Five Patterns Every Great Digital Project of 2025 Got Right

1. They treated architecture like an investment, not a checkbox 

Ask any developer about "headless" or "hybrid" or "composable" and you'll get either a passionate sermon or a deep sigh. I get both, often in the same meeting. 

What was striking this year is how quietly the winners handled it. Nobody was doing architecture for the buzzword. They were doing it because they wanted to stop fighting their own platform. 

MND Energie, one of the largest independent gas and electricity suppliers in Czechia, didn't just upgrade, they consolidated two content management systems into one hybrid-headless setup, with the Content Hub as a single source of truth. That's not a redesign. That's a deliberate decision about how the business will operate for the next five years. 

Translink (with MSQ DX) was the very first organization to upgrade from Kentico 12 all the way to Xperience by Kentico, while serving 1.5 million passengers and maintaining PCI compliance. Building a custom commerce engine while the trains keep running. That's the kind of work that ages you ten years and then makes you proud forever. 

And ASLA, the American Society of Landscape Architects (with BlueModus), folded two large, fragmented legacy systems into one platform, and chose SaaS hosting on top of that, because they're thinking about the next decade, not the next campaign. The pattern: the great projects pick architecture that lets the business move faster, even if it is a challenge now. 

2. The partner-client relationship did half the work

This is the one I wish more case studies talked about openly. Technology gets the byline, but the relationship gets the win. 

Look at Bob W., our Community Choice winner. Their site, built with IDHL, didn't just translate a brand, it translated a character. Bob, the imaginary friend who runs your apartment in Tallinn or Vienna, has personality. To get that on the screen, the agency and the client had to spend a lot of time together not talking about CMS at all. They were talking about tone. Charm. Friction. The result is a site that has personality, with memorable branding that truly stands out, which is a rare accomplishment on the internet today. 

Pierret, a Belgian door and window manufacturer, had the kind of timeline that gives project managers nightmares: launch before the summer holidays so SEO peaks in September, just as a brand-new factory opens in 2026. Delaware Consulting and Pierret had to ride that train together, and you don't get there without trust on both sides. 

I'll say something a little unfashionable here. The technology was never the bottleneck on these projects. The conversations were. The winners are the teams who learned to have hard conversations early: about scope, about priorities, about what "done" actually means. 

3. Content strategy stopped being a side dish

For years I would read project debriefs that went, "Phase 1: build the platform. Phase 2: figure out the content." Phase 2 rarely happened. Or it happened in a panic the week before launch. 

That's not what 2025 looked like. 

Cleveland Metroparks (with Americaneagle.com) is a great example. They run Parks, a Zoo, and a Golf division: three audiences, three voices, one digital backbone. The Content Hub, plus a thoughtful permissions model, meant that the editor managing the Zoo's events doesn't accidentally rewrite the golf course's tee-time policy. That sounds boring, and it is, until it isn't. Boring infrastructure = a lot of saved Sunday evenings. 

ROC van Amsterdam & Flevoland (with TrueLime) had to migrate from an end-of-life CMS while serving over 300 programs, and they leaned hard into persona-based content. A 17-year-old looking at a hospitality program is not the same person as a career-changer at 35 looking at adult education. Their site knows that. 

And Oppenheimer (with Sagepath Reply) pulled off a five-month migration from CrownPeak with single-source content management feeding multiple microsites and channels. Five months. I want to put that in bold for anyone who has ever shipped a website. 

The pattern: the winners treated content as the product, and the platform as the kitchen. Not the other way around. 

4. They optimized for experience, not for feature lists

Here is a small confession that VPs of Partnership rarely say out loud: I find feature checklists boring. I know I shouldn't. But when I read a brief that says, "We need a CMS with personalization, AI, headless, multilingual, commerce, and a member portal," I know that brief was written by a committee, and somewhere along the way consideration for the actual user disappeared. 

The 2025 winners didn't ignore the user. They obsessed over them.

LeShuttle Freight (with Crafted) operates 24/7 between the UK and France with 35-minute crossings. Their winning idea? A real-time service status on the homepage. That's it. That's the whole story. A driver pulls into the queue, glances at the homepage, and knows whether they're rolling on or sitting in their cab. Multilingual support in English, French, German and Spanish on top of that. That's a site built for someone with diesel on their hands, not for a marketing committee. 

Bob W. did the same thing on the consumer side: smart filtering, comparison tools, orchestrated guest journeys. None of those words are exciting on a slide, but all of them matter when you're trying to book an apartment for a long weekend. 

The lesson here is not to "ignore features." It's start from the user, work backwards to the feature. Almost nobody does this. But the winners did. 

5. They thought in ecosystems, not in islands

This one didn't show up in every project, but it showed up in enough of them to be a pattern. 

The best sites of 2025 weren't islands. They lived inside a complete ecosystem of tools, CRM, marketing automation, analytics, identity, commerce, and the platform's job was to play well with all of them. 

Cleveland Metroparks, for example, integrated with Shopify, Mapbox, OneLogin (SSO), and RecTrac. ROC integrated with Eduarte, BlueConic, and Spotler. Pierret built on .NET, Typesense, Azure and React/TypeScript. None of them tried to be a walled off garden. 

This matters because the day of "buy one platform, and then fit your business to it" is over. The winners assumed their stack would change, their tools would change, their AI tools would change, and they built for that.

Make it open by default, flexible by design. And don't forget to be humble about how much you can predict. 

What it takes to create winning digital projects in 2026

When I started in partnerships ten years ago, "great digital experience" mostly meant a good-looking site that doesn't crash. Honestly, at the time, the bar was high. 

What 2025 showed me is that the bar moved. The great projects this year were not the prettiest, or the most technically clever, or the loudest about AI. They were the ones where the team  (client and partner together) made hard, considered choices about architecture, content, and experience, and then trusted each other to ship them. 

If you're starting a project in 2026, my unsolicited advice is: pick the partner before you pick the features. Argue about the architecture before you argue about the homepage hero. And when in doubt, design for the person who is actually going to use the thing, not for the slide that announces it. 

Congratulations again to every winner and every honorable mention. You made my weekend shorter than it should have been, and I'm grateful for it.