type, but the underlying structure consistent across all of them.

Frequently Asked Questions

How long does a custom website take to build?

Most SMB custom websites take 6–10 weeks from discovery to launch. Simpler projects come in around 6 weeks; larger marketing sites with 15+ pages typically run 10–14 weeks. We provide a realistic timeline before the project starts based on actual scope, not a generic estimate.

How much of my time will a custom website project take?

Your time investment is heavily front-loaded — heaviest in weeks 1–2 (discovery and wireframe review), moderate in weeks 3–4 (design reviews), and light in weeks 5–8 (mostly waiting for development and providing final content). Plan for roughly 8–12 hours of your time spread across the project, concentrated in the first half.

What happens if I want to make a change mid-project?

Small adjustments within the existing scope are part of the process. Material changes to the scope — adding pages, restructuring sections after design approval, adding integrations — go through a documented change-order process so you see the impact on timeline and cost before we do the work. Surprise invoices for “extras” aren’t a thing in our process.

Can the timeline be compressed?

Yes, but with tradeoffs. Rush timelines under 4 weeks carry a premium and compress the design-review phase, which is the place we don’t recommend compressing. Discovery and strategy can’t meaningfully be shortcut without producing a generic site. We’ll tell you honestly whether your timeline is achievable.

What happens after launch?

The first 30 days post-launch include monitoring, performance tuning, and any small adjustments based on real user behavior — included with every build. From there, an optional maintenance partnership covers ongoing security, performance, content updates, and new features. See our guide on whether you need a maintenance plan for what that actually includes.

What if I’ve never gone through a custom website project before?

Most of our clients haven’t. The process above is built to make a first custom build feel manageable — you’re heavily supported in the high-decision phases (weeks 1–2 and 3–4), and you’re mostly waiting in the development phase. The most important thing for a first-timer is taking the discovery session seriously; everything downstream depends on it.

Bottom Line

The diagram is true. The texture is what owners learn the hard way on their first project — that discovery is heavier than expected, that wireframes reveal more than designs do, that launch is anticlimactic, and that the time investment is concentrated in the first half. Knowing that going in makes the experience match the pitch.

If you’re evaluating a custom website project and want to walk through what your specific timeline and process would look like, book a free call. We’ll talk through your project shape, timeline expectations, and the realistic version of what your first few weeks would look like with us.

For the structured reference version of our process, see our process page. For pricing detail, the Denver web design cost guide covers the full picture, and the cost calculator gives you an instant estimate.

Sources: Standish Group — CHAOS Project Resolution Benchmark.