Almost every business owner shops for a website the same way. Collect three quotes, look at three portfolios, pick the one that feels the most professional for the money. Six months later the site is live, it looks fine, and nothing in the business has changed. Same call volume, same booking rate, same weeks where the phone just does not ring.
That is not usually a bad vendor. That is a bad question. "Who should build my website" is a shopping question. The buying question is "what should this thing do for me that my current one does not," and almost nobody asks it out loud before signing.
Decide What the Site Is Supposed to Do First
A website is not a deliverable. It is a machine with a job, and the job is different for every business.
A roofing supply yard needs a site that makes it obvious what is in stock and gets a contractor to the counter. A cosmetic dentist needs a site that turns a nervous browser into a consultation. A wildfire defense company needs a site that qualifies a homeowner before anyone spends an hour on a property assessment. Those are three completely different machines. They can all be beautiful and only one of them is right for you.
Write down the single job before you take the first call. One sentence. "I want more of X kind of customer to do Y without me chasing them." Now you have a standard to hold every proposal against, and you will be shocked how many proposals cannot answer it.
The Questions That Actually Separate Vendors
Price and portfolio are the two things every vendor is prepared for. These are not.
- What happens after someone fills out the form? If the answer is "you get an email," that is a brochure with a mailbox, not a system. Ask whether the lead lands in your CRM, whether the customer gets an instant confirmation, and whether anything happens automatically if nobody responds in an hour.
- Can a visitor book themselves? Not request a callback. Book. If the answer is no, ask why not, because the reason is usually that it was never scoped, not that it does not fit your business.
- What will you build that my competitors do not have? Estimators, visualizers, qualifiers, instant quote tools. Something that gives a visitor a reason to stay and a reason to hand over their information.
- Who owns the site, the domain, and the data when we stop working together? The correct answer is you. Get it in writing.
- What will you show me in ninety days, and what number will it be? If the vendor cannot name the number they are accountable for, they are accountable for nothing.
That is the whole list. If a vendor answers all five without deflecting, you are talking to a builder. If they steer everything back to design revisions and page counts, you are talking to a decorator.
Portfolios Tell You Less Than You Think
A portfolio proves someone can make things look good. It does not prove the things worked. Screenshots have no outcomes attached.
So ask about outcomes directly, and listen for specificity. A vendor who works on results can tell you what they measure and where the number came from. On our side, that sounds like Green Forever Arizona: 475+ leads tracked, 27.6% lead-to-appointment rate. For TurfSpa, it is 34.7% returning customers. Those are not bragging numbers, they are just proof that somebody was watching the whole path instead of the launch date.
You are not asking for guarantees. Nobody honest gives those. You are checking whether they instrument what they build.
Watch for the Quiet Red Flags
The obvious warnings are easy. The quiet ones cost more.
Beware the vendor who never asks about your sales process. Your site is the front of a machine that ends with somebody signing something, and a builder who does not ask how that part works is going to hand you a very pretty dead end. Beware the proposal that is all nouns and no verbs, pages and sections and revisions with no mention of what a visitor is supposed to do. Beware anyone who cannot explain in plain language how you will edit your own content, or who treats a small copy change as a new project.
And be honest about the flip side. If you cannot get anyone in your company to answer a form fill within the hour, no vendor can fix that with design.
This Week
Write the one sentence about what the site is supposed to do. Then go fill out your own contact form from your phone, on your data, like a stranger would, and time how long it takes to hear back from a human. Whatever you learn in those two exercises is the actual brief. Hand it to every vendor you talk to and let them tell you what they would build.