We Tested 200 Dominican Business Websites. Almost All Are Slow on Mobile.

Which platforms do Dominican businesses build their websites on, and does the choice make them faster? To find out we took 200 business websites from Google Maps across 8 sectors and 6 cities, identified the platform behind each one from its page source, and tested every site with Google PageSpeed Insights on a mobile connection.

We build custom websites, so we have an interest in the answer. We report every result below as plainly as we can, including the ones that don't favour custom builds, and we don't name any business.

The short version

• Almost every site is slow on mobile. Only 3 of 166 measurable sites show their main content within Google's 2.5-second target.
• WordPress runs about a third of the sites. 69 of 200 (35%), and 67% of those use the Elementor page builder.
• Custom builds scored a little higher than WordPress, and the gap is uncertain. Median 63 vs 57 out of 100.
• A well-built WordPress site beats the typical custom site. The best quarter of WordPress sites has a median of 65.5.
• About 1 in 8 websites listed on Google Maps doesn't work. 24 of 200 lead to a dead domain, an error page, a placeholder or somewhere that isn't the business.

Almost every site is slow on mobile

Google's PageSpeed test simulates a mid-range phone on a slow mobile connection and scores the result from 0 to 100. The median Dominican business site in our sample scored 57, and its main content took 9.2 seconds to appear. Google considers 2.5 seconds or less "good". Only 3 of the 166 sites we could measure reach that, and only 6 score 90 or more.

Platform choice moves the median by a few points. The bigger story is that the whole market is far from the speed Google rewards, whatever the site is built with.

What the sites are built with

WordPress is the most common platform by far, with 69 sites. The next largest groups are sites we could not identify or load, hosted industry platforms (website services for rentals, real estate and hotels), and hand-built HTML. Shopify appears almost only in retail.

Elementor dominates WordPress here: 67% of the WordPress sites use it. Elementor sites were not slower by score (median 58 against 54 for other WordPress sites), although their pages were heavier.

WordPress or custom: what the data says

Custom builds (Next.js, React apps and hand-built HTML) had the highest median score of the main groups: 63, against 57 for WordPress and 55 for site builders such as Wix and Squarespace.

That 6-point gap is smaller than it looks. Each site was tested once, and PageSpeed scores move 5 to 10 points between runs. Allowing for that uncertainty, the true gap could plausibly be anywhere from -3 to +17 points, which includes no difference at all.

The best WordPress sites beat the typical custom site. The top quarter of WordPress sites has a median score of 65.5, above the custom median of 63. And 40% of custom sites score below the WordPress median. A custom build does not guarantee speed, and a carefully built WordPress site is not slow by default.

Where custom builds do differ is the share of very poor results: 14% of custom sites scored below 50, against 30% of WordPress sites. Sector plays a part too. Professional-services sites were the fastest and include the most custom builds, but the gap remains when that sector is left out, and after comparing each site with its own sector it is still about 6 points.

Industry platforms scored lowest

The weakest group was not a general site builder. It was the hosted services built for one industry: booking-site builders for vacation rentals, listing platforms for real-estate agencies, and hotel website systems. Their median score was 45.5. They are convenient because listings and bookings update themselves, but the business has little control over how fast the pages load.

One in eight Maps websites doesn't work

24 of the 200 businesses (12%) still show a website on their Google Maps listing that leads nowhere useful. For a customer who taps the link, that looks the same as a business that has closed.

One of those domains has been taken over and now shows gambling spam. When a domain registration lapses, someone else can buy it, and the old Google Maps link starts sending your customers to them.

The basics

In a market that serves tourists and expats, only 18% of sites offer real versions in both languages, and fewer than half (38%) link to WhatsApp, the channel many Dominican customers use first.

By sector

What this means if you run a business

• Tap your own Google Maps link. It takes ten seconds, and for one business in eight it leads nowhere.
• Test your site on a phone. Run it through PageSpeed Insights and look at the mobile result, not desktop.
• Don't switch platform expecting speed. The spread inside every platform is far larger than the gap between them. Heavy images, too many plugins and bloated themes slow a site down on any platform.
• Renew your domain on auto-pay. An expired domain can end up selling someone else's product to your customers.

How we did it

Sample. We set quotas across 8 sectors and 6 cities: Santo Domingo, Punta Cana/Bávaro, Santiago, Puerto Plata/Sosúa/Cabarete, Las Terrenas and La Romana. For each of 141 Google Maps searches we loaded the full result list and opened listings in a random order, so small businesses far down the list had the same chance as the top-ranked ones. We opened 1,654 listings, and 831 of them had no website. Only the business's own domain counted: no Facebook pages, booking sites, directories or international chains' global sites. The list of 200 sites was fixed before any of them was measured.

Measurement. We identified each platform from strings in the page source or server headers and recorded the exact evidence for every classification. Speed is one mobile PageSpeed Insights test per site, run between 30 September and 1 October 2026. Pages that were not the business's working site are left out of the speed figures and counted as broken instead.

Limitations. PageSpeed simulates a slow mobile connection from Google's servers; it is not a measurement of real Dominican networks. The sample reflects what Google Maps lists, and vacation rentals are under-represented because almost none have their own website. Some sites we call hand-built showed no platform signature at all and may run on platforms we could not identify. This study shows how platform and speed relate in this sample; it cannot show that a platform causes a site to be fast or slow. Webflow, GoDaddy, Joomla and Drupal had too few sites to compare.

Our conflict of interest

DR Web Studio builds custom websites, so a result showing custom builds are faster would suit us. That is why we report the uncertainty around that gap and the finding that the best WordPress sites beat the typical custom site, and why the sample was fixed before we measured anything.

Download the data

The aggregate figures behind every table and chart in this article are available as a CSV file. It contains totals, medians and sample sizes only, with no individual businesses.