I've spent twenty-plus years building websites in and around Los Angeles — for NBCUniversal on one end and a family-owned cleaning company on the other. The most common question I get hasn't changed in two decades: "what should this cost?" The honest answer is that the LA market quotes anywhere from $500 to $250,000 for "a website," and both extremes are usually the wrong purchase. Here's how I actually see the market break down, and how to know which band you belong in.
Band one: roughly $1,500–$5,000. A focused site for a local service business — five to ten pages, built on a proven platform, professional but not custom-engineered. This is the right purchase for most restaurants, cleaning companies, contractors and solo professionals. The mistake people make here isn't overpaying; it's underpaying — a $500 template job with stock text that says nothing specific about the business converts nobody and ranks for nothing.
Band two: roughly $5,000–$25,000. Custom design and development: real information architecture, copy written for your customers, conversion paths that were actually thought about, e-commerce, bilingual versions, integrations. Medical practices, law firms, established B2B companies live here. Most of my client work over the years — from Beverly Hills Plastic Surgery Group to Calpiping — falls in this band.
Band three: $25,000 and far beyond. This isn't a website anymore; it's a product. Platforms, marketplaces, apps with accounts and data. YESHVA, the community platform I founded, lives here. If someone quotes you $60,000 for a brochure site, walk. If someone quotes you $8,000 for a two-sided marketplace, walk faster — they don't understand what they're agreeing to build.
Page count matters much less than people think. What moves the number, in rough order: content (who writes it — writing is half the work on a good site), e-commerce (catalog size, payments, shipping logic), bilingual builds (a proper Spanish version is closer to 40% more work, not 5% — see why word-for-word translation fails), integrations (booking, CRMs, member areas), and photography (real photos of your actual work outperform stock every single time, and someone has to take them).
Three patterns, over and over. Paying monthly for platform tiers and plugins nobody uses — I've audited businesses spending $400 a month to run what is functionally a five-page brochure. Rebuilding sites that weren't broken because a salesperson said "modern" enough times — a redesign without a conversion problem to solve is decoration. And the big one lately: paying agency prices for unedited AI-generated pages. If your ten service pages read like everyone else's ten service pages, you paid for something Google and AI search engines have already decided to ignore.
I work three ways: a defined project with a fixed scope and schedule, an ongoing monthly rhythm for businesses that want the site to keep improving, or a one-time strategic audit when you need a roadmap before committing real money. The first conversation is free, and if the right answer for you is band one and a good Google Business profile, I'll tell you that — it costs me a project and saves you a mistake, which is a trade I've never regretted.
For a band-one local business site, two to four weeks. Custom projects usually run six to twelve weeks, and most of the variance is content: sites stall waiting for text and photos, not code. Platforms are measured in months, not weeks.
Hosting and a domain, yes — usually $10–$40 a month for a normal business site. Beyond that, be suspicious. My rule is that you should own everything: your domain, your hosting account, your content. If leaving your web company means losing your website, you were renting, not buying.
Yes — when you're testing a new business and need a presentable placeholder while the real budget goes to the work itself. It becomes the wrong call the moment the website is supposed to bring in customers, because template sites with generic copy don't rank and don't convert.
You can, and for a hobby project you probably should. For a business, the builder isn't the problem — the output is. AI builders produce structurally identical sites with interchangeable copy, and search engines increasingly treat that sameness as a signal to rank something more specific instead.
Have a question these pages don’t answer? Ask me directly — or ask my assistant right now.
Start a Conversation