A template is enough for most small business sites, as long as it is fast, accessible and carries your own brand. Custom website design pays when you need flows no theme supports, when looking like rivals costs you sales, or when a key page has measured friction. The best middle path is a template with your own design system on top.
Why is this a hard choice?
Both sides sell you a myth. Template sellers say any site can look premium in an afternoon. Custom studios say templates hurt your SEO and kill conversions.
The evidence is calmer than either. Speed, accessibility and first impressions depend on how a site is built and filled, far more than on where the layout came from. I'll go through each claim, then give you a decision tree you can use today.
This article skips two things. Platforms get their own comparison in WordPress vs Webflow vs Wix vs Shopify. Budgets live in what a small business website costs.
What do "template" and "custom" actually mean?
A template (or theme) is a ready-made set of page layouts, styles and components. You swap in your words, photos and colours. Most sell to thousands of buyers.
A custom design starts from your business. A designer works out your pages, your content and your look, then builds or configures it. It can still run on WordPress, Webflow or Shopify.
Most real sites sit between the two. A designer often starts from a theme or a component library and replaces the parts that matter. That middle ground is where most of the value is, and I come back to it below.
What does the evidence say about each claim?
Here is the short version. Each row has its own section further down.
| Claim | Best evidence | What it shows |
|---|---|---|
| Templates hurt SEO | Google Search Central docs | False. Relevance ranks first |
| Templates make sites slow | HTTP Archive, millions of sites | Depends on build. Spread within a platform is huge |
| Custom always converts better | No controlled study found | No evidence. Familiar layouts score better |
| Templates are less accessible | WebAIM Million, 1M home pages | Mixed. Hosted builders score better on average |
| Templates make you look generic | Theme sales, brand research | True if untouched. Your assets fix it |
| Visitors judge design instantly | Lab studies, 50 ms and 17 ms | Strong. Simple and familiar wins |
Do templates hurt SEO?
No. Google has no penalty for using a theme, and it ranks pages mainly on relevance.
Google's page experience guidance says its systems always try to show the most relevant content, even when page experience is weak.1 Core Web Vitals are used by its ranking systems, but Google warns that chasing a perfect score for SEO alone may not be worth it.1
The "shared code" worry is also old news. Back in 2008, Google said plainly that there is no such thing as a duplicate content penalty, outside deliberate scraping and spam.2 Your header and footer code matching 100,000 other sites is fine. Your words matching them is the problem.
What hurts SEO is thin copy, missing headings, slow pages and broken structure. A custom site can have all four. A well set-up template can have none.
Are template sites slower than custom ones?
It depends on the build, not the label. The spread inside one platform is far wider than the gap between platforms.
The 2025 Web Almanac, built on Chrome field data from real visitors, found big differences in mobile Core Web Vitals pass rates by system.3 85% of Duda sites passed, against 74% of Wix sites and 45% of WordPress sites.3 Hosted builders with locked-down templates did best.
Mobile sites passing Core Web Vitals, by CMS
Those are averages, though. The 2024 Web Almanac showed how wide WordPress pages range in mobile weight.4 The lightest 10% weighed 598 KB or less. The heaviest 10% weighed 7,826 KB or more.4 That is a 13-fold gap on the same platform, and much of it comes from what people add.
Same platform, thirteen times the weight
The 2025 chapter names the cause. It says variation across extensible systems comes more from themes, plugins, page builders and third-party tools than from the core software.3 About 60% of WordPress sites use a page builder, and builders often add heavier code.3
So a template can be fast or slow. A custom build can be fast or slow. The thing to check is the actual page. Run a theme's demo page, and then your own draft, through the page weight checker before you commit.
Does custom design convert better?
I found no controlled study that compares template and custom sites on conversion. The claim is a sales line.
What the research does show points the other way. In a 2012 study from the University of Basel and Google, Tuch and colleagues showed 59 people 119 website screenshots for as little as 50 ms. They then showed 82 people 80 screenshots for as little as 17 ms.5 Sites that looked simple and typical for their category got the highest beauty ratings.5 Google's write-up of the study says users strongly prefer designs that look simple and familiar.6
Jakob Nielsen made the same point in 2023 with Jakob's Law. Users expect your site to work like the other sites they already know.7 Templates are built from those shared patterns. That is a strength.
Where conversion does move, it moves in the details of a flow. Baymard Institute, which runs large usability studies and also sells audits, reports average cart abandonment of 70.22%.8 It says 17% of shoppers leave because checkout is too long or complex. It also says large stores could lift conversion by up to 35% with better checkout design.8 These are a research vendor's figures. They point at fixing specific steps, which you can do on a template. The checkout page design guide lists those steps.
Are templates worse for accessibility?
On average, no. Sites on hosted builders show fewer automated errors than the web as a whole.
