Icelandic websites on the move

Guest article on Vísir — Thoughts on content management systems, part 2 of 2

In my previous piece, I discussed the quiet demise of custom Icelandic content management systems and promised to dive into the data next. Here it is. I relied on two different metrics: BuiltWith, which analyzes the technology powering websites in each country, and Semrush, which measures where search interest is heading, both locally and globally. The former tells us where the market is; the latter, where it is going. Together, they paint a picture that surprised me in a few places.

Half the country on a single platform

According to BuiltWith, just over 22,000 websites in Iceland are identified with a recognizable CMS. WordPress reigns supreme with about 11,000 sites, accounting for nearly half of all detections—a status that mirrors its global standing. Next come the simple site builders: Wix with nearly 3,000 sites and Squarespace with around 1,800. Duda powers about 550 sites, Framer around 520, Joomla around 460, Webflow just over 400, and Prismic around 350.

Further down the list, interesting figures emerge. Dacoda, one of the old legacy Icelandic systems, still powers 176 sites, making it the largest of the local platforms detected. Headless systems remain a niche choice here: Sanity is detected on 90 sites and Contentful on just over 50. Umbraco is registered on only eight sites, a figure I take with a grain of salt; the system runs on the backend and is difficult to detect externally. There are undoubtedly many more, as Icelandic digital agencies have specialized in Umbraco for years.

The same caveat applies to other custom Icelandic systems. Moya, Eplica, LiSA, and VEVA do not show up in the analysis because foreign scanning tools do not recognize them due to their small footprint. The actual distribution of Icelandic websites is therefore slightly different than the list suggests, but the big picture remains: WordPress holds about half the market, simple site builders take up about a quarter, and the rest share what remains.

Interest shifts elsewhere

Now to the momentum. I exported search traffic data from Semrush for the websites of various CMS providers, both globally and from the Icelandic database, spanning a year back.

A quick warning before we proceed: there is a heavy flurry of percentages ahead. It will pass quickly, and the story the numbers tell is well worth the read.

Globally, one platform stands out. Search traffic to Framer’s website grew by 66% in a single year, rising from nearly 490,000 monthly visits to over 800,000. In comparison, combined traffic to Sanity, Prismic, and Contentful is under 300,000 monthly visits; Framer alone is nearly three times larger than the three headless systems combined and is growing exponentially faster (Contentful grew by 21%, Sanity by 17%, and Prismic contracted by 11%). Those familiar with these platforms know it is not quite an apples-to-apples comparison: Framer appeals to both individuals and enterprises, while headless systems are geared almost exclusively toward engineering teams, which explains some of the volume difference. Still, the growth rate speaks for itself. The same pattern holds in the US market: Framer grew by 45% while WordPress declined by 36% and Duda by 21%.

In Iceland, the contrast is even sharper. Framer is the only platform showing growth, up 51% in a single year from 95 monthly searches to 143. Everything else is slipping: Contentful dropped by 86% (from 211 to 29), WordPress fell by 66% (from 1,056 to 361), Duda by 54%, Prismic by 52%, and Webflow by 51% (from 336 to 164). Sanity and Payload barely register in local searches. Webflow still holds a narrow lead over Framer in absolute numbers, 164 monthly searches to 143, but the trajectories are heading in opposite directions; if this trend continues, Framer will overtake it in the coming months.

Two caveats are necessary here. Search interest does not measure actual usage; it measures what people are exploring and learning. Also, part of the WordPress decline is likely due to information searches shifting from Google to generative AI, as I touched upon in my last article. Even so, the pattern is decisive: yesterday's majority has stopped growing, and new curiosity is almost entirely concentrated in one place.

The systems saying goodbye

The data tells a clear story. The custom local systems are already on their way out, as detailed in my previous piece. The WordPress empire still anchors half the market but has ceased to grow; the system is showing its age, the maintenance burden is very real, and plugin vulnerabilities remain the most common entry point for attacks. Those with a well-maintained WordPress site have no immediate reason to leave, but I would think twice before building a new site on the platform as we head into 2026. Simple builders like Wix, Squarespace, and Duda still solve straightforward tasks quickly and affordably, but they have not kept pace with the AI evolution, and their search volume reflects that. The same applies to Webflow here at home, even if its position remains stronger across the Atlantic.

Regarding VEVA, my thoughts from the previous article stand: I have no basis to evaluate the system itself, but any buyer must price in the risk of being locked into a single digital agency in a market that has recently shown how that story tends to end.

Three platforms I am betting on

I see three systems continuing to gain ground: Framer, Sanity, and Payload—the latter having been acquired last year by Figma, the world’s leading design tool. What they share is a modern technical foundation where AI is woven directly into the workflow rather than tacked on as an afterthought. Their content schemas are open and portable, preventing vendor lock-in, and they are backed by fast-growing global companies and vibrant, active developer communities. These are precisely the traits I argued would matter most in the coming years.

The choice between them comes down to the task at hand. Framer is, in my view, the finest tool for marketing and informational websites, which constitute the vast majority of local sites: design and deployment in a single environment, fluid for editors, and leading the way in native AI. Our experience shows that development here is roughly three times faster than conventional workflows. Sanity and Payload are the right choices when content needs to live beyond a single website or when heavy custom backend logic is required, such as behind portals; they require a development team but offer complete freedom in frontend presentation and integrations. The fact that headless options remain a niche choice here despite global adoption tells me that the Icelandic market has yet to take that step, rather than it being the wrong path to walk.

The wild card in all of this is generative AI and the sheer velocity of its evolution. What was cutting-edge six months ago is taken for granted today, and the platforms mentioned above update on a monthly cadence rather than a yearly one. That is why now is the right time to clean up your digital foundation: structure your data architecture, refine content flows, and polish your narrative, because AI only performs as well as the structured ground beneath it. The same applies to team capabilities; the availability of AI training has never been richer. The organizations that do this homework well will adapt to the new landscape far quicker than the competition.

The journey has begun

Icelandic websites are on the move. The migration away from legacy platforms will be gradual but significant, accelerated by the AI-driven shifts I described in my previous article. For website owners, the takeaway is the same, now backed by data: begin with your actual needs, select a platform with a global future, and keep your content portable. The market is moving, and for the first time in a quarter-century, the cost of switching is swiftly declining. There has never been a better time to build for the future.

The author is the founder and CEO of REVERA and co-founder of Optise. Data sources: BuiltWith Trends for Iceland (July 2026) and exported Semrush datasets covering global, US, and Icelandic databases from July 2025 to July 2026.