Fifteen Years of sheepCRM

In the summer of 2011, I was between consulting contracts and volunteering some time with Adventure Plus, a charity that works with young people getting them outdoors. The founder, Jon Cox, is a good friend. I'd started helping with grant applications — useful work, but not exactly where my skills were sharpest.

Then the conversation turned to their database.

Adventure Plus had an Access database built by a volunteer years earlier. It had been a brilliant solution for its time, but the volunteer had moved on, the system had grown stale, and nobody could maintain it. I knew I could help with this.

The Salesforce false start

My first instinct was Salesforce. It was the world leader, it would be great experience for me, and surely it would be a good fit for a small charity. I picked it up and started working through it.

It wasn't a good fit. The language, the workflow, the assumptions — everything was built around sales funnels and lead conversion. It would have worked, technically, but it would have felt like wearing someone else's clothes. A system that was always slightly wrong, always making you compromise.

So I made what I later described on the CTO Playbook podcast as "a classic foolish mistake of an entrepreneurial software engineer" — I decided to build my own.

That summer

I spent the summer of 2011 writing a system for Adventure Plus. Somewhere in that process, a thought surfaced: I wonder if there's a business model here.

The first paying customers arrived in early 2012.

Two technology decisions from those early months have proved remarkably durable. I hosted the system on Amazon Web Services — a genuine novelty at the time, especially for a tool aimed at charities. And I chose MongoDB, which gave us flexible data storage that would have been painful to achieve with a traditional relational database. Both were contrarian calls in 2011. Both still underpin sheepCRM today.

A third architectural decision — separating the API from the front-end client — was perhaps the most consequential. It meant the system could evolve on both sides independently, and it's the reason we've been able to modernise the product over fifteen years without rewriting it from scratch.

Finding the niche

In its earliest incarnation, sheepCRM was a fairly generic database. One early contact described it as "Access in the cloud," which was accurate and slightly deflating. The system would work for almost anyone who turned up, which felt like a strength but was actually a weakness.

I was too broad. I was building what was needed to win the next piece of work, which meant the product grew wide but lacked depth. If I were starting again, the single thing I'd do differently is focus earlier — have the confidence to say no to work that didn't fit, because of a bigger yes to the market I wanted to serve.

That focus came gradually, not through strategy but through listening. A series of membership organisations found us, and memberships — a specific sub-sector of subscriptions — became the core. We weren't tracking leads through a funnel. We were helping organisations maintain long-term relationships with their members: richer data, ongoing engagement, and facilitating whatever mission brought those members together in the first place.

Over the years, we've tightened further. Today, sheepCRM serves principally professional associations.

The name

People ask about the name. There isn't a dramatic origin story. I was aware of a generation of products with animal names — MailChimp, SurveyMonkey, Firefox — and I wanted something informal. Most CRMs of that era were boring, commercial, and had poor interfaces. I wanted something softer, fluffier, approachable.

sheepCRM was always designed for the small guys, not for big commercials. The measure I set myself was that anyone who could use Facebook should be able to pick it up without a training manual.

Scaling up

For the first few years, sheepCRM was a side project alongside consulting work. In 2016, I went all in.

In 2019, I merged the company with Authentic Digital, run by brothers Joe and Will Jefferies. The merger was a shortcut to a much bigger team and access to valuable experience in sales and operations that I didn't have. It was the step-change the product needed.

Today there are ten of us. We serve just over a hundred organisations, ranging from professional bodies with close to a hundred thousand members down to smaller associations with around a thousand. Adventure Plus — the charity where it all started — is still a customer.

What I've learned

Building and operating a SaaS product for fifteen years teaches you things that consulting and advisory work cannot. You live with your own decisions. The architectural choices you made in year one are still running in production in year fifteen. The shortcuts you took to win early customers become the technical debt you're still paying down a decade later.

It also teaches you about the gap between where organisations think they are and where they actually are. This became the foundation of what I now call the Three Mountains model — a way of helping organisations honestly assess their operational maturity before trying to leap ahead to ambitious outcomes. The model came directly from years of customer conversations where people arrived wanting data warehousing and predictive analytics, but couldn't yet reliably answer the question: how many members do we have?

The thing I'm most proud of isn't technical. It's seeing sheepCRM make a difference to organisations doing good work — professional bodies supporting their members, charities reaching the people they exist to serve. The technology is in service of that, and always has been.

Still going

Fifteen years in, sheepCRM is still evolving. We're applying AI to internal processes and exploring what the new generation of tools means for membership organisations. The mountains keep shifting — every time you reach what felt like a summit, the landscape has changed and there's further to go.

That's probably the nature of building software. You never arrive. But if you're honest about where you are, focused on the right problem, and willing to do the hard work of the valleys between the peaks, you keep making progress.

Adventure Plus is still using the system. Jon Cox is still a good friend. And I still think that making the jump from Salesforce to writing my own system was one of the best foolish decisions I ever made.