A lot of the conversations I have with one-truck operators and small crews start the same way: “I’ve been burned by a web shop before. Tell me what working with you actually looks like.”
Fair question. Here’s the honest, step-by-step.
Before the call
When you reach out (form, email, or LinkedIn), I usually reply inside a few hours, and definitely within one business day. I’m a real person, not a sales team. The first reply is short and asks two or three questions to make sure we’re a fit before we put a calendar slot on the books.
If we’re not a fit, I’ll say so. The two most common honest “nos” are:
- Your budget is genuinely tighter than a custom build supports. A real custom build starts around $4,000. If your business is smaller than that, a website builder will serve you better. (How to choose between custom and a builder.)
- You don’t have a stable list of services or service areas yet. Custom builds work when the scope is clear. If you’re still figuring out what your business does, start with something cheaper and let it iterate before paying for a custom one.
Neither of those is an insult. They’re me trying to make sure you don’t spend money on the wrong thing.
The discovery call
Thirty minutes. No slide deck. No “let me tell you about Salt & Scale.” We talk about your business.
What we usually cover:
- What you actually do. Specific services, not categories. “Tankless water heater installs” beats “plumbing.”
- Where you actually do it. Specific cities and neighborhoods. Service-area pages live or die on this.
- How people find you today. Google Business Profile, word of mouth, repeat customers, paid ads, all of the above.
- What’s broken about your current setup. Site that doesn’t rank, form that doesn’t work, phone that doesn’t ring, design that doesn’t look like the business you actually run.
- What “won” looks like a year from now. More leads, better-qualified leads, a brand you’re proud to hand a business card for. Usually all three.
At the end of the call I tell you whether I think a custom build is the right call, or whether something simpler would serve you. If it’s the wrong call, I’ll point you at the right one. I’d rather lose a project than ship the wrong thing.
The written proposal
Three business days after the discovery call, you get a written proposal. Usually one PDF, two to four pages. What’s in it:
- Scope. Every page that will exist on the site, in a bulleted list. Homepage, services, service areas, about, contact, plus anything specific to your business.
- Deliverables. What you receive at the end: a live site, a GitHub repo, hosting under your account, a sitemap submitted to Search Console, a walkthrough Loom.
- Timeline. Week-by-week. “Week 1: design. Week 2: build. Week 3: revisions and launch.” Real dates, not “starting soon.”
- Price. One fixed number. Not “starting at.” Not “depending on.” One number, in writing.
- What’s not included. Photography, copywriting if you’d rather hire a copywriter for the long-form, ongoing maintenance after the included post-launch window. Listed explicitly so there are no surprises.
You take a few days to read it. You sign it, or you ask questions, or you walk away. No pressure, no “this price expires Friday” tactics.
Design before any production code
This is where most engagements get derailed when shops skip it. Before any production code gets written at scale, you see a high-fidelity design of:
- The homepage. With your real services, real service areas, real photos if I have them. Not a stock template with “Lorem ipsum.”
- One inner page. Usually a services page or a service-area page, because those carry the heaviest layout work.
You review. You request changes. We agree on direction, typography, color, structure. Then I build.
Two revision rounds are built into the price. Most projects use one. A few use two. I’ve never had a project need three rounds where the issue wasn’t a fundamental scope miss on my end, which is on me to fix, not on the meter.
The build
This is the part most owners worry about and shouldn’t. The build is the most predictable phase.
- Staging URL from day one. You can look at the site at any point. No “we’ll show you in two weeks.” You see it as it gets built.
- Weekly check-ins by text, email, or a quick call. Usually a 10-minute “here’s what shipped this week, here’s what’s next.” If you don’t want a call, you don’t get one. Some owners want a weekly call. Some prefer a Friday email. Either is fine.
- AI accelerates the scaffolding. That’s how a 15-page site ships in under three weeks. But every line is reviewed before it goes on the staging URL. AI doesn’t ship; I do.
