Most service business owners hire a web designer the same way they hire a contractor: by asking around, picking the friend-of-a-friend with the best portfolio, signing the contract, and hoping. The hope-based approach works often enough that it has not been replaced. But the failure cases are expensive: a site that took six months to ship, a designer who disappeared after launch, a contract that locked the code into the designer's account, a project that ballooned past the original quote.
Seven questions, asked early, separate the designers who will help your business from the ones who will quietly drain it. Each one has a clear right answer, a clear wrong answer, and a yellow zone where the designer's response tells you more than the words themselves.
1. "Who actually writes the code?"
Right answer: "I do" (for a freelancer) or "the lead developer on the team, who you will meet on the kickoff call" (for an agency).
Wrong answer: "We have a development partner" (vague), "It is outsourced overseas" (cheap, with the predictable consequences), or any answer that does not name a specific human being.
Why it matters: the person writing the code shapes the site for the next ten years. If you cannot name them, you cannot reach them later. Most agency builds route through a pipeline of three to five people; the brief gets distorted at every hand-off and the final code is something nobody on the team can fully explain.
2. "What stack will the site be built on?"
Right answer: a specific platform name with a reason. "Custom-coded with Eleventy because it is fast and you can host it on Cloudflare for almost nothing" is a good answer. "WordPress with the X theme because the client prefers a drag-and-drop editor" is a fine answer if it matches the client's needs.
Wrong answer: "Whatever you want," "we work in everything," or any answer that suggests the designer has no opinion. A designer with no stack opinion is a designer who has not built enough sites to form one, and that is the wrong person to learn from on your project.
Why it matters: the stack determines the site's lifetime cost. WordPress is cheap to start and expensive to maintain; custom code is the inverse. Wix and Squarespace are cheap on both ends but cap your speed and SEO ceilings. There is no universally right answer; there is a right answer for your specific business, and the designer should be able to explain why.
3. "What happens if I want to leave?"
Right answer: the domain is in your name. The content (text, photos, brand assets) is yours. You receive a clean export of the site that runs on any host. The designer hands you everything you need to keep operating without them. The hosting account itself can either stay with the designer (common, and fine if the export is clean) or live in your name from day one (also fine), but the test is whether you can leave without losing anything.
Wrong answer: "The site stays on our platform" (lock-in), "We retain the source code" (a real red flag), "You will need to rebuild from scratch to leave," or any answer that makes leaving expensive or difficult. This is the question most likely to expose a designer who is selling a contract trap rather than a service.
Why it matters: the relationship will end someday. Either the designer retires, the agency goes under, the price changes, the work quality drops, or your business outgrows them. The exit terms matter more than the entry terms. Read the agreement before you sign — and if there is no agreement, that is itself a red flag.
4. "How fast will the site load on a phone?"
Right answer: a specific number, with a method. "Under two seconds, with a Google PageSpeed score above 90" is a good answer. "Under one second, mobile and desktop, and I can show you live runs on existing client sites" is a great answer.
Wrong answer: "It will be fast," "we do not really measure that," or "PageSpeed is not really meaningful." All three are excuses. Page speed is the single most important conversion factor on a service business site. A designer who cannot commit to a specific number does not actually know what their builds will produce.
5. "What is the total first-year cost, in writing?"
Right answer: a specific number that includes design, build, hosting, SSL, maintenance, and any recurring fees. "$2,100 a year, all in. Here is the line-item breakdown" is a good answer.
Wrong answer: a build cost without ongoing costs ("$5,000 for the build, hosting is separate"), or any answer that says "it depends on what you need" without offering a concrete bracketed range. This is where retainer creep hides.
Why it matters: the build cost is the smaller half of the lifetime cost in most engagements. A separate post runs the math in detail. The honest first-year number lets you compare apples to apples; the build-cost-only number does not.
6. "How will I make changes after launch?"
Right answer: a specific path. "Email me, I do it within one business day, included in the monthly fee" is a good answer. "There is a CMS dashboard you can edit from, and I help you with anything beyond that" is a fine answer if you actually want to do your own edits.
Wrong answer: "We bill change requests at $X per hour" (no included edits is unusual outside agency engagements), "You will need to learn WordPress" (only if you want to), or any answer that makes routine changes painful. The site will need changes every month for the next ten years; the change cost compounds.
Why it matters: the change-request friction is the #1 reason service business owners give up on their websites and let them go stale. If updating the site requires a sales call, an estimate, an invoice, and a two-week wait, the updates do not happen. The site that does not get updated is the site that stops earning.
7. "Can I see three real clients I can call?"
Right answer: a list of names with phone numbers or emails, with the client's permission. The designer who has done good work for service businesses for any length of time has a list of clients who will speak well of them, and the introduction is easy to make.
Wrong answer: "I cannot share that for privacy reasons," "Most of our clients prefer not to be contacted," or any answer that converts a routine reference request into something complicated. Reference checks are normal in every other service-business engagement; web design should be no different.
Why it matters: the portfolio shows the work. The references show what the designer is like to work with — which matters more, on a six-month or six-year engagement, than the visual quality of the portfolio.
Putting it together
A designer worth hiring will answer all seven questions cleanly and specifically. The yellow-zone signals are usually obvious: vague answers, defensiveness, a pivot to selling, an upsell on something you did not ask about, or a new question back at you ("what is your budget?") before they have answered the question you asked.
Seven questions takes about fifteen minutes to ask. The web design relationship will run for years. Fifteen minutes well spent up front saves a lot of pain later.
For reference, the answers to all seven, for the agency you are reading right now, are public on the site. The pricing page covers cost in detail; the service agreement covers what happens when you want to leave; the about page covers who writes the code; the methodology page covers the stack and the engineering posture. None of those are sales pages; they are the same answers any reasonable buyer should expect from any vendor.
If you are at the "considering web designers" stage, the next step is the free five-point audit on your existing site. The audit is independent of any decision about who builds the next one — it just gives you a written baseline to compare proposals against.
Every one of these questions, answered straight.
Custom-coded, $175 a month flat, no design deposit, twelve-month minimum. The agreement, the process, and the pricing are all public on this site.