Every time we publish a study, somebody asks the same question: fine, but what is the headline number?

It is a reasonable thing to want. We have spent a year measuring Dominican business websites — hotels, restaurants, veterinarians, dive shops, immigration lawyers, real estate agents, security companies, web agencies. Roughly 460 distinct businesses across four studies. At some point you should be able to say what all of it adds up to.

So here is the attempt. One number, with the working shown, including the parts that make the number weaker than it sounds.

What we actually measured

Four separate pieces of fieldwork, all in September and October 2026.

• 36 Punta Cana and Bayahibe hotel websites, tested with Google's PageSpeed Insights on a simulated mid-range phone, plus a direct fetch of each homepage.

• 20 small hotels and apartment-hotels in Bávaro and Bayahibe, checked by hand for whether they have a website of their own at all.

• 64 restaurants across seven markets, each opened in a live browser to see whether a customer can read a menu with prices.

• 369 businesses named by four AI assistants across 24 queries, each one verified by hand against its own site, its Google listing, an industry register or local news.

The hotel and restaurant samples overlap with the AI study: six hotels and at least twenty-two restaurants appear in both. That is why the total is about 460 and not 489.

The bar

To produce one number you have to define one standard. Here is ours. It is deliberately low — nothing here is a design opinion or a marketing preference.

A business clears the bar if all four of these are true:

• It has a working web page it controls. Not a dead link, not an expired domain, not only a Facebook page or a directory listing.

• That page says what the business sells, as readable text. A photograph of a price list is not readable text. Neither is a PDF, a Scribd upload or a 2019 flipbook.

• Its Google listing points at that working page.

• The page's main content appears in under four seconds on a mid-range phone.

Four seconds is already generous: Google's own threshold for "good" is 2.5 seconds. We used four because we wanted the bar to be beatable.

What each study says against that bar

Restaurants. Of 64 restaurants, 10 publish a menu as readable text with prices on their own website. That is 16%. Of the rest, some have no website, some have the menu as a photograph or PDF, some send you to a delivery app, and 22% have a website or menu link that is dead, expired or a placeholder. The full study is here.

Hotels, on speed. Of 36 hotel websites measured on mobile, exactly one loaded its main content in under four seconds. One. The median was 17.1 seconds. Not a single site in the sample reached Google's 2.5-second "good" threshold, and none scored 90 or above on PageSpeed. The median page weighed just over 8 MB. The full study is here.

Small accommodation, on existence. Of 20 small hotels and apartment-hotels listed in Bávaro and Bayahibe, 4 had a website of their own. Twelve had nothing but aggregator listings, Facebook pages and directory entries. Two we could not identify as a trading business at all under the name they were listed under.

Service businesses, on being correctly described. Of 117 service businesses that AI assistants named in their answers, 7 could not be verified to exist. One had permanently closed. Twenty-nine — 25% — had their name rendered inaccurately by at least one assistant. That study is here.

Why you cannot multiply these together

This is the part most people skip, and it is why the single number has to be stated carefully.

Each study measured a different check on a different sample. We tested the restaurants for readable menus, not for speed. We tested the hotels for speed, not for menus. If the two were independent, you could multiply 16% by 3% and announce that about one Dominican business in two hundred clears the whole bar. That number would be nonsense. We never measured any single business against all four checks.

What we can say is the thing the arithmetic does support: on no sample we measured did more than 16% clear the check we applied to it. And the true all-four-checks figure cannot be higher than the lowest single-check pass rate on a comparable sample, because a business that fails one check fails the bar.

The number

At the very best, roughly one in six. On the strictest single check we ran, one in thirty-six.

Across everything measured, the share of Dominican businesses clearing a basic, deliberately low standard of online usefulness never exceeded 16%, and on the one dimension we measured most precisely it was 3%.

If you want a single sentence to carry around: fewer than one Dominican business in six has an online presence that does the job, and the ones that fail do not fail by a little.

The number that makes it worse, and the one that makes it fixable

Here is the finding that reframes everything else. Among the 64 restaurants, 98% had a Google Business Profile.

The Dominican market is not absent from the internet. It is overwhelmingly present on the internet. Almost every business has a listing, a Facebook page, photos, reviews, hours. What it does not have is a working destination at the end of that listing. Roughly one in six of those Google listings pointed at something dead.

That is a much better problem to have than invisibility, because it is cheap to fix. A business with 98% presence and 16% usefulness is not starting from zero. It is starting from a front door that works and a room behind it that does not.

There is one more number worth knowing from the AI study. Service businesses with their own website were named by AI assistants 2.52 times on average; businesses without one were named 1.50 times. Roughly a 40% penalty for not owning a page. That gap is the clearest evidence we have that the website still does work, even in a world where customers increasingly ask a machine instead of typing a query.

What this number is not

Worth being blunt about the limits, because an unqualified statistic travels further than it should.

It is not a measurement of revenue. Nothing here says a restaurant with no online menu earns less than one with a menu. Plenty of excellent Dominican businesses are full every night with a website that does not exist. We measured what is published, not what it earns.

It is not nationally representative. The samples were built from search results, tourism rankings and Google Maps listings, not from a sampling frame of all Dominican businesses. We did not draw them from a register of every Dominican business, and the official business statistics that do exist are several years old.

It is not stable. Every figure here is a snapshot of September and October 2026. Websites get rebuilt, domains expire, AI models get retrained. A number this specific has a shelf life.

And it is not neutral. We build websites for a living, so a finding that most Dominican websites are bad is convenient for us. That is why we publish the raw counts, the sample lists and the cases that cut against us, and why we explain the whole arrangement.

What to do with it

If you run a Dominican business and you want to know which side of the 16% you are on, you do not need an audit. You need four checks, in this order, and they take about twenty minutes.

• Search your business name on your phone. Tap the website link in your Google listing. Does a working page load?

• Time it. Count the seconds until you can read something. Over four, you have the most common Dominican website problem.

• Look for your prices, menu or service list. Is it text you could select and copy, or is it a picture?

• Open your listing and check the hours, phone and address are current.

Most businesses fail at least one of those. That is the whole finding, and it is why the thirty-minute Google Business Profile fix is still the highest-return hour of work available to almost anyone reading this.

If you would rather have someone run the checks with you and tell you honestly which ones matter for your business, DR Web Studio works with businesses across the Dominican Republic — contact us for a free consultation.