The 2026 WebAIM Million scanned 1,000,000 home pages in February 2026. It found 95.9% had detected WCAG failures, with 56.1 errors per page on average.9 Squarespace home pages averaged 33.0 errors and Wix 33.3. WordPress averaged 52.8 and Shopify 75.1.9
Detected accessibility errors per home page
WebAIM is careful here. It says extra errors that come with a technology can't always be blamed on that technology.9 Big stores have more content, and that adds errors.
The practical lesson is simple. A theme starts you with tested buttons, menus and headings. Most errors arrive later with low-contrast colours, images without alt text and form fields without labels. Those come from whoever fills the template, so check contrast and labels whichever route you take.
Will a template make my business look generic?
Yes, if you leave it as it ships. The most popular themes sell to huge numbers of buyers.
Avada, the top seller on ThemeForest, shows 1,075,434 sales on its product page.10 That is the vendor's own count. Many of those buyers keep the demo layout, the demo fonts and stock photos. Visitors may not name the theme, but they sense they've seen it before.
Brand research explains why that matters. Jenni Romaniuk of the Ehrenberg-Bass Institute measures brand assets on two things, fame and uniqueness.11 An asset works when people link it to you and to nobody else. A demo layout fails the uniqueness test by design.
The fix is cheaper than a full custom build. Your logo, your colour system, your type pairing, your photography and your voice carry the brand. The grid underneath can stay familiar. If you're picking colours, my guide to colour psychology for brands shows which claims hold up.
How fast do visitors judge a site's design?
Within a twentieth of a second. This finding is old, but it has held up.
In 2006, Gitte Lindgaard's team at Carleton University showed people home pages for 500 ms and then for 50 ms.12 Visual appeal ratings were highly consistent between the two exposures.12 Tuch's 2012 study found effects at 17 ms.5
That cuts both ways. Visitors can't judge your code in 50 ms. They judge clutter, contrast and whether the site looks like what they expect from a business like yours. A clean template does well on that test. A bold custom design that breaks category norms can do worse.
When is a template enough?
A template is enough when most of these are true.
- Your site is mostly pages people read, such as services, about, contact and a few articles.
- Your competitors don't win on looks. Trust, reviews and clear prices matter more.
- The theme's demo passes Core Web Vitals and stays light with your real images.
- You will add your own logo, colours, type and photos.
- You need to launch in weeks, and you'll learn what to change from real visitors.
When does custom design pay?
Custom pays when one of these is true.
- You need a flow no theme handles well, such as a quote builder, a booking rule set or a product configurator.
- Looking like rivals costs you money. Premium, design-led and crowded local markets are the usual cases.
- A key page has measured friction. Analytics or user tests show where people drop out.
- You'll run the site for years and want one system your team can extend without breaking it.
Look at what's missing from that list. "Custom ranks better" and "custom converts better" are not reasons on their own.
Is there a middle path?
Yes, and it's what I recommend most often. Start from a solid template or component set, then design your own system on top.
A design system is a short set of rules. It covers colour tokens, a type scale, spacing, button styles and a few page patterns. You apply it across the template, so every page looks like you and nobody else.
Then you go custom only where it earns its keep. That is usually the home page hero, one or two key landing pages and any money flow such as checkout or a quote form. Everything else stays on tested template parts.
This keeps the template's strengths, which are familiar layouts, tested accessibility and fast launch. It fixes its weakness, the generic look. On my web design work this is the default starting point unless a client's flows rule it out.
Which should you pick?
Answer five questions in order. Each split rests on one piece of evidence from above.
Template or custom? Five questions, six outcomes
What should you do first?
Here's the order I'd follow.
Write down your must-have flows. List anything a visitor must do beyond reading and contacting you. If nothing is unusual, a template is in play.
Test two or three template demos. Run each demo through the page weight checker and Google PageSpeed Insights on mobile. Drop any that fail before you add a single photo.
Collect your brand assets. Logo, two or three colours, a type pairing and a photo style. If you don't have these, that is the first piece of design work to pay for.
Turn those assets into a small design system. Colour tokens, a type scale, spacing and button styles. Apply them to the template everywhere.
Go custom only where the data says so. After launch, watch where people drop off. Redesign that page or flow, and leave the rest alone.
Pick the right person for the job. A template setup and a custom build need different skills. My guide on how to choose a website designer covers what to ask.
Questions people ask
Is a website design template bad for SEO?
No. Google ranks pages mainly on relevance and has no penalty for themes or shared code. Thin copy and slow pages hurt rankings on any site.
Is custom website design faster than a template?
Only if it's built well. Speed depends on page weight, plugins and images. The spread inside one platform is far wider than the gap between template and custom.
What makes a website design look professional?
Simple, familiar layouts with clear contrast and consistent type. Research shows people rate simple, typical designs as the most attractive within 50 ms.
Can I start with a template and move to custom later?
Yes. Build your design system first. Then you can rebuild pages one by one without changing how the brand looks.
How do I make a template look unique?
Replace the demo fonts, colours and photos with your own. Add a design system and one or two custom sections on key pages. Your assets carry the brand, and the grid doesn't have to.