What's wrong with most small business websites

Most small business websites we see fall into the same pattern: a drag-and-drop builder template, a stock photo hero, a contact form that may or may not send, and a Google PageSpeed score somewhere in the 30s.

This isn’t a moral failing. The owners were sold a “$29/month, no developer needed” plan and left to figure it out. This post is about what’s actually wrong, when the template is fine, and when it isn’t.

When a template is fine

Let’s start here. If you run a business where your website is a glorified business card — address, hours, phone number, menu — a Wix or Squarespace site is often the correct choice. It’s cheap, you can update it yourself, and your customers don’t care whether it loads in 2 seconds or 4.

A lot of small businesses are in this category. We tell them so. Paying a developer for a custom marketing site is a waste when you’d get the same business outcome from a template.

When it isn’t

The template stops being fine when any of the following is true:

• SEO matters. You want to rank for “dentist in Etobicoke” or “tiffin delivery Brampton.” Core Web Vitals and semantic HTML are ranking factors, and most template builders are weak on both. After the March 2026 core update, the gap between fast, specific sites and slow, generic ones got wider, not smaller.
• You have custom functionality. Booking with availability, quote calculators, customer accounts, subscription management. Templates handle these poorly, if at all.
• You’re competing on brand. If your competitor uses the same Wix template as you, you’re differentiated only on price. That’s a tough position.
• You’re paying more than you realize. A $49/month Wix plan plus $30/month for their booking app plus $20 for their email app adds up to over $1,200/year. Three years of that is the cost of a custom site.

What “custom” actually gets you

Modern static and server-rendered frameworks (Astro, Next.js, Eleventy) produce sites that load quickly by default. For a marketing site, this means:

• Sub-second page loads, even on a mid-range phone on a slow connection.
• Low or zero JavaScript sent to the browser. Astro, specifically, ships zero JS per page unless you opt into it.
• Clean HTML that Google can index properly.
• Real content ownership. Your markdown files are yours. You don’t lose them if the platform decides to change their export format.

Custom also means we can wire in whatever backend services you actually use — a real booking engine, a CRM integration, a custom payment flow — without wedging them into a builder’s plugin marketplace.

What custom doesn’t cost

The common assumption is that a custom site is $10,000 and a six-month project. Neither is usually true.

For a well-scoped small-business marketing site (5–15 pages, a contact form, some SEO work, a simple CMS), we typically quote in the low four figures and deliver in 2–4 weeks. That’s comparable to a year of template-builder plans and the switching costs that come with them.

Where the number gets bigger is when the scope grows — custom booking, multi-vendor support, multilingual, accounts, billing. Those are real projects, with real engineering, and they cost accordingly. But that’s because they’re actually doing more than a marketing site, not because “custom” itself is expensive.

How we think about small-business sites

Our process for a typical small-business site looks like this:

1. A 30-minute call to understand what the site is for. If it’s truly a business card, we tell you to use Squarespace.
2. A rough scope and fixed price, written down, before any design work starts.
3. A clickable design prototype in the first week so you can see the thing before we build it.
4. Build on Astro, deploy to Vercel or Netlify, integrate whatever tools you actually use (Google Analytics, Mailchimp, Stripe).
5. Hand over a CMS so you can update copy and images without calling us. This is non-negotiable.
6. Two weeks of post-launch support included, plus an optional retainer if you want someone on call.

That’s the whole thing. It’s not a radical methodology and there’s no secret sauce. Most of the value is in being straightforward about what a website is for and not building more than that.

A 2026 note

Since we first wrote this, two things have made the template-vs-custom trade worse for owners who need real performance. First, Google’s Interaction to Next Paint metric became part of Core Web Vitals, and most builders ship enough JavaScript that they fail it on mid-range phones. Second, the March 2026 core update leaned harder into rewarding specific, first-hand content over generic pages, which is exactly what templates are worst at producing. The decision framework above hasn’t changed. The cost of being in the wrong half of it has.

If you’re trying to figure out whether your current site is costing you business, we’re happy to take a look.