Most builds in this range take 1 to 3 weeks of calendar time. The fastest so far, Elevated Water Solutions, went from idea to production in seven days. Bigger scopes (more pages, custom integrations, multiple decision-makers) can push toward four weeks. If a project genuinely needs more than four weeks, that’s in the original proposal in writing, not a surprise halfway through.
Launch day
Three things happen in roughly this order:
- Final review. You walk the staging URL one more time, top to bottom. We fix any last-minute typos or photo swaps.
- DNS cutover. I move the domain to point at the new site. There’s usually 5 to 30 minutes of propagation; sometimes a few hours for stragglers. Email is unaffected; only the website moves.
- Sitemap submission and post-launch checks. Sitemap submitted to Google Search Console. Structured data validated. PageSpeed and accessibility re-run on the live site. Any post-launch fixes that come up in the first 48 hours are on me, not on the bill.
You also get a walkthrough Loom. A recorded 10-to-15-minute video showing you the few things you’ll actually touch: updating phone numbers, swapping a photo, adding a new service area. Most owners spend less than 10 minutes a month on the site after launch. The Loom covers that.
After launch
A few common patterns, none of them required.
- Small tweaks. A new photo, a new service, a new city, a new license number. Send a Loom or a text. Most things land within a few business days. The first 30 days are included; after that, small tweaks bill by the change or against a retainer.
- Retainer. Some owners keep me on an hour or two a month for ongoing tweaks, performance monitoring, and content updates. Most don’t need it after the first few months.
- New features. A new section, a new integration, a new conversion funnel. We do a quick discovery on the phone, I send a small written estimate, you decide.
I’m not chasing you for upsells. If the site is doing its job and your phone is ringing, the right answer is “leave it alone.” I’ll tell you that.
What I won’t do
Spelling this out because it’s the other half of the deal.
- No hourly billing. Fixed-price quotes only. If a project takes me 20% longer than I estimated, that’s my problem, not yours. The flip side is that I scope conservatively and don’t take projects where the scope is unclear.
- No junior handoffs. I write the code. I review the code. I deploy the code. If you’re paying for senior-engineer judgment, you’re getting senior-engineer judgment, not a Slack message routed to someone you’ve never met.
- No “we own your site” relationships. The repo lives under a GitHub account you control. The domain stays in your name. The Vercel hosting account is yours. If we part ways tomorrow, you have everything you need to keep running the site or hand it to another engineer.
- No change-orders for “scope creep” that I created. If I missed something in the original scope, I fix it on my time. Change-orders only apply when you genuinely change the scope (a new feature, a major redesign, a new page you didn’t ask for in the proposal).
- No agency markup on hosting or tools. You pay Vercel what Vercel charges, you pay Resend what Resend charges. I don’t resell anything.
The cost of getting this wrong
The reason this process is structured the way it is: I’ve seen what happens when it isn’t.
- Hourly billing without a cap leads to bills that double the original estimate, then double again. The shop you hire knows how to bill; you don’t know how to challenge it.
- No design approval before building leads to a site you hate at launch and either a fight over revisions or a launch you’re embarrassed to share.
- Junior handoffs lead to “the senior engineer who sold you the project will be back next quarter” while a junior makes decisions they don’t have the context to make.
- Vendor-owned domains and hosting lead to ransoms when the relationship ends. I’ve seen $5,000 invoices for “transfer assistance.”
None of those things happen on a project I run. The proposal is the proposal. The price is the price. You own everything when we’re done.
The bottom line
If you run a one-truck operation or a small crew and you’ve been burned before, the relevant questions to ask any vendor are:
- Can I see a written proposal with a fixed price before I sign anything?
- Who actually writes the code?
- Do I own the repo, the domain, and the hosting account at launch?
- What happens after launch if something breaks?
Those are also the things I lead with on the discovery call, because they’re the questions that should have been answered on every previous engagement and probably weren’t.
If you want to see how this works for your specific business, the next step is a 30-minute call. No slide deck. No sales pitch. Just questions.
Ready for the discovery call? Schedule a project. Most replies land within a few hours, and definitely inside one business day.