← All projects
Side project2019 – 2026Archived

Wings

Website-as-a-service for campaign organisations, 2019 to 2026

Wings social card showing a petition page on a phone next to the Wings logo

Wings was a website-as-a-service platform for progressive campaign organisations: a centralised CMS with a Next.js front end, built so that a campaigner could launch a petition, a fundraiser or a sign-up page in an afternoon without going through a developer. At its height it ran sites for FNV, GroenLinks, Wakker Dier, Milieudefensie, Humanistisch Verbond, BIJ1 and a long tail of smaller organisations and one-off campaigns.

I built it at Bureau Bolster, the agency I founded in 2014, and for a while it was the thing the agency was really about. The insight held up well: campaign organisations have a genuinely distinct set of needs, they are underserved by generic website tooling, and the recurring cost of stitching together five American SaaS products is both financial and political.

Why it ended

The platform outlived its architecture. Wings ran on US infrastructure at a moment when its own users were increasingly, and correctly, uncomfortable with that. Its CMS was proprietary and its sites were not transferable, which meant every client was locked to us whether or not that served them. The forms layer was limited in ways that a decade of campaign work kept running into. And maintaining a bespoke CMS is a large fixed cost to carry for a small number of sites.

None of that is a story about the idea being wrong. It is a story about the implementation aging, and about a market that moved underneath it.

What happened to it

Wings was wound down in 2026. Sites with live forms, recurring donations and ongoing editorial work were migrated to Dreamkit, which takes the same premise and rebuilds it on WordPress, on European hosting, with the site transferable to the client. Sites that only needed to stay readable were mirrored to static archives, where the design and text survive and the forms are allowed to stop working.

I have a lot of affection for this project and I learned more from winding it down than from building it. The parts worth keeping went into Dreamkit; the parts that were load-bearing only for us did not.

Succeeded by Dreamkit.