Case Study Nonprofit

Nobody could say what all the servers were for, and every one of them had a bill

The International Living Future Institute ran two websites across a collection of self-managed servers that had grown for years. We mapped what was there, moved what mattered, and switched off the rest.

  • 2025 client since
  • $500+ a month off the old server bill
  • 0 downtime during the move
  • 1 team to call for both sites

A small staff, two websites, and a growing pile of servers

In their own words, "The mission of the International Living Future Institute (Living Future) is to cultivate a society that is socially just, culturally rich, and ecologically restorative." The institute is the nonprofit behind the Living Building Challenge. It runs two websites: its main site, where programs, certifications, and events live, and Trim Tab, its online magazine.

The people who work there are program, design, and communications professionals. Nobody on staff is a server administrator, and nobody should have to be. But the websites ran on servers the institute managed itself at DigitalOcean, and over the years that collection had grown.

Each server had arrived for a good reason. The reasons left with the people who set them up. The bills stayed.

Nobody could say what each server was for

By the time we met them, the institute was paying every month for a group of servers, and no one could say with confidence what each one did. Some ran the two websites. Others had been set up for projects that had long since ended. Nothing was written down.

That leaves an organization stuck in a particular way. Switching a server off might save money, or it might take down something important that nobody remembered. So nothing got switched off, and the person approving the bill each month approved it again.

Servers that an organization manages itself also need looking after: software updates, backups, and someone watching them. That work belonged to no one in particular. When a site had a problem, the first question was whose problem it was.

We made a map before we moved anything

We started with an inventory. Before touching anything, we went through every server and wrote down what was running on it, what connected to it, and whether anything still depended on it. For the first time, the institute had a plain list: these servers run the websites, and these do nothing anyone needs. The list was worth having on its own, because it turned a vague worry into a decision somebody could make.

We moved the two websites first. Both went to managed hosting, where the server upkeep, the security, and the backups are our job, inside one predictable monthly cost. We checked each site in its new home before visitors were pointed to it, and neither site went down during the move. The sites kept the same addresses and staff kept the same logins, so the people who use them every day had nothing new to learn.

We switched off the old servers last. Only when both sites were settled, and we had confirmed that nothing else relied on the old machines, did the institute shut them down. Working in that order meant turning something off was never a gamble. The old server bill dropped by more than $500 a month. In its place is one fee for hosting and support, and it is no longer a mystery.

We left the websites alone. The sites themselves were in good shape and staff knew how to use them, so nothing changed about how they look or how staff edit them. What the institute needed was solid ground under the websites it already had.

What staff ask us for now

Today nobody at the institute manages a server. Updates to both sites run every week without anyone asking. When the magazine was due for a major new version of WordPress, it was done the day staff raised it, and they were told that kind of request no longer needs to be on their list.

The requests we get now are about the institute's programs. A banner for the annual conference, a registration page that has to change on a set date, new names on the staff page. Their buildings team has also been working with us on a new resource library, which staff are filling with content before it opens.

The person who approves the bill sees one line for hosting and knows what it pays for. When something does go wrong, there is one team to call, and it is a team that already knows both sites.

What changed

  • Every server was identified and written down before anything was switched off
  • Both websites moved to managed hosting with no downtime, with security and backups included
  • Servers that were no longer needed were retired, taking more than $500 a month off the server bill
  • Routine updates happen every week without anyone at the institute asking
  • Staff have one team to call for a fix or for something new

Under the hood

For the developer evaluating us
  • 2WordPress sites on block themes
  • 28plugins on the main site
  • 33plugins on Trim Tab
  • Weeklycore, plugin, and theme updates

What We Built

  • A written inventory of every server and what depended on it
  • Both websites moved to managed hosting and checked before visitors were pointed to them
  • Old servers retired once nothing relied on them
  • Security, backups, and weekly updates handled as part of hosting
  • Ongoing support and development for both sites from one team

The inventory

For each server we recorded what it was running, which domains pointed at it, and whether either production website depended on it. The result separated the servers running production from the ones that could be retired.

The order of the move

Copy each site to managed cloud hosting, verify it there against the original, point visitors to the new home, watch it, and only then retire the old server. Decommissioning came last and only for machines with nothing depending on them.

The two sites

Both are WordPress sites built on block themes that extend Twenty Twenty-Four, with custom blocks registered through the current block API. The main site adds a case studies content type and site search through Relevanssi. Trim Tab runs a separate installation with its own child theme. The two share a design language and are maintained as separate sites on separate servers.

Hosting and upkeep

Each site runs as its own application on managed cloud servers behind Cloudflare Enterprise (WAF and CDN), with Imunify360 on the server, managed SSL, and scheduled offsite backups. WordPress core, plugins, and themes update on a weekly schedule with checks before and after.

Figures current as of September 26, 2026

More stories like this

Want a web partner who picks up the phone?

Tell us what your site needs to do, and we will tell you how we would approach